1. |
- Hagberg, Eva, et al.
(author)
-
Semi-supervised learning with natural language processing for right ventricle classification in echocardiography—a scalable approach
- 2022
-
In: Computers in Biology and Medicine. - : Elsevier BV. - 0010-4825 .- 1879-0534. ; 143
-
Journal article (peer-reviewed)abstract
- We created a deep learning model, trained on text classified by natural language processing (NLP), to assess right ventricular (RV) size and function from echocardiographic images. We included 12,684 examinations with corresponding written reports for text classification. After manual annotation of 1489 reports, we trained an NLP model to classify the remaining 10,651 reports. A view classifier was developed to select the 4-chamber or RV-focused view from an echocardiographic examination (n = 539). The final models were two image classification models trained on the predicted labels from the combined manual annotation and NLP models and the corresponding echocardiographic view to assess RV function (training set n = 11,008) and size (training set n = 9951. The text classifier identified impaired RV function with 99% sensitivity and 98% specificity and RV enlargement with 98% sensitivity and 98% specificity. The view classification model identified the 4-chamber view with 92% accuracy and the RV-focused view with 73% accuracy. The image classification models identified impaired RV function with 93% sensitivity and 72% specificity and an enlarged RV with 80% sensitivity and 85% specificity; agreement with the written reports was substantial (both κ = 0.65). Our findings show that models for automatic image assessment can be trained to classify RV size and function by using model-annotated data from written echocardiography reports. This pipeline for auto-annotation of the echocardiographic images, using a NLP model with medical reports as input, can be used to train an image-assessment model without manual annotation of images and enables fast and inexpensive expansion of the training dataset when needed. © 2022
|
|
2. |
- Abbaspour, S., et al.
(author)
-
Real-Time and Offline Evaluation of Myoelectric Pattern Recognition for the Decoding of Hand Movements
- 2021
-
In: Sensors. - : MDPI AG. - 1424-8220. ; 21:16
-
Journal article (peer-reviewed)abstract
- Pattern recognition algorithms have been widely used to map surface electromyographic signals to target movements as a source for prosthetic control. However, most investigations have been conducted offline by performing the analysis on pre-recorded datasets. While real-time data analysis (i.e., classification when new data becomes available, with limits on latency under 200-300 milliseconds) plays an important role in the control of prosthetics, less knowledge has been gained with respect to real-time performance. Recent literature has underscored the differences between offline classification accuracy, the most common performance metric, and the usability of upper limb prostheses. Therefore, a comparative offline and real-time performance analysis between common algorithms had yet to be performed. In this study, we investigated the offline and real-time performance of nine different classification algorithms, decoding ten individual hand and wrist movements. Surface myoelectric signals were recorded from fifteen able-bodied subjects while performing the ten movements. The offline decoding demonstrated that linear discriminant analysis (LDA) and maximum likelihood estimation (MLE) significantly (p < 0.05) outperformed other classifiers, with an average classification accuracy of above 97%. On the other hand, the real-time investigation revealed that, in addition to the LDA and MLE, multilayer perceptron also outperformed the other algorithms and achieved a classification accuracy and completion rate of above 68% and 69%, respectively.
|
|
3. |
- Guerrero Orozco, Laura, 1996, et al.
(author)
-
Microwave Antenna System for Muscle Rupture Imaging with a Lossy Gel to Reduce Multipath Interference
- 2022
-
In: Sensors. - : MDPI AG. - 1424-8220. ; 22:11
-
Journal article (peer-reviewed)abstract
- Injuries to the hamstring muscles are an increasing problem in sports. Imaging plays a key role in diagnosing and managing athletes with muscle injuries, but there are several problems with conventional imaging modalities with respect to cost and availability. We hypothesized that microwave imaging could provide improved availability and lower costs and lead to improved and more accurate diagnostics. In this paper, a semicircular microwave imaging array with eight antennae was investigated. A key component in this system is the novel antenna design, which is based on a monopole antenna and a lossy gel. The purpose of the gel is to reduce the effects of multipath signals and improve the imaging quality. Several different gels have been manufactured and evaluated in imaging experiments. For comparison, corresponding simulations were performed. The results showed that the gels can effectively reduce the multipath signals and the imaging experiments resulted in significantly more stable and repeatable reconstructions when a lossy gel was used compared to when an almost non-lossy gel was used.
|
|
4. |
- Khodadad, Davood, 1985-, et al.
(author)
-
The Value of Phase Angle in Electrical Impedance Tomography Breath Detection
- 2018
-
In: 2018 Progress in Electromagnetics Research Symposium (PIERS-Toyama). - : Electromagnetics Academy. - 9784885523168 - 9781538654552 ; , s. 1040-1043
-
Conference paper (peer-reviewed)abstract
- The objective of this paper is to report our investigation demonstrating that the phase angle information of complex impedance could be a simple indicator of a breath cycle in chest Electrical Impedance Tomography (EIT). The study used clinical neonatal EIT data. The results show that measurement of the phase angle from complex EIT data can be used as a complementary information for improving the conventional breath detection algorithms.
|
|
5. |
- Bondesson, Johan, 1991, et al.
(author)
-
Definition of Tubular Anatomic Structures from Arbitrary Stereo Lithographic Surface
- 2017
-
In: Initiative Seminar Engineering Health, 8-9 November 2017, Chalmers.
-
Conference paper (other academic/artistic)abstract
- An accurate description of anatomies and dynamics of vessels is crucial to understand their characteristics and improve surgical techniques, thus it is the basis, in addition to surgeon experience, on which stent design and operation procedures rely. The process of producing this description is user intensive, and recent improvement in image processing of medical3D imaging allows for a more automated workflow. However, there is a need to bridge the gap from a processed geometry to a robust mathematical computational grid. By sequentially segmenting a tubular anatomic structure, here defined by a stereo lithographic (STL) surface, an initial centerline is formed by connecting centroids of orthogonal cross-sectional contours along the length of the structure. Relying on the initial centerline, a set of non-overlapping 2D cross sectional contours are defined along the centerline, a centerline which is updated after the 2D contours are produced. After a second iteration of producing 2D contours and updating the centerline, a full description of the structure is created. Our method for describing vessel geometry shows good coherence to existing method. The main advantages of our method include the possibility of having arbitrary triangulated STL surface input, automated centerline definition, safety against intersecting cross-sectional contours and automatic clean-up of local kinks and wrinkles.
|
|
6. |
- Bäcklin, Emelie, et al.
(author)
-
Pulmonary volumes and signs of chronic airflow limitation in quantitative computed tomography
- 2024
-
In: Clinical Physiology and Functional Imaging. - : Wiley. - 1475-0961 .- 1475-097X. ; 44:4, s. 340-348
-
Journal article (peer-reviewed)abstract
- BackgroundComputed tomography (CT) offers pulmonary volumetric quantification but is not commonly used in healthy individuals due to radiation concerns. Chronic airflow limitation (CAL) is one of the diagnostic criteria for chronic obstructive pulmonary disease (COPD), where early diagnosis is important. Our aim was to present reference values for chest CT volumetric and radiodensity measurements and explore their potential in detecting early signs of CAL.MethodsFrom the population-based Swedish CArdioPulmonarybioImage Study (SCAPIS), 294 participants aged 50–64, were categorized into non-CAL (n = 258) and CAL (n = 36) groups based on spirometry. From inspiratory and expiratory CT images we compared lung volumes, mean lung density (MLD), percentage of low attenuation volume (LAV%) and LAV cluster volume between groups, and against reference values from static pulmonary function test (PFT).ResultsThe CAL group exhibited larger lung volumes, higher LAV%, increased LAV cluster volume and lower MLD compared to the non-CAL group. Lung volumes significantly deviated from PFT values. Expiratory measurements yielded more reliable results for identifying CAL compared to inspiratory. Using a cut-off value of 0.6 for expiratory LAV%, we achieved sensitivity, specificity and positive/negative predictive values of 72%, 85% and 40%/96%, respectively.ConclusionWe present volumetric reference values from inspiratory and expiratory chest CT images for a middle-aged healthy cohort. These results are not directly comparable to those from PFTs. Measures of MLD and LAV can be valuable in the evaluation of suspected CAL. Further validation and refinement are necessary to demonstrate its potential as a decision support tool for early detection of COPD.
|
|
7. |
- Reza, Salim, 1985-, et al.
(author)
-
Smart dosimetry by pattern recognition using a single photon counting detector system in time over threshold mode
- 2012
-
In: Journal of Instrumentation. - 1748-0221. ; 7:1
-
Journal article (peer-reviewed)abstract
- The function of a dosimeter is to determine the absorbed dose of radiation, for those cases in which, generally, the particular type of radiation is already known. Lately, a number of applications have emerged in which all kinds of radiation are absorbed and are sorted by pattern recognition, such as the Medipix2 application in [1]. This form of smart dosimetry enables measurements where not only the total dosage is measured, but also the contributions of different types of radiation impacting upon the detector surface. Furthermore, the use of a photon counting system, where the energy deposition can be measured in each individual pixel, ensures measurements with a high degree of accuracy in relation to the pattern recognition. In this article a Timepix [2] detector system has been used in the creation of a smart dosimeter for Alpha, Beta and Gamma radiation. When a radioactive particle hits the detector surface it generates charge clusters and those impacting upon the detector surface are read out and image processing algorithms are then used to classify each charge cluster. The individual clusters are calculated and as a result, the dosage for each type of radiation is given. In some cases, several particles can impact in roughly the same place, forming overlapping clusters. In order to handle this problem, a cluster separation method has been added to the pattern recognition algorithm. When the clusters have been separated, they are classified by shape and sorted into the correct type of radiation. The algorithms and methods used in this dosimeter have been developed so as to be simple and computationally effective, in order to enable implementation on a portable device. © 2012 IOP Publishing Ltd and SISSA.
|
|
8. |
- Ge, Chenjie, 1991, et al.
(author)
-
Enlarged Training Dataset by Pairwise GANs for Molecular-Based Brain Tumor Classification
- 2020
-
In: IEEE Access. - 2169-3536 .- 2169-3536. ; 8:1, s. 22560-22570
-
Journal article (peer-reviewed)abstract
- This paper addresses issues of brain tumor subtype classification using Magnetic Resonance Images (MRIs) from different scanner modalities like T1 weighted, T1 weighted with contrast-enhanced, T2 weighted and FLAIR images. Currently most available glioma datasets are relatively moderate in size, and often accompanied with incomplete MRIs in different modalities. To tackle the commonly encountered problems of insufficiently large brain tumor datasets and incomplete modality of image for deep learning, we propose to add augmented brain MR images to enlarge the training dataset by employing a pairwise Generative Adversarial Network (GAN) model. The pairwise GAN is able to generate synthetic MRIs across different modalities. To achieve the patient-level diagnostic result, we propose a post-processing strategy to combine the slice-level glioma subtype classification results by majority voting. A two-stage course-to-fine training strategy is proposed to learn the glioma feature using GAN-augmented MRIs followed by real MRIs. To evaluate the effectiveness of the proposed scheme, experiments have been conducted on a brain tumor dataset for classifying glioma molecular subtypes: isocitrate dehydrogenase 1 (IDH1) mutation and IDH1 wild-type. Our results on the dataset have shown good performance (with test accuracy 88.82%). Comparisons with several state-of-the-art methods are also included.
|
|
9. |
- Y Banaem, Hossein, et al.
(author)
-
Brain tumor modeling : glioma growth and interaction with chemotherapy
- 2011
-
In: International Conference on Graphic and Image Processing (ICGIP 2011). - : SPIE. ; 8285
-
Conference paper (peer-reviewed)abstract
- In last decade increasingly mathematical models of tumor growths have been studied, particularly on solid tumors which growth mainly caused by cellular proliferation. In this paper we propose a modified model to simulate the growth of gliomas in different stages. Glioma growth is modeled by a reaction-advection-diffusion. We begin with a model of untreated gliomas and continue with models of polyclonal glioma following chemotherapy. From relatively simple assumptions involving homogeneous brain tissue bounded by a few gross anatomical landmarks (ventricles and skull) the models have been expanded to include heterogeneous brain tissue with different motilities of glioma cells in grey and white matter. Tumor growth is characterized by a dangerous change in the control mechanisms, which normally maintain a balance between the rate of proliferation and the rate of apoptosis (controlled cell death). Result shows that this model closes to clinical finding and can simulate brain tumor behavior properly.
|
|
10. |
- Hamid Muhammed, Hamed
(author)
-
Optimization of radiation doses in panoramic X-ray examination using automated image processing
- 2014
-
In: IST 2014 - 2014 IEEE International Conference on Imaging Systems and Techniques, Proceedings. - 9781479967483 ; , s. 361-364
-
Conference paper (peer-reviewed)abstract
- Radiological techniques based on X-rays are well established in medical diagnostics and there are known risks associated with the use of ionizing radiation like X-rays. That explains why the X-ray technology is constantly under development in the pursuit of new technologies that can contribute to reduce radiation dose to patients. Since the reduction of a radiation dose generally results in a poorer image quality, we have investigated whether the use of digital image processing can provide panoramic radiographs with enhanced image quality. An automated image processing algorithm was proposed and employed for this purpose. Panoramic X-ray examination is an important and common tool in dental radiology, used especially for children and teenagers. The technique is used to create an overview of a patient's jaw.
|
|
11. |
- del Aguila Pla, Pol, 1990-
(author)
-
Inverse problems in signal processing : Functional optimization, parameter estimation and machine learning
- 2019
-
Doctoral thesis (other academic/artistic)abstract
- Inverse problems arise in any scientific endeavor. Indeed, it is seldom the case that our senses or basic instruments, i.e., the data, provide the answer we seek. It is only by using our understanding of how the world has generated the data, i.e., a model, that we can hope to infer what the data imply. Solving an inverse problem is, simply put, using a model to retrieve the information we seek from the data.In signal processing, systems are engineered to generate, process, or transmit signals, i.e., indexed data, in order to achieve some goal. The goal of a specific system could be to use an observed signal and its model to solve an inverse problem. However, the goal could also be to generate a signal so that it reveals a parameter to investigation by inverse problems. Inverse problems and signal processing overlap substantially, and rely on the same set of concepts and tools. This thesis lies at the intersection between them, and presents results in modeling, optimization, statistics, machine learning, biomedical imaging and automatic control.The novel scientific content of this thesis is contained in its seven composing publications, which are reproduced in Part II. In five of these, which are mostly motivated by a biomedical imaging application, a set of related optimization and machine learning approaches to source localization under diffusion and convolutional coding models are presented. These are included in Publications A, B, E, F and G, which also include contributions to the modeling and simulation of a specific family of image-based immunoassays. Publication C presents the analysis of a system for clock synchronization between two nodes connected by a channel, which is a problem of utmost relevance in automatic control. The system exploits a specific node design to generate a signal that enables the estimation of the synchronization parameters. In the analysis, substantial contributions to the identifiability of sawtooth signal models under different conditions are made. Finally, Publication D brings to light and proves results that have been largely overlooked by the signal processing community and characterize the information that quantized linear models contain about their location and scale parameters.
|
|
12. |
- Saboori, Arash, et al.
(author)
-
Unsupervised segmentation and data augmentation in image sequences of skeletal muscle contraction by cycle-consistent generative adversarial network
- 2023
-
In: 2023 international conference on modeling, simulation & intelligent computing (MoSICom). - : IEEE. - 9798350393415 - 9798350393422 ; , s. 474-479
-
Conference paper (peer-reviewed)abstract
- This paper investigates a method addressing the unsupervised segmentation and joint data augmentation in medical ultrasound imaging based on the modified CycleGAN. Accurate quantification of fascia and muscle is the key for the diagnostics of neuromuscular disorders based on the analysis of image sequences of skeletal muscle contraction. Although the Deep Learning (DL) models represent encouraging results, some challenges exist. The traditional models don't consider the complex interaction between tissues within a muscle and its surroundings, which reduces the performance of the fascia segmentation. Also, the DL requires many annotated datasets, which ignores dealing with noisy and complex ultrasound images. To overcome these issues, we propose a method to generate realistic images, and then present an unsupervised fascia segmentation method. The results show that our method improves the segmentation accuracy in noisy and complex ultrasound images compared to the traditional methods.
|
|
13. |
- Alonso-Fernandez, Fernando, 1978-, et al.
(author)
-
Very Low-Resolution Iris Recognition Via Eigen-Patch Super-Resolution and Matcher Fusion
- 2016
-
In: 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS). - Piscataway : IEEE. - 9781467397339 - 9781467397346
-
Conference paper (peer-reviewed)abstract
- Current research in iris recognition is moving towards enabling more relaxed acquisition conditions. This has effects on the quality of acquired images, with low resolution being a predominant issue. Here, we evaluate a super-resolution algorithm used to reconstruct iris images based on Eigen-transformation of local image patches. Each patch is reconstructed separately, allowing better quality of enhanced images by preserving local information. Contrast enhancement is used to improve the reconstruction quality, while matcher fusion has been adopted to improve iris recognition performance. We validate the system using a database of 1,872 near-infrared iris images. The presented approach is superior to bilinear or bicubic interpolation, especially at lower resolutions, and the fusion of the two systems pushes the EER to below 5% for down-sampling factors up to a image size of only 13×13.
|
|
14. |
- Yousefi-Banaem, Hossein, et al.
(author)
-
An improved spatial FCM algorithm for cardiac image segmentation
- 2013
-
In: 2013 13th Iranian Conference on Fuzzy Systems (IFSC). - Piscataway, NJ : IEEE. - 9781479912285 ; , s. 1-4
-
Conference paper (peer-reviewed)abstract
- Image segmentation is one of challenging field in medical image processing. Segmentation of cardiac wall is one of challenging work and it is very important step in evaluation of heart functionality by existing methods. For cardiac image analysis, Fuzzy C- Means (FCM) algorithm proved to be superior over the other clustering approaches in segmentation field. However, the nave FCM algorithm is sensitive to noise because of not considering the spatial information in the image. In this paper an improved FCM algorithm is formulated by incorporating the spatial domain neighborhood information into the membership function for clustering (ISFCM). In this paper we applied improved Fuzzy c-Means with spatial information for left ventricular wall segmentation. Obtained results showed that the proposed method can segment cardiac wall automatically with acceptable accuracy. The comparison of proposed method with nave FCM proved that ISFCM can segment with more accuracy than nave FCM.
|
|
15. |
- Qiao, Sibo, et al.
(author)
-
A Pseudo-Siamese Feature Fusion Generative Adversarial Network for Synthesizing High-quality Fetal Four-chamber Views
- 2023
-
In: IEEE journal of biomedical and health informatics. - : Institute of Electrical and Electronics Engineers (IEEE). - 2168-2194 .- 2168-2208. ; 27:3, s. 1193-1204
-
Journal article (peer-reviewed)abstract
- Four-chamber (FC) views are the primary ultrasound (US) images that cardiologists diagnose whether the fetus has congenital heart disease (CHD) in prenatal diagnosis and screening. FC views intuitively depict the developmental morphology of the fetal heart. Early diagnosis of fetal CHD has always been the focus and difficulty of prenatal screening. Furthermore, deep learning technology has achieved great success in medical image analysis. Hence, applying deep learning technology in the early screening of fetal CHD helps improve diagnostic accuracy. However, the lack of large-scale and high-quality fetal FC views brings incredible difficulties to deep learning models or cardiologists. Hence, we propose a Pseudo-Siamese Feature Fusion Generative Adversarial Network (PSFFGAN), synthesizing high-quality fetal FC views using FC sketch images. In addition, we propose a novel Triplet Generative Adversarial Loss Function (TGALF), which optimizes PSFFGAN to fully extract the cardiac anatomical structure information provided by FC sketch images to synthesize the corresponding fetal FC views with speckle noises, artifacts, and other ultrasonic characteristics. The experimental results show that the fetal FC views synthesized by our proposed PSFFGAN have the best objective evaluation values: SSIM of 0.4627, MS-SSIM of 0.6224, and FID of 83.92, respectively. More importantly, two professional cardiologists evaluate healthy FC views and CHD FC views synthesized by our PSFFGAN, giving a subjective score that the average qualified rate is 82% and 79%, respectively, which further proves the effectiveness of the PSFFGAN.
|
|
16. |
- Khodadad, Davood, 1985-, et al.
(author)
-
Optimized breath detection algorithm in electrical impedance tomography
- 2018
-
In: Physiological Measurement. - : IOP Publishing. - 0967-3334 .- 1361-6579. ; 39:9
-
Journal article (peer-reviewed)abstract
- Objective: This paper defines a method for optimizing the breath delineation algorithms used in electrical impedance tomography (EIT). In lung EIT the identification of the breath phases is central for generating tidal impedance variation images, subsequent data analysis and clinical evaluation. The optimisation of these algorithms is particularly important in neonatal care since the existing breath detectors developed for adults may give insufficient reliability in neonates due to their very irregular breathing pattern.Approach: Our approach is generic in the sense that it relies on the definition of a gold standard and the associated definition of detector sensitivity and specificity, an optimisation criterion and a set of detector parameters to be investigated. The gold standard has been defined by 11 clinicians with previous experience with EIT and the performance of our approach is described and validated using a neonatal EIT dataset acquired within the EU-funded CRADL project.Main results: Three different algorithms are proposed that improve the breath detector performance by adding conditions on (1) maximum tidal breath rate obtained from zero-crossings of the EIT breathing signal, (2) minimum tidal impedance amplitude and (3) minimum tidal breath rate obtained from time-frequency analysis. As a baseline a zero-crossing algorithm has been used with some default parameters based on the Swisstom EIT device.Significance: Based on the gold standard, the most crucial parameters of the proposed algorithms are optimised by using a simple exhaustive search and a weighted metric defined in connection with the receiver operating characterics. This provides a practical way to achieve any desirable trade-off between the sensitivity and the specificity of the detectors.
|
|
17. |
- Lei, Xiangyu, 1992, et al.
(author)
-
Model-Based Parametric Study of Surface-Breaking Defect Characterization Using Half-Skip Total Focusing Method
- 2023
-
In: Journal of nondestructive evaluation. - : Springer Nature. - 0195-9298 .- 1573-4862. ; 42:2
-
Journal article (peer-reviewed)abstract
- As the demand of structural integrity in manufacturing industries is increasing, the ultrasonic array technique has drawn more attention thanks to its inspection flexibility and versatility. By taking advantage of the possibility of individual triggering of each array element, full matrix capture (FMC) data acquisition strategy has been developed that contains the entire information of an inspection scenario. Total focusing method (TFM) as one of the ultrasonic imaging algorithms, is preferably applied to FMC dataset since it uses all information in FMC to synthetically focus the sound energy at every image pixel in the region of interest. Half-skip TFM (HSTFM) is proposed in multi-mode TFM imaging that involves a backwall reflection wave path, so that the defect profile could be reconstructed for accurate defect characterization. In this paper, a method involving Snell’s law-based wave mode conversion is proposed to account for more reasonable wave propagation time when wave mode conversion happens at backwall reflection in HSTFM. A series of model based simulations (in software simSUNDT) are performed for parametric studies, with the intention of investigating the capability of defect characterization using HSTFM with varying tilt angle and relative position of surface-breaking notch to array probe. The results show that certain TFM modes could help with defect characterization, but the effectiveness is limited with varying defect features. It is inappropriate to address a certain mode for all characterization perspectives but rather a combination, i.e., multi-mode TFM, should be adopted for possible interpretation and characterization of defect features.
|
|
18. |
- Lekamlage, Charitha Dissanayake, et al.
(author)
-
Mini-DDSM : Mammography-based Automatic Age Estimation
- 2020
-
In: ACM International Conference Proceeding Series. - New York, NY, USA : Association for Computing Machinery. - 9781450389044 ; , s. 1-6
-
Conference paper (peer-reviewed)abstract
- Age estimation has attracted attention for its various medical applications. There are many studies on human age estimation from biomedical images. However, there is no research done on mammograms for age estimation, as far as we know. The purpose of this study is to devise an AI-based model for estimating age from mammogram images. Due to lack of public mammography data sets that have the age attribute, we resort to using a web crawler to download thumbnail mammographic images and their age fields from the public data set; the Digital Database for Screening Mammography. The original images in this data set unfortunately can only be retrieved by a software which is broken. Subsequently, we extracted deep learning features from the collected data set, by which we built a model using Random Forests regressor to estimate the age automatically. The performance assessment was measured using the mean absolute error values. The average error value out of 10 tests on random selection of samples was around 8 years. In this paper, we show the merits of this approach to fill up missing age values. We ran logistic and linear regression models on another independent data set to further validate the advantage of our proposed work. This paper also introduces the free-access Mini-DDSM data set. © 2020 ACM.
|
|
19. |
- Hallström, Erik, et al.
(author)
-
Label-free deep learning-based species classification of bacteria imaged by phase-contrast microscopy
- 2023
-
In: PloS Computational Biology. - : Public Library of Science (PLoS). - 1553-734X .- 1553-7358. ; 19:11
-
Journal article (peer-reviewed)abstract
- Reliable detection and classification of bacteria and other pathogens in the human body, animals, food, and water is crucial for improving and safeguarding public health. For instance, identifying the species and its antibiotic susceptibility is vital for effective bacterial infection treatment. Here we show that phase contrast time-lapse microscopy combined with deep learning is sufficient to classify four species of bacteria relevant to human health. The classification is performed on living bacteria and does not require fixation or staining, meaning that the bacterial species can be determined as the bacteria reproduce in a microfluidic device, enabling parallel determination of susceptibility to antibiotics. We assess the performance of convolutional neural networks and vision transformers, where the best model attained a class-average accuracy exceeding 98%. Our successful proof-of-principle results suggest that the methods should be challenged with data covering more species and clinically relevant isolates for future clinical use. Bacterial infections are a leading cause of premature death worldwide, and growing antibiotic resistance is making treatment increasingly challenging. To effectively treat a patient with a bacterial infection, it is essential to quickly detect and identify the bacterial species and determine its susceptibility to different antibiotics. Prompt and effective treatment is crucial for the patient's survival. A microfluidic device functions as a miniature "lab-on-chip" for manipulating and analyzing tiny amounts of fluids, such as blood or urine samples from patients. Microfluidic chips with chambers and channels have been designed for quickly testing bacterial susceptibility to different antibiotics by analyzing bacterial growth. Identifying bacterial species has previously relied on killing the bacteria and applying species-specific fluorescent probes. The purpose of the herein proposed species identification is to speed up decisions on treatment options by already in the first few imaging frames getting an idea of the bacterial species, without interfering with the ongoing antibiotics susceptibility testing. We introduce deep learning models as a fast and cost-effective method for identifying bacteria species. We envision this method being employed concurrently with antibiotic susceptibility tests in future applications, significantly enhancing bacterial infection treatments.
|
|
20. |
|
|
21. |
- Ge, Chenjie, 1991, et al.
(author)
-
Multiscale Deep Convolutional Networks for Characterization and Detection of Alzheimer's Disease using MR Images
- 2019
-
In: Proceedings - International Conference on Image Processing, ICIP. - 1522-4880. ; 2019-September, s. 789-793
-
Conference paper (peer-reviewed)abstract
- This paper addresses the issues of Alzheimer's disease (AD) characterization and detection from Magnetic Resonance Images (MRIs). Many existing AD detection methods use single-scale feature learning from brain scans. In this paper, we propose a multiscale deep learning architecture for learning AD features. The main contributions of the paper include: (a) propose a novel 3D multiscale CNN architecture for the dedicated task of AD detection; (b) propose a feature fusion and enhancement strategy for multiscale features; (c) empirical study on the impact of several settings, including two dataset partitioning approaches, and the use of multiscale and feature enhancement. Experiments were conducted on an open ADNI dataset (1198 brain scans from 337 subjects), test results have shown the effectiveness of the proposed method with test accuracy of 93.53%, 87.24% (best, average) on subject separated dataset, and 99.44%, 98.80% (best, average) on random brain scan-partitioned dataset. Comparison with eight existing methods has provided further support to the proposed method.
|
|
22. |
- Kothapalli, Satya V.V.N. 1985-
(author)
-
Nano-Engineered Contrast Agents : Toward Multimodal Imaging and Acoustophoresis
- 2015
-
Doctoral thesis (other academic/artistic)abstract
- Diagnostic ultrasound (US) is safer, quicker and cheaper than other diagnostic imaging modalities. Over the past two decades, the applications of US imaging has been widened due to the development of injectable, compressible and encapsulated microbubbles (MBs) that provide an opportunity to improve conventional echocardiographic imaging, blood flow assessment and molecular imaging. The encapsulating material is manufactured by different biocompatible materials such as proteins, lipids or polymers. In current research, researchers modify the encapsulated shell with the help of advanced molecular chemistry techniques to load them with dyes (for fluorescent imaging), nanoparticles and radioisotopes (for multimodal imaging) or functional ligands or therapeutic gases (for local drug delivery). The echogenicity and the radial oscillation of MBs is the result of their compressibility, which undoubtedly varies with the encapsulated shell characteristics such as rigidity or elasticity.In this thesis, we present acoustic properties of novel type of polyvinyl alcohol (PVA)-shelled microbubble (PVA-MB) that was further modified with superparamagnetic iron oxide nanoparticles (SPIONs) to work as a dual-modal contrast agent for magnetic resonance (MR) imaging along with US imaging. Apparently, the shell modification changes their mechanical characteristics, which affects their acoustic properties. The overall objective of the thesis is to investigate the acoustic properties of modified and unmodified PVA-MBs at different ultrasound parameters.The acoustic and mechanical characterization of SPIONs modified PVA-MBs revealed that the acoustical response depends on the SPION inclusion strategy. However they retain the same structural characteristics after the modification. The modified MBs with SPIONs included on the surface of the PVA shell exhibit a soft-shelled behavior and produce a higher echogenicity than the MBs with the SPIONs inside the PVA shell. The fracturing mechanism of the unmodified PVA-MBs was identified to be different from the other fracturing mechanisms of conventional MBs. With the interaction of high-pressure bursts, the air gas core is squeezed out through small punctures in the PVA shell. During the fracturing, the PVA-MBs exhibit asymmetric (other modes) oscillations, resulting in sub- and ultra-harmonic generation. Exploiting the US imaging at the other modes of the oscillation of the PVA-MBs would provide an opportunity to visualize very low concentrations of (down to single) PVA-MBs. We further introduced the PVA-MBs along with particles mimicking red blood cells in an acoustic standing-wave field to observe the acoustic radiation force effect. We observed that the compressible PVA-MBs drawn toward pressure antinode while the solid blood phantoms moved toward the pressure node. This acoustic separation method (acoustophoresis) could be an efficient tool for studying the bioclearance of the PVA-MBs in the body, either by collecting blood samples (in-vitro) or by using the extracorporeal medical procedure (ex-vivo) at different organs.Overall, this work contributes significant feedback for chemists (to optimize the nanoparticle inclusion) and imaging groups (to develop new imaging sequences), and the positive findings pave new paths and provide triggers to engage in further research.
|
|
23. |
- Mignardi, Marco, et al.
(author)
-
Bridging Histology and Bioinformatics : Computational analysis of spatially resolved transcriptomics
- 2017
-
In: Proceedings of the IEEE. - 0018-9219 .- 1558-2256. ; 105:3, s. 530-541
-
Journal article (peer-reviewed)abstract
- It is well known that cells in tissue display a large heterogeneity in gene expression due to differences in cell lineage origin and variation in the local environment. Traditional methods that analyze gene expression from bulk RNA extracts fail to accurately describe this heterogeneity because of their intrinsic limitation in cellular and spatial resolution. Also, information on histology in the form of tissue architecture and organization is lost in the process. Recently, new transcriptome-wide analysis technologies have enabled the study of RNA molecules directly in tissue samples, thus maintaining spatial resolution and complementing histological information with molecular information important for the understanding of many biological processes and potentially relevant for the clinical management of cancer patients. These new methods generally comprise three levels of analysis. At the first level, biochemical techniques are used to generate signals that can be imaged by different means of fluorescence microscopy. At the second level, images are subject to digital image processing and analysis in order to detect and identify the aforementioned signals. At the third level, the collected data are analyzed and transformed into interpretable information by statistical methods and visualization techniques relating them to each other, to spatial distribution, and to tissue morphology. In this review, we describe state-of-the-art techniques used at all three levels of analysis. Finally, we discuss future perspective in this fast-growing field of spatially resolved transcriptomics.
|
|
24. |
- Qiao, Sibo, et al.
(author)
-
HCMMNet : Hierarchical Conv-MLP-Mixed Network for Medical Image Segmentation in Metaverse for Consumer Health
- 2024
-
In: IEEE transactions on consumer electronics. - : Institute of Electrical and Electronics Engineers (IEEE). - 0098-3063 .- 1558-4127. ; 70:1, s. 2078-2089
-
Journal article (peer-reviewed)abstract
- In the burgeoning metaverse for consumer health (MCH), medical image segmentation methods with high accuracy and generalization capability are essential to drive personalized healthcare solutions and enhance the patient experience. To address the inherent challenges of capturing complex structures and features in medical image segmentation, we propose a convolutional neural network (CNN) and multi-layer-perceptron (MLP) mixed module named HCMM, which hierarchically incorporates local priors of CNN into fully-connected (FC) layers, ingeniously capturing specific details and a broader range of contextual information of the focused object from diverse perspectives. Then, we propose an MLP-based information fusion module (MIF) designed to dynamically merge feature maps of varying levels from different pathways, enhancing feature expression and discriminative power. Based on the above-proposed modules, we design a novel segmentation model, HCMMNet, which can adeptly capture feature maps from input medical images at different scales and perspectives. Through comparative experiments, we demonstrate the outstanding performance of the HCMMNet for medical image segmentation on three publicly available datasets and one self-organized dataset. Notably, our HCMMNet showcases remarkable efficacy while maintaining an extraordinarily lightweight profile, weighing in at a mere 3M, rendering it ideal for MCH application.
|
|
25. |
|
|
26. |
- Ge, Chenjie, 1991, et al.
(author)
-
3D Multi-Scale Convolutional Networks for Glioma Grading Using MR Images
- 2018
-
In: Proceedings - International Conference on Image Processing, ICIP. - 1522-4880. - 9781479970612 ; , s. 141-145
-
Conference paper (peer-reviewed)abstract
- This paper addresses issues of grading brain tumor, glioma, from Magnetic Resonance Images (MRIs). Although feature pyramid is shown to be useful to extract multi-scale features for object recognition, it is rarely explored in MRI images for glioma classification/grading. For glioma grading, existing deep learning methods often use convolutional neural networks (CNNs) to extract single-scale features without considering that the scales of brain tumor features vary depending on structure/shape, size, tissue smoothness, and locations. In this paper, we propose to incorporate the multi-scale feature learning into a deep convolutional network architecture, which extracts multi-scale semantic as well as fine features for glioma tumor grading. The main contributions of the paper are: (a) propose a novel 3D multi-scale convolutional network architecture for the dedicated task of glioma grading; (b) propose a novel feature fusion scheme that further refines multi-scale features generated from multi-scale convolutional layers; (c) propose a saliency-aware strategy to enhance tumor regions of MRIs. Experiments were conducted on an open dataset for classifying high/low grade gliomas. Performance on the test set using the proposed scheme has shown good results (with accuracy of 89.47%).
|
|
27. |
- Sun, Lilei, et al.
(author)
-
Two-view attention-guided convolutional neural network for mammographic image classification
- 2022
-
In: CAAI Transactions on Intelligence Technology. - : Institution of Engineering and Technology (IET). - 2468-6557 .- 2468-2322.
-
Journal article (peer-reviewed)abstract
- Deep learning has been widely used in the field of mammographic image classification owing to its superiority in automatic feature extraction. However, general deep learning models cannot achieve very satisfactory classification results on mammographic images because these models are not specifically designed for mammographic images and do not take the specific traits of these images into account. To exploit the essential discriminant information of mammographic images, we propose a novel classification method based on a convolutional neural network. Specifically, the proposed method designs two branches to extract the discriminative features from mammographic images from the mediolateral oblique and craniocaudal (CC) mammographic views. The features extracted from the two-view mammographic images contain complementary information that enables breast cancer to be more easily distinguished. Moreover, the attention block is introduced to capture the channel-wise information by adjusting the weight of each feature map, which is beneficial to emphasising the important features of mammographic images. Furthermore, we add a penalty term based on the fuzzy cluster algorithm to the cross-entropy function, which improves the generalisation ability of the classification model by maximising the interclass distance and minimising the intraclass distance of the samples. The experimental results on The Digital database for Screening Mammography INbreast and MIAS mammography databases illustrate that the proposed method achieves the best classification performance and is more robust than the compared state-of-the-art classification methods.
|
|
28. |
- Gómez-de-Mariscal, E., et al.
(author)
-
DeepImageJ : A user-friendly environment to run deep learning models in ImageJ
- 2021
-
In: Nature Methods. - : Springer Nature. - 1548-7091 .- 1548-7105. ; 18:10, s. 1192-1195
-
Journal article (peer-reviewed)abstract
- DeepImageJ is a user-friendly solution that enables the generic use of pre-trained deep learning models for biomedical image analysis in ImageJ. The deepImageJ environment gives access to the largest bioimage repository of pre-trained deep learning models (BioImage Model Zoo). Hence, nonexperts can easily perform common image processing tasks in life-science research with deep learning-based tools including pixel and object classification, instance segmentation, denoising or virtual staining. DeepImageJ is compatible with existing state of the art solutions and it is equipped with utility tools for developers to include new models. Very recently, several training frameworks have adopted the deepImageJ format to deploy their work in one of the most used softwares in the field (ImageJ). Beyond its direct use, we expect deepImageJ to contribute to the broader dissemination and reuse of deep learning models in life sciences applications and bioimage informatics.
|
|
29. |
- Fakhouri, Hussam N., et al.
(author)
-
A cognitive deep learning approach for medical image processing
- 2024
-
In: Scientific Reports. - : Springer Nature. - 2045-2322. ; 14:1
-
Journal article (peer-reviewed)abstract
- In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation. © 2024. The Author(s).
|
|
30. |
- Klintström, Eva, et al.
(author)
-
Trabecular bone microstructure analysis on data from a novel twin robotic X-ray device
- 2023
-
In: Acta Radiologica. - : SAGE Publications. - 0284-1851 .- 1600-0455. ; 64:4, s. 1566-1572
-
Journal article (peer-reviewed)abstract
- Background: Bone strength is related to both mineral density (BMD) and the bone microstructure. However, only the assessment of BMD is available in clinical routine care today. Purpose: To analyze and study the correlation of trabecular bone microstructure from the imaging data of a prototype Multitom Rax system, using micro-computed tomography (CT) data as the reference method (Skyscan 1176).Material and Methods: Imaging data from 14 bone samples from the human radius were analyzed regarding six bone structure parameters, i.e. trabecular nodes, separation, spacing, and thickness as well as bone volume (BV/TV) and structural model index (SMI).Results: All six structure parameters showed strong correlations to micro-CT with Spearman correlation coefficients in the range of 0.83–0.93. BV/TV and SMI had a correlation >0.90. Two of the parameters, namely, separation and number, had mean values in the same range as micro-CT. The other four were either over- or underestimated.Conclusion: The strong correlation between micro-CT and the clinical imaging system, indicates the possibility for analyzing bone microstructure with potential to add value in fracture assessment using the studied device in a clinical workflow.
|
|
31. |
- Svensson, Björn, 1979-
(author)
-
A Multidimensional Filtering Framework with Applications to Local Structure Analysis and Image Enhancement
- 2008
-
Doctoral thesis (other academic/artistic)abstract
- Filtering is a fundamental operation in image science in general and in medical image science in particular. The most central applications are image enhancement, registration, segmentation and feature extraction. Even though these applications involve non-linear processing a majority of the methodologies available rely on initial estimates using linear filters. Linear filtering is a well established cornerstone of signal processing, which is reflected by the overwhelming amount of literature on finite impulse response filters and their design.Standard techniques for multidimensional filtering are computationally intense. This leads to either a long computation time or a performance loss caused by approximations made in order to increase the computational efficiency. This dissertation presents a framework for realization of efficient multidimensional filters. A weighted least squares design criterion ensures preservation of the performance and the two techniques called filter networks and sub-filter sequences significantly reduce the computational demand.A filter network is a realization of a set of filters, which are decomposed into a structure of sparse sub-filters each with a low number of coefficients. Sparsity is here a key property to reduce the number of floating point operations required for filtering. Also, the network structure is important for efficiency, since it determines how the sub-filters contribute to several output nodes, allowing reduction or elimination of redundant computations.Filter networks, which is the main contribution of this dissertation, has many potential applications. The primary target of the research presented here has been local structure analysis and image enhancement. A filter network realization for local structure analysis in 3D shows a computational gain, in terms of multiplications required, which can exceed a factor 70 compared to standard convolution. For comparison, this filter network requires approximately the same amount of multiplications per signal sample as a single 2D filter. These results are purely algorithmic and are not in conflict with the use of hardware acceleration techniques such as parallel processing or graphics processing units (GPU). To get a flavor of the computation time required, a prototype implementation which makes use of filter networks carries out image enhancement in 3D, involving the computation of 16 filter responses, at an approximate speed of 1MVoxel/s on a standard PC.
|
|
32. |
|
|
33. |
- Landré, Jérôme, et al.
(author)
-
A deformable model-based system for 3D analysis and visualization of tumor in PET/CT images
- 2008
-
In: Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology". - : IEEE. - 9781424418145 - 9781424418152 ; , s. 3130-3133
-
Conference paper (peer-reviewed)abstract
- This paper presents a tumor detecting system that allows interactive 3D tumor visualization and tumor volume measurements. An improved level set method is proposed to automatically segment the tumor images slice by slice. PET images are used to detect the tumor while CT images make a 3D representation of the patient's body possible. An initial slice with a seed within the tumor is firstly chosen by the operator. The system then performs automatically the tumor volume segmentation that allows the clinician to visualize the tumor, to measure it and to evaluate the best medical treatment adapted to the patient.
|
|
34. |
- Mendrik, AM, et al.
(author)
-
MRBrainS Challenge: Online Evaluation Framework for Brain Image Segmentation in 3T MRI Scans
- 2015
-
In: Computational Intelligence and Neuroscience. - : Hindawi Publishing Corporation. - 1687-5265 .- 1687-5273. ; 2015
-
Journal article (peer-reviewed)abstract
- Many methods have been proposed for tissue segmentation in brain MRI scans. The multitude of methods proposed complicates the choice of one method above others. We have therefore established the MRBrainS online evaluation framework for evaluating (semi)automatic algorithms that segment gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) on 3T brain MRI scans of elderly subjects (65–80 y). Participants apply their algorithms to the provided data, after which their results are evaluated and ranked. Full manual segmentations of GM, WM, and CSF are available for all scans and used as the reference standard. Five datasets are provided for training and fifteen for testing. The evaluated methods are ranked based on their overall performance to segment GM, WM, and CSF and evaluated using three evaluation metrics (Dice, H95, and AVD) and the results are published on the MRBrainS13 website. We present the results of eleven segmentation algorithms that participated in the MRBrainS13 challenge workshop at MICCAI, where the framework was launched, and three commonly used freeware packages: FreeSurfer, FSL, and SPM. The MRBrainS evaluation framework provides an objective and direct comparison of all evaluated algorithms and can aid in selecting the best performing method for the segmentation goal at hand.
|
|
35. |
- Eklund, Anders, et al.
(author)
-
Using Real-Time fMRI to Control a Dynamical System
- 2009
-
In: ISMRM 17th Scientific Meeting & Exhibition. - Linköping : Linköping University Electronic Press.
-
Conference paper (peer-reviewed)abstract
- We present e method for controlling a dynamical system using real-time fMRI. The objective for the subject in the MR scanner is to balance an inverse pendulum by activating the left or right hand or resting. The brain activity is clasified each second by a neural network and the classification is sent to a pendulum simulator to change the state of the pendulum. The state of the inverse pendulum is shown to the subject in a pair of VR goggles. The subject was able to balance the inverse pendulum during a 7 minute test run.
|
|
36. |
- Isaksson-Daun, Johan
(author)
-
A Sound Approach Toward a Mobility Aid for Blind and Low-Vision Individuals
- 2023
-
Doctoral thesis (other academic/artistic)abstract
- Reduced independent mobility of blind and low-vision individuals (BLVIs) cause considerable societal cost, burden on relatives, and reduced quality of life for the individuals, including increased anxiety, depression symptoms, need of assistance, risk of falls, and mortality. Despite the numerous electronic travel aids proposed since at least the 1940’s, along with ever-advancing technology, the mobility issues persist. A substantial reason for this is likely several and severe shortcomings of the field, both in regards to aid design and evaluation.In this work, these shortcomings are addressed with a generic design model called Desire of Use (DoU), which describes the desire of a given user to use an aid for a given activity. It is then applied on mobility of BLVIs (DoU-MoB), to systematically illuminate and structure possibly all related aspects that such an aid needs to aptly deal with, in order for it to become an adequate aid for the objective. These aspects can then both guide user-centered design as well as choice of test methods and measures.One such measure is then demonstrated in the Desire of Use Questionnaire for Mobility of Blind and Low-Vision Individuals (DoUQ-MoB), an aid-agnostic and comprehensive patient-reported outcome measure. The question construction originates from the DoU-MoB to ensure an encompassing focus on mobility of BLVIs, something that has been missing in the field. Since it is aid-agnostic it facilitates aid comparison, which it also actively promotes. To support the reliability of the DoUQ-MoB, it utilizes the best known practices of questionnaire design and has been validated once with eight orientation and mobility professionals, and six BLVIs. Based on this, the questionnaire has also been revised once.To allow for relevant and reproducible methodology, another tool presented herein is a portable virtual reality (VR) system called the Parrot-VR. It uses a hybrid control scheme of absolute rotation by tracking the user’s head in reality, affording intuitive turning; and relative movement where simple button presses on a controller moves the virtual avatar forward and backward, allowing for large-scale traversal while not walking physically. VR provides excellent reproducibility, making various aggregate movement analysis feasible, while it is also inherently safe. Meanwhile, the portability of the system facilitates testing near the participants, substantially increasing the number of potential blind and low-vision recruits for user tests.The thesis also gives a short account on the state of long-term testing in the field; it being short is mainly due to that there is not much to report. It then provides an initial investigation into possible outcome measures for such tests by taking instruments in use by Swedish orientation and mobility professionals as a starting point. Two of these are also piloted in an initial single-session trial with 19 BLVIs, and could plausibly be used for long-term tests after further evaluation.Finally, a discussion is presented regarding the Audomni project — the development of a primary mobility aid for BLVIs. Audomni is a visuo-auditory sensory supplementation device, which aims to take visual information and translate it to sound. A wide field-of-view, 3D-depth camera records the environment, which is then transformed to audio through the sonification algorithms of Audomni, and finally presented in a pair of open-ear headphones that do not block out environmental sounds. The design of Audomni leverages the DoU-MoB to ensure user-centric development and evaluation, in the aim of reaching an aid with such form and function that it grants the users better mobility, while the users still want to use it.Audomni has been evaluated with user tests twice, once in pilot tests with two BLVIs, and once in VR with a heterogenous set of 19 BLVIs, utilizing the Parrot-VR and the DoUQ-MoB. 76 % of responders (13 / 17) answered that it was very or extremely likely that they would want use Audomni along with their current aid. This might be the first result in the field demonstrating a majority of blind and low-vision participants reporting that they actually want to use a new electronic travel aid. This shows promise that eventual long-term tests will demonstrate an increased mobility of blind and low-vision users — the overarching project aim. Such results would ultimately mean that Audomni can become an aid that alleviates societal cost, reduces burden on relatives, and improves users’ quality of life and independence.
|
|
37. |
- Haj-Hosseini, Neda, 1980-, et al.
(author)
-
Early Detection of Oral Potentially Malignant Disorders: A Review on Prospective Screening Methods with Regard to Global Challenges
- 2024
-
In: Journal of Maxillofacial & Oral Surgery. - New Delhi, India : Springer Science and Business Media LLC. - 0972-8279 .- 0974-942X. ; 23:1, s. 23-32
-
Journal article (peer-reviewed)abstract
- Oral cancer is a cancer type that is widely prevalent in low-and middle-income countries with a high mortality rate, and poor quality of life for patients after treatment. Early treatment of cancer increases patient survival, improves quality of life and results in less morbidity and a better prognosis. To reach this goal, early detection of malignancies using technologies that can be used in remote and low resource areas is desirable. Such technologies should be affordable, accurate, and easy to use and interpret. This review surveys different technologies that have the potentials of implementation in primary health and general dental practice, considering global perspectives and with a focus on the population in India, where oral cancer is highly prevalent. The technologies reviewed include both sample-based methods, such as saliva and blood analysis and brush biopsy, and more direct screening of the oral cavity including fluorescence, Raman techniques, and optical coherence tomography. Digitalisation, followed by automated artificial intelligence based analysis, are key elements in facilitating wide access to these technologies, to non-specialist personnel and in rural areas, increasing quality and objectivity of the analysis while simultaneously reducing the labour and need for highly trained specialists.
|
|
38. |
- Kylberg, Gustaf, et al.
(author)
-
Segmentation of virus particle candidates in transmission electron microscopy images
- 2012
-
In: Journal of Microscopy. - : Blackwell Publishing. - 0022-2720 .- 1365-2818. ; 245:2, s. 140-147
-
Journal article (peer-reviewed)abstract
- In this paper, we present an automatic segmentation method that detects virus particles of various shapes in transmission electron microscopy images. The method is based on a statistical analysis of local neighbourhoods of all the pixels in the image followed by an object width discrimination and finally, for elongated objects, a border refinement step. It requires only one input parameter, the approximate width of the virus particles searched for. The proposed method is evaluated on a large number of viruses. It successfully segments viruses regardless of shape, from polyhedral to highly pleomorphic.
|
|
39. |
|
|
40. |
- Yu, X., et al.
(author)
-
Cardiac LGE MRI Segmentation with Cross-Modality Image Augmentation and Improved U-Net
- 2023
-
In: IEEE journal of biomedical and health informatics. - : Institute of Electrical and Electronics Engineers (IEEE). - 2168-2194 .- 2168-2208. ; 27:2, s. 588-597
-
Journal article (peer-reviewed)abstract
- Image segmentation is a challenging problem in imaging informatics, which stems from the intersection of imaging techniques, computer science and biomedicine. In particular, accurate segmentation of cardiac structures in late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) is of great clinical importance for cardiac function assessment and myocardial disease diagnosis. However, it is a well-known challenge due to its special imaging modality and the lack of labeled LGE samples. In this paper, we propose an unsupervised ventricular segmentation algorithm that can perform biventricular segmentation of LGE images in the absence of labeled LGE data. There are two primary modules, the data augmentation procedure and the segmentation network. The easily available annotated balanced-Steady State Free Precession (bSSFP) images are employed for cross-modal data augmentation by image translation, where a single bSSFP image is converted into multiple synthetic LGE images while preserving the original morphological structure. Then, the proposed segmentation network is trained with the synthetic LGE images and used for segmenting real LGE images. Validation experiments demonstrated the effectiveness and advantages of the proposed algorithm.
|
|
41. |
- Lindgren, Erik, et al.
(author)
-
Autoencoder-Based Anomaly Detection in Industrial X-ray Images
- 2021
-
In: <em>Proceedings of 2021 48th Annual Review of Progress in Quantitative Nondestructive Evaluation</em>. Virtual, Online. July 28–30, 2021.. - : ASME Press. - 9780791885529 ; , s. 28-30
-
Conference paper (peer-reviewed)abstract
- Within many quality-critical industries, e.g. the aerospace industry, industrial X-ray inspection is an essential as well as a resource intense part of quality control. Within such industries the X-ray image interpretation is typically still done by humans, therefore, increasing the interpretation automatization would be of great value. We claim, that safe automatic interpretation of industrial X-ray images, requires a robust confidence estimation with respect to out-of-distribution (OOD) data. In this work we have explored if such a confidence estimation can be achieved by comparing input images with a model of the accepted images. For the image model we derived an autoencoder which we trained unsupervised on a public dataset with X-ray images of metal fusion-welds. We achieved a true positive rate at 80–90% at a 4% false positive rate, as well as correctly detected an OOD data example as an anomaly.
|
|
42. |
- Koriakina, Nadezhda, 1991-, et al.
(author)
-
Deep multiple instance learning versus conventional deep single instance learning for interpretable oral cancer detection
- 2024
-
In: PLOS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 19:4 April
-
Journal article (peer-reviewed)abstract
- The current medical standard for setting an oral cancer (OC) diagnosis is histological examination of a tissue sample taken from the oral cavity. This process is time-consuming and more invasive than an alternative approach of acquiring a brush sample followed by cytological analysis. Using a microscope, skilled cytotechnologists are able to detect changes due to malignancy; however, introducing this approach into clinical routine is associated with challenges such as a lack of resources and experts. To design a trustworthy OC detection system that can assist cytotechnologists, we are interested in deep learning based methods that can reliably detect cancer, given only per-patient labels (thereby minimizing annotation bias), and also provide information regarding which cells are most relevant for the diagnosis (thereby enabling supervision and understanding). In this study, we perform a comparison of two approaches suitable for OC detection and interpretation: (i) conventional single instance learning (SIL) approach and (ii) a modern multiple instance learning (MIL) method. To facilitate systematic evaluation of the considered approaches, we, in addition to a real OC dataset with patient-level ground truth annotations, also introduce a synthetic dataset—PAP-QMNIST. This dataset shares several properties of OC data, such as image size and large and varied number of instances per bag, and may therefore act as a proxy model of a real OC dataset, while, in contrast to OC data, it offers reliable per-instance ground truth, as defined by design. PAP-QMNIST has the additional advantage of being visually interpretable for non-experts, which simplifies analysis of the behavior of methods. For both OC and PAP-QMNIST data, we evaluate performance of the methods utilizing three different neural network architectures. Our study indicates, somewhat surprisingly, that on both synthetic and real data, the performance of the SIL approach is better or equal to the performance of the MIL approach. Visual examination by cytotechnologist indicates that the methods manage to identify cells which deviate from normality, including malignant cells as well as those suspicious for dysplasia. We share the code as open source.
|
|
43. |
- Borrelli, P., et al.
(author)
-
Artificial intelligence-aided CT segmentation for body composition analysis: a validation study
- 2021
-
In: European Radiology Experimental. - : Springer Science and Business Media LLC. - 2509-9280. ; 5:1
-
Journal article (peer-reviewed)abstract
- BackgroundBody composition is associated with survival outcome in oncological patients, but it is not routinely calculated. Manual segmentation of subcutaneous adipose tissue (SAT) and muscle is time-consuming and therefore limited to a single CT slice. Our goal was to develop an artificial-intelligence (AI)-based method for automated quantification of three-dimensional SAT and muscle volumes from CT images.MethodsEthical approvals from Gothenburg and Lund Universities were obtained. Convolutional neural networks were trained to segment SAT and muscle using manual segmentations on CT images from a training group of 50 patients. The method was applied to a separate test group of 74 cancer patients, who had two CT studies each with a median interval between the studies of 3days. Manual segmentations in a single CT slice were used for comparison. The accuracy was measured as overlap between the automated and manual segmentations.ResultsThe accuracy of the AI method was 0.96 for SAT and 0.94 for muscle. The average differences in volumes were significantly lower than the corresponding differences in areas in a single CT slice: 1.8% versus 5.0% (p <0.001) for SAT and 1.9% versus 3.9% (p < 0.001) for muscle. The 95% confidence intervals for predicted volumes in an individual subject from the corresponding single CT slice areas were in the order of 20%.Conclusions The AI-based tool for quantification of SAT and muscle volumes showed high accuracy and reproducibility and provided a body composition analysis that is more relevant than manual analysis of a single CT slice.
|
|
44. |
- Ge, Chenjie, 1991, et al.
(author)
-
Multi-Stream Multi-Scale Deep Convolutional Networks for Alzheimer's Disease Detection using MR Images
- 2019
-
In: Neurocomputing. - : Elsevier BV. - 0925-2312 .- 1872-8286. ; 350, s. 60-69
-
Journal article (peer-reviewed)abstract
- This paper addresses the issue of Alzheimer's disease (AD) detection from Magnetic Resonance Images (MRIs). Existing AD detection methods rely on global feature learning from the whole brain scans, while depending on the tissue types, AD related features in dierent tissue regions, e.g. grey matter (GM), white matter (WM), and cerebrospinal uid (CSF), show different characteristics. In this paper, we propose a deep learning method for multi-scale feature learning based on segmented tissue areas. A novel deep 3D multi-scale convolutional network scheme is proposed to generate multi-resolution features for AD detection. The proposed scheme employs several parallel 3D multi-scale convolutional networks, each applying to individual tissue regions (GM, WM and CSF) followed by feature fusions. The proposed fusion is applied in two separate levels: the rst level fusion is applied on different scales within the same tissue region, and the second level is on dierent tissue regions. To further reduce the dimensions of features and mitigate overtting, a feature boosting and dimension reduction method, XGBoost, is utilized before the classication. The proposed deep learning scheme has been tested on a moderate open dataset of ADNI (1198 scans from 337 subjects), with excellent test performance on randomly partitioned datasets (best 99.67%, average 98.29%), and good test performance on subject-separated partitioned datasets (best 94.74%, average 89.51%). Comparisons with state-of-the-art methods are also included.
|
|
45. |
- Brusini, Irene
(author)
-
Methods for the analysis and characterization of brain morphology from MRI images
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Brain magnetic resonance imaging (MRI) is an imaging modality that produces detailed images of the brain without using any ionizing radiation. From a structural MRI scan, it is possible to extract morphological properties of different brain regions, such as their volume and shape. These measures can both allow a better understanding of how the brain changes due to multiple factors (e.g., environmental and pathological) and contribute to the identification of new imaging biomarkers of neurological and psychiatric diseases. The overall goal of the present thesis is to advance the knowledge on how brain MRI image processing can be effectively used to analyze and characterize brain structure.The first two works presented in this thesis are animal studies that primarily aim to use MRI data for analyzing differences between groups of interest. In Paper I, MRI scans from wild and domestic rabbits were processed to identify structural brain differences between these two groups. Domestication was found to significantly reshape brain structure in terms of both regional gray matter volume and white matter integrity. In Paper II, rat brain MRI scans were used to train a brain age prediction model. This model was then tested on both controls and a group of rats that underwent long-term environmental enrichment and dietary restriction. This healthy lifestyle intervention was shown to significantly affect the predicted brain age trajectories by slowing the rats' aging process compared to controls. Furthermore, brain age predicted on young adult rats was found to have a significant effect on survival.Papers III to V are human studies that propose deep learning-based methods for segmenting brain structures that can be severely affected by neurodegeneration. In particular, Papers III and IV focus on U-Net-based 2D segmentation of the corpus callosum (CC) in multiple sclerosis (MS) patients. In both studies, good segmentation accuracy was obtained and a significant correlation was found between CC area and the patient's level of cognitive and physical disability. Additionally, in Paper IV, shape analysis of the segmented CC revealed a significant association between disability and both CC thickness and bending angle. Conversely, in Paper V, a novel method for automatic segmentation of the hippocampus is proposed, which consists of embedding a statistical shape prior as context information into a U-Net-based framework. The inclusion of shape information was shown to significantly improve segmentation accuracy when testing the method on a new unseen cohort (i.e., different from the one used for training). Furthermore, good performance was observed across three different diagnostic groups (healthy controls, subjects with mild cognitive impairment and Alzheimer's patients) that were characterized by different levels of hippocampal atrophy.In summary, the studies presented in this thesis support the great value of MRI image analysis for the advancement of neuroscientific knowledge, and their contribution is mostly two-fold. First, by applying well-established processing methods on datasets that had not yet been explored in the literature, it was possible to characterize specific brain changes and disentangle relevant problems of a clinical or biological nature. Second, a technical contribution is provided by modifying and extending already-existing brain image processing methods to achieve good performance on new datasets.
|
|
46. |
- Gårdhagen, Roland, 1978-, et al.
(author)
-
Assessment of Geometrical Influence on WSS Estimation in the Human Aorta
- 2006
-
In: WSEAS Transactions on Fluid Mechanics. - 1790-5087. ; 4:1, s. 318-326
-
Journal article (peer-reviewed)abstract
- Computational fluid dynamics simulations were performed on a stenosed human aorta with poststenotic dilatation, in order to estimate wall shear stress (WSS). WSS is important due to its correlation with atherosclerosis. Both steady-state and non-stationary simulations were conducted. Three different models were created from a set of MRI images. Comparison of geometrically different models was accomplished by using geometrical landmarks and a comparison parameter. Geometrical differences had larger influence on WSS magnitude than inflow rotation in steady-state results for the models used. In non-stationary flow the largest differences in WSS are found when the flow velocity near the wall is low e.g. when the inflow is low or in recirculation regions.
|
|
47. |
- Sund, Patrik, et al.
(author)
-
Detection of low contrast test patterns on an LCD with different luminance and illuminance settings
- 2008
-
In: Progress in Biomedical Optics and Imaging - Proceedings of SPIE. - : SPIE. - 1605-7422. ; 6917
-
Journal article (other academic/artistic)abstract
- The DICOM part 14 grayscale standard display function provides one way of harmonizing image appearance under different monitor luminance settings. This function is based on ideal observer conditions where the eye is always adapted to the target luminance and thereby also at peak contrast sensitivity. Clinical workstations are however often exposed to variations in ambient light due to a sub-optimal reading room light environment. Also, clinical images are inhomogeneous and low-contrast patterns must be detected even at luminance levels that differ from the eye adaptation level. All deviations from ideal luminance conditions cause the observer to detect patterns with reduced eye sensitivity but the magnitude of this reduction is unclear. The purpose of this paper was to quantify the effect different luminance settings have on the contrast threshold. A method to display well-defined sinusoidal low-contrast test patterns on an LCD has previously been developed and was used in this study. The observers were exposed to light from three different areas: 1) A small sinusoidal test pattern. 2) The remaining of the display surface. 3) Ambient light from outside the display area covering most of the observer's field of view. By adjusting the luminance from each of these three areas, two major effects could be quantified. The first effect was similar to Barten's f-factor where the target luminance differs from the observer's adaptation level while the second effect concerned the influence of areas outside the display surface. When a luminance range of 1-350 cd/m2 was used, the contrast needed to detect a dark object in a gray surrounding was almost doubled compared to a dark object in a dark surrounding. Ambient light from outside the display area has a moderate effect on the contrast threshold, except for the combination of high ambient light and dark objects where the contrast threshold increased considerably.
|
|
48. |
- Wetzer, Elisabeth
(author)
-
Representation Learning and Information Fusion : Applications in Biomedical Image Processing
- 2023
-
Doctoral thesis (other academic/artistic)abstract
- In recent years Machine Learning and in particular Deep Learning have excelled in object recognition and classification tasks in computer vision. As these methods extract features from the data itself by learning features that are relevant for a particular task, a key aspect of this remarkable success is the amount of data on which these methods train. Biomedical applications face the problem that the amount of training data is limited. In particular, labels and annotations are usually scarce and expensive to obtain as they require biological or medical expertise. One way to overcome this issue is to use additional knowledge about the data at hand. This guidance can come from expert knowledge, which puts focus on specific, relevant characteristics in the images, or geometric priors which can be used to exploit the spatial relationships in the images. This thesis presents machine learning methods for visual data that exploit such additional information and build upon classic image processing techniques, to combine the strengths of both model- and learning-based approaches. The thesis comprises five papers with applications in digital pathology. Two of them study the use and fusion of texture features within convolutional neural networks for image classification tasks. The other three papers study rotational equivariant representation learning, and show that learned, shared representations of multimodal images can be used for multimodal image registration and cross-modality image retrieval.
|
|
49. |
- Khodadad, Davood, 1985-, et al.
(author)
-
B-spline based free form deformation thoracic non-rigid registration of CT and PET images
- 2011
-
In: International Conference on Graphic and Image Processing (ICGIP 2011). - : SPIE - International Society for Optical Engineering. ; 8285
-
Conference paper (peer-reviewed)abstract
- Accurate attenuation correction of emission data is mandatory for quantitative analysis of PET images. One of the main concerns in CT-based attenuation correction(CTAC) of PET data in multimodality PET/CT imaging is misalignment between PET and CT images. The aim of this study, is to proposed a hybrid method which is simple, fast and accurate, for registration of PET and CT data which affected from respiratory motion in order to improve the quality of CTAC. The algorithm is composed of three methods: First, using B-spline Free Form Deformation to describe both images and deformation field. Then applying a pre-filtering on both PET and CT images before segmentation of structures in order to reduce the respiratory related attenuation correction artifacts of PET emission data. In this approach, B-spline using FFD provide more accurate adaptive transformation to align the images, and structure constraints obtained from prefiltering applied to guide the algorithm to be more fast and accurate. Also it helps to reduce the radiation dose in PET/CT by avoiding repetition of CT imaging. These advances increase the potential of the method for routine clinical application.
|
|
50. |
- Fredén Jansson, Karl-Johan, 1988
(author)
-
The Balanced Electromagnetic Separation Transducer for Bone Conduction Audiometry and Hearing Rehabilitation
- 2017
-
Doctoral thesis (other academic/artistic)abstract
- Hearing via air conduction (AC) and bone conduction (BC) are attributed to bethe natural ways of conducting sound to the cochlea. With AC hearing, air pressurevariations are transmitted to the cochlea via the ear canal, whereas with BChearing, sound vibrations are transmitted through the skull bone to the cochlea.Patients with a hearing loss in the cochlea or auditory nerve are commonly rehabilitatedwith conventional AC hearing aids in the ear canal, but also using cochlearimplants. If the pathway for AC sound to reach the cochlea is obstructed, patientscan often benet from bone conduction devices (BCDs). In order to determinethe type and degree of hearing loss, the BC hearing thresholds are measured usinga bone conduction vibrator, and then analyzed together with the AC hearingthresholds for the diagnosis and to suggest an appropriate rehabilitation alternative.The motor unit in conventional BCDs and bone vibrators are known togenerate high amount of distortion at low frequencies where the Balanced ElectromagneticSeparation Transducer (BEST) principle may oer a new era in BChearing rehabilitation and audiometry.This thesis combines two BC hearing related topics, where the rst topic is anevaluation of a new audiometric bone vibrator, Radioear B81, which is assumedto oer more accurate BC hearing threshold measurements. The second topic isrelated to a new type of active transcutaneous BCD, called the Bone ConductionImplant (BCI), which leaves the skin intact by using a wireless solution thatdoes not require a permanent skin penetration. Even though the applications aredierent, both devices use the BEST principle as motor unit in their design.The audiometric bone vibrator Radioear B81 was found to have an improvedperformance at low frequencies where it can produce higher output levels with lessharmonic distortion than the conventional Radioear B71. In a clinical study of therst six patients, the BCI was found as ecient as already commercially availableBCDs, and with the advantage of not needing a skin penetration. In a technicalevaluation of the BCI, it was shown to be a mechanically robust design and totolerate magnetic resonance imaging at 1.5 Tesla.
|
|