| 1. |
- Astaraki, Mehdi, PhD Student, 1984-
(author)
-
Advanced Machine Learning Methods for Oncological Image Analysis
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Cancer is a major public health problem, accounting for an estimated 10 million deaths worldwide in 2020 alone. Rapid advances in the field of image acquisition and hardware development over the past three decades have resulted in the development of modern medical imaging modalities that can capture high-resolution anatomical, physiological, functional, and metabolic quantitative information from cancerous organs. Therefore, the applications of medical imaging have become increasingly crucial in the clinical routines of oncology, providing screening, diagnosis, treatment monitoring, and non/minimally-invasive evaluation of disease prognosis. The essential need for medical images, however, has resulted in the acquisition of a tremendous number of imaging scans. Considering the growing role of medical imaging data on one side and the challenges of manually examining such an abundance of data on the other side, the development of computerized tools to automatically or semi-automatically examine the image data has attracted considerable interest. Hence, a variety of machine learning tools have been developed for oncological image analysis, aiming to assist clinicians with repetitive tasks in their workflow.This thesis aims to contribute to the field of oncological image analysis by proposing new ways of quantifying tumor characteristics from medical image data. Specifically, this thesis consists of six studies, the first two of which focus on introducing novel methods for tumor segmentation. The last four studies aim to develop quantitative imaging biomarkers for cancer diagnosis and prognosis.The main objective of Study I is to develop a deep learning pipeline capable of capturing the appearance of lung pathologies, including lung tumors, and integrating this pipeline into the segmentation networks to leverage the segmentation accuracy. The proposed pipeline was tested on several comprehensive datasets, and the numerical quantifications show the superiority of the proposed prior-aware DL framework compared to the state of the art. Study II aims to address a crucial challenge faced by supervised segmentation models: dependency on the large-scale labeled dataset. In this study, an unsupervised segmentation approach is proposed based on the concept of image inpainting to segment lung and head-neck tumors in images from single and multiple modalities. The proposed autoinpainting pipeline shows great potential in synthesizing high-quality tumor-free images and outperforms a family of well-established unsupervised models in terms of segmentation accuracy.Studies III and IV aim to automatically discriminate the benign from the malignant pulmonary nodules by analyzing the low-dose computed tomography (LDCT) scans. In Study III, a dual-pathway deep classification framework is proposed to simultaneously take into account the local intra-nodule heterogeneities and the global contextual information. Study IV seeks to compare the discriminative power of a series of carefully selected conventional radiomics methods, end-to-end Deep Learning (DL) models, and deep features-based radiomics analysis on the same dataset. The numerical analyses show the potential of fusing the learned deep features into radiomic features for boosting the classification power.Study V focuses on the early assessment of lung tumor response to the applied treatments by proposing a novel feature set that can be interpreted physiologically. This feature set was employed to quantify the changes in the tumor characteristics from longitudinal PET-CT scans in order to predict the overall survival status of the patients two years after the last session of treatments. The discriminative power of the introduced imaging biomarkers was compared against the conventional radiomics, and the quantitative evaluations verified the superiority of the proposed feature set. Whereas Study V focuses on a binary survival prediction task, Study VI addresses the prediction of survival rate in patients diagnosed with lung and head-neck cancer by investigating the potential of spherical convolutional neural networks and comparing their performance against other types of features, including radiomics. While comparable results were achieved in intra-dataset analyses, the proposed spherical-based features show more predictive power in inter-dataset analyses.In summary, the six studies incorporate different imaging modalities and a wide range of image processing and machine-learning techniques in the methods developed for the quantitative assessment of tumor characteristics and contribute to the essential procedures of cancer diagnosis and prognosis.
|
|
| 2. |
- Bilic, Patrick, et al.
(author)
-
The Liver Tumor Segmentation Benchmark (LiTS)
- 2023
-
In: Medical Image Analysis. - : Elsevier BV. - 1361-8415 .- 1361-8423. ; 84, s. 102680-
-
Journal article (peer-reviewed)abstract
- In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.
|
|
| 3. |
- Jimenez-del-Toro, Oscar, et al.
(author)
-
Cloud-Based Evaluation of Anatomical Structure Segmentation and Landmark Detection Algorithms : VISCERAL Anatomy Benchmarks
- 2016
-
In: IEEE Transactions on Medical Imaging. - : Institute of Electrical and Electronics Engineers (IEEE). - 0278-0062 .- 1558-254X. ; 35:11, s. 2459-2475
-
Journal article (peer-reviewed)abstract
- Variations in the shape and appearance of anatomical structures in medical images are often relevant radiological signs of disease. Automatic tools can help automate parts of this manual process. A cloud-based evaluation framework is presented in this paper including results of benchmarking current state-of-the-art medical imaging algorithms for anatomical structure segmentation and landmark detection: the VISCERAL Anatomy benchmarks. The algorithms are implemented in virtual machines in the cloud where participants can only access the training data and can be run privately by the benchmark administrators to objectively compare their performance in an unseen common test set. Overall, 120 computed tomography and magnetic resonance patient volumes were manually annotated to create a standard Gold Corpus containing a total of 1295 structures and 1760 landmarks. Ten participants contributed with automatic algorithms for the organ segmentation task, and three for the landmark localization task. Different algorithms obtained the best scores in the four available imaging modalities and for subsets of anatomical structures. The annotation framework, resulting data set, evaluation setup, results and performance analysis from the three VISCERAL Anatomy benchmarks are presented in this article. Both the VISCERAL data set and Silver Corpus generated with the fusion of the participant algorithms on a larger set of non-manually-annotated medical images are available to the research community.
|
|
| 4. |
- Kaushik, Sandeep S., et al.
(author)
-
Region of interest focused MRI to synthetic CT translation using regression and segmentation multi-task network
- 2023
-
In: Physics in Medicine and Biology. - : Institute of Physics (IOP). - 0031-9155 .- 1361-6560. ; 68:19
-
Journal article (peer-reviewed)abstract
- Objective: In MR-only clinical workflow, replacing CT with MR image is of advantage for workflow efficiency and reduces radiation to the patient. An important step required to eliminate CT scan from the workflow is to generate the information provided by CT via an MR image. In this work, we aim to demonstrate a method to generate accurate synthetic CT (sCT) from an MR image to suit the radiation therapy (RT) treatment planning workflow. We show the feasibility of the method and make way for a broader clinical evaluation.Approach: We present a machine learning method for sCT generation from zero-echo-time (ZTE) MRI aimed at structural and quantitative accuracies of the image, with a particular focus on the accurate bone density value prediction. The misestimation of bone density in the radiation path could lead to unintended dose delivery to the target volume and results in suboptimal treatment outcome. We propose a loss function that favors a spatially sparse bone region in the image. We harness the ability of the multi-task network to produce correlated outputs as a framework to enable localization of region of interest (RoI) via segmentation, emphasize regression of values within RoI and still retain the overall accuracy via global regression. The network is optimized by a composite loss function that combines a dedicated loss from each task.Main results: We have included 54 brain patient images in this study and tested the sCT images against reference CT on a subset of 20 cases. A pilot dose evaluation was performed on 9 of the 20 test cases to demonstrate the viability of the generated sCT in RT planning. The average quantitative metrics produced by the proposed method over the test set were-(a) mean absolute error (MAE) of 70 ± 8.6 HU; (b) peak signal-to-noise ratio (PSNR) of 29.4 ± 2.8 dB; structural similarity metric (SSIM) of 0.95 ± 0.02; and (d) Dice coefficient of the body region of 0.984 ± 0.Significance: We demonstrate that the proposed method generates sCT images that resemble visual characteristics of a real CT image and has a quantitative accuracy that suits RT dose planning application. We compare the dose calculation from the proposed sCT and the real CT in a radiation therapy treatment planning setup and show that sCT based planning falls within 0.5% target dose error. The method presented here with an initial dose evaluation makes an encouraging precursor to a broader clinical evaluation of sCT based RT planning on different anatomical regions.
|
|
| 5. |
- Kofler, Florian, et al.
(author)
-
Approaching Peak Ground Truth
- 2023
-
In: Proceedings - International Symposium on Biomedical Imaging. - 1945-7928 .- 1945-8452. - 9781665473583
-
Conference paper (peer-reviewed)abstract
- Machine learning models are typically evaluated by computing similarity with reference annotations and trained by maximizing similarity with such. Especially in the biomedical domain, annotations are subjective and suffer from low inter-and intra-rater reliability. Since annotations only reflect one interpretation of the real world, this can lead to sub-optimal predictions even though the model achieves high similarity scores. Here, the theoretical concept of Peak Ground Truth (PGT) is introduced. PGT marks the point beyond which an increase in similarity with the reference annotation stops translating to better Real World Model Performance (RWMP). Additionally, a quantitative technique to approximate PGT by computing inter- and intra-rater reliability is proposed. Finally, four categories of PGT-aware strategies to evaluate and improve model performance are reviewed.
|
|
| 6. |
- Maier, O., et al.
(author)
-
ISLES 2015-A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI
- 2017
-
In: Medical Image Analysis. - : Elsevier BV. - 1361-8415 .- 1361-8423. ; 35, s. 250-269
-
Journal article (peer-reviewed)abstract
- Ischemic stroke is the most common cerebrovascular disease, and its diagnosis, treatment, and study relies on non-invasive imaging. Algorithms for stroke lesion segmentation from magnetic resonance imaging (MRI) volumes are intensely researched, but the reported results are largely incomparable due to different datasets and evaluation schemes. We approached this urgent problem of comparability with the Ischemic Stroke Lesion Segmentation (ISLES) challenge organized in conjunction with the MICCAI 2015 conference. In this paper we propose a common evaluation framework, describe the publicly available datasets, and present the results of the two sub-challenges: Sub-Acute Stroke Lesion Segmentation (SISS) and Stroke Perfusion Estimation (SPES). A total of 16 research groups participated with a wide range of state-of-the-art automatic segmentation algorithms. A thorough analysis of the obtained data enables a critical evaluation of the current state-of-the-art, recommendations for further developments, and the identification of remaining challenges. The segmentation of acute perfusion lesions addressed in SPES was found to be feasible. However, algorithms applied to sub-acute lesion segmentation in SISS still lack accuracy. Overall, no algorithmic characteristic of any method was found to perform superior to the others. Instead, the characteristics of stroke lesion appearances, their evolution, and the observed challenges should be studied in detail. The annotated ISLES image datasets continue to be publicly available through an online evaluation system to serve as an ongoing benchmarking resource (www.isles-challenge.org).
|
|
| 7. |
- Mehta, Raghav, et al.
(author)
-
QU-BraTS : MICCAI BraTS 2020 Challenge on QuantifyingUncertainty in Brain Tumor Segmentation - Analysis of Ranking Scores and Benchmarking Results
- 2022
-
In: Journal of Machine Learning for Biomedical Imaging. - 2766-905X. ; , s. 1-54
-
Journal article (peer-reviewed)abstract
- Deep learning (DL) models have provided the state-of-the-art performance in a wide variety of medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g., tumor and lesion sub-regions) is particularly challenging, and potential errors hinder the translation of DL models into clinical workflows. Quantifying the reliability of DL model predictions in the form of uncertainties, could enable clinical review of the most uncertain regions, thereby building trust and paving the way towards clinical translation. Recently, a number of uncertainty estimation methods have been introduced for DL medical image segmentation tasks. Developing scores to evaluate and compare the performance of uncertainty measures will assist the end-user in making more informed decisions. In this study, we explore and evaluate a score developed during the BraTS 2019-2020 task on uncertainty quantification (QU-BraTS), and designed to assess and rank uncertainty estimates for brain tumor multi-compartment segmentation. This score (1) rewards uncertainty estimates that produce high confidence in correct assertions, and those that assign low confidence levels at incorrect assertions, and (2) penalizes uncertainty measures that lead to a higher percentages of under-confident correct assertions. We further benchmark the segmentation uncertainties generated by 14 independent participating teams of QU-BraTS 2020, all of which also participated in the main BraTS segmentation task. Overall, our findings confirm the importance and complementary value that uncertainty estimates provide to segmentation algorithms, and hence highlight the need for uncertainty quantification in medical image analyses. Our evaluation code is made publicly available at https://github.com/RagMeh11/QU-BraTS
|
|
| 8. |
- Wieslander, Håkan
(author)
-
Application, Optimisation and Evaluation of Deep Learning for Biomedical Imaging
- 2022
-
Doctoral thesis (other academic/artistic)abstract
- Microscopy imaging is a powerful technique when studying biology at a cellular and sub-cellular level. When combined with digital image analysis it creates an invaluable tool for investigating complex biological processes and phenomena. However, imaging at the cell and sub-cellular level tends to generate large amounts of data which can be difficult to analyse, navigate and store. Despite these difficulties, large data volumes mean more information content which is beneficial for computational methods like machine learning, especially deep learning. The union of microscopy imaging and deep learning thus provides numerous opportunities for advancing our scientific understanding and uncovering interesting and useful biological insights.The work in this thesis explores various means for optimising information extraction from microscopy data utilising image analysis with deep learning. The focus is on three different imaging modalities: bright-field; fluorescence; and transmission electron microscopy. Within these modalities different learning-based image analysis and processing techniques are explored, ranging from image classification and detection to image restoration and translation. The main contributions are: (i) a computational method for diagnosing oral and cervical cancer based on smear samples and bright-field microscopy; (ii) a hierarchical analysis of whole-slide tissue images from fluorescence microscopy and introducing a confidence based measure for pixel classifications; (iii) an image restoration model for motion-degraded images from transmission electron microscopy with an evaluation of model overfitting on underlying textures; and (iv) an image-to-image translation (virtual staining) of cell images from bright-field to fluorescence microscopy, optimised for biological feature relevance. A common theme underlying all the investigations in this thesis is that the evaluation of the methods used is in relation to the biological question at hand.
|
|
