| 1. |
- Ranisch, Robert, et al.
(författare)
-
Ethics of digital contact tracing apps for the Covid-19 pandemic response
- 2020
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- There is a growing interest in contact tracing apps (CT apps) for pandemic man- agement. These apps raise significant moral concerns. It is therefore crucial to consider ethical requirements before and while implementing such apps. Public trust is of major importance for population uptake of contact tracing apps. Hasty, ill-prepared or badly communicated implementations of CT apps will likely under- mine public trust, and as such, risk impeding general effectiveness. In response to these demands, to meet ethical requirements and find a basis for justified trust, this background introduces an ethical framework for a responsible design and implementation of CT apps. However, even prudently chosen measures of digital contact tracing carry moral costs, which makes it necessary address different trade-offs. This background paper aims to inform developers, researchers and decision-makers be- fore and throughout the process of implementing contact tracing apps.
|
|
| 2. |
- Petersson, Jesper, 1974, et al.
(författare)
-
Off the record: The invisibility work of doctors in a patient-accessible electronic health record information service.
- 2021
-
Ingår i: Sociology of health & illness. - : Wiley. - 1467-9566 .- 0141-9889. ; 43:5, s. 1270-1285
-
Tidskriftsartikel (refereegranskat)abstract
- In this article, we draw on Michael Lipsky's work on street-level bureaucrats and discretion to analyse a real case setting comprising an interview study of 30 Swedish doctors regarding their experiences of changes in clinical work following patients being given access to medical records information online. We introduce the notion of invisibility work to capture how doctors exercise discretion to preserve the invisibility of their work, in contrast to the well-established notion of invisible work, which denotes work made invisible by parties other than those performing it. We discuss three main forms of invisibility work in relation to records: omitting information, cryptic writing and parallel note writing. We argue that invisibility work is a way for doctors to resolve professional tensions arising from the political decision to provide patients with online access to record information. Although invisibility work is understood by doctors as a solution to government-initiated visibility, we highlight how it can create difficulties for doctors concerning accountability towards patients, peers and authorities.
|
|
| 3. |
- Johansson, Martin L, et al.
(författare)
-
Non-invasive sampling procedure revealing the molecular events at different abutments of bone-anchored hearing systems–A prospective clinical pilot study
- 2022
-
Ingår i: Frontiers in Neuroscience. - : Frontiers Media SA. - 1662-4548 .- 1662-453X. ; 16
-
Tidskriftsartikel (refereegranskat)abstract
- Purpose: To investigate the molecular activities in different compartments around the bone-anchored hearing system (BAHS) with either electropolished or machined abutments and to correlate these activities with clinical and microbiological findings. Materials and methods: Twelve patients received machined or electropolished abutments after implant installation of BAHS. Peri-abutment fluid and tissue were collected from baseline to 12 months. Gene expression of cytokines and factors related to tissue healing and inflammation, regeneration and remodelling, as well as bacterial recognition were determined using quantitative-polymerase chain reaction (qPCR). The clinical status was evaluated using the Holgers scoring system, and bacterial colonisation was investigated by culturing. Results: The gene expression of inflammatory cytokines (IL-8, IL-1β, and IL-10) and bacteria-related Toll-like receptors (2 and 4) was higher in the peri-abutment fluid than at baseline and in the peri-abutment tissue at 3 and 12 months. Conversely, the expression of genes related to tissue regeneration (Coll1a1 and FOXO1) was higher in the tissue samples than in the peri-abutment fluid at 3 and 12 months. Electropolished abutments triggered higher expression of inflammatory cytokines (IL-8 and IL-1β) (in peri-abutment fluid) and regeneration factor FOXO1 (in peri-abutment tissue) than machined abutments. Several cytokine genes in the peri-abutment fluid correlated positively with the detection of aerobes, anaerobes and Staphylococcus species, as well as with high Holger scores. Conclusion: This study provides unprecedented molecular information on the biological processes of BAHS. Despite being apparently healed, the peri-abutment fluid harbours prolonged inflammatory activity in conjunction with the presence of different bacterial species. An electropolished abutment surface appears to be associated with stronger proinflammatory activity than that with a machined surface. The analysis of the peri-abutment fluid deserves further verification as a non-invasive sampling and diagnostic procedure of BAHS.
|
|
| 4. |
- Ge, Chenjie, 1991, et al.
(författare)
-
Enlarged Training Dataset by Pairwise GANs for Molecular-Based Brain Tumor Classification
- 2020
-
Ingår i: IEEE Access. - 2169-3536 .- 2169-3536. ; 8:1, s. 22560-22570
-
Tidskriftsartikel (refereegranskat)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.
|
|
| 5. |
- Robinson, Yohan, 1977, et al.
(författare)
-
AI och framtidens försvarsmedicin
- 2020
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- Medicinskt legitimerad personal är, och kommer med stor sannolikhet fortsattatt vara, en knapp resurs inom Försvarsmaktens sjukvårdsorganisation. I denna rapport ges en översikt över pågående och planerade ansatser baserade påartificiell intelligens (AI) inom akutsjukvård med särskild tonvikt på omhändertagandet av traumapatienter, där lösningarna skulle kunna bidra till att Försvarsmakten kan bibehålla sin sjukvårdskapacitet i kritiska lägen. Rapporten är ett resultat av samarbetet mellan FM, FOI, FMV, FHS och KI, och vänder sig i första hand till Försvarsmaktens strategiska ledning.Användningen av AI-teknik i framtida beslutsstöd kan skapa nya möjligheter till avlastning av personal och resurseffektivisering. Tekniken ger möjligheter att i realtid samla in, bearbeta och analysera stora mängder blandadinformation om förbands hälsoläge och fysiska stridsvärde. Bedömning av skadade kan t.ex. göras av triagedrönare och den efterföljande evakueringen kanunderlättas av intelligenta autonoma plattformar. Införandet av AI-system ställer dock vårdgivaren inför svåra etiska och medikolegala överväganden.Försvarsmedicin har en central roll i Försvarsmaktens krigföringsförmåga och för samhällets uthållighet. För att nyttja hela AI-teknikens framfart till Försvarsmaktens nytta måste dess innebörd och konsekvens för försvarsmedicinen förstås. Därför rekommenderar denna studie att Försvarsmaktens framtida satsningar inom AI och autonomi inkluderar den försvarsmedicinska teknikutveckling som är beskriven i denna rapport.
|
|
| 6. |
- Abbaspour, S., et al.
(författare)
-
Real-Time and Offline Evaluation of Myoelectric Pattern Recognition for the Decoding of Hand Movements
- 2021
-
Ingår i: Sensors. - : MDPI AG. - 1424-8220. ; 21:16
-
Tidskriftsartikel (refereegranskat)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.
|
|
| 7. |
- Hellstrand Tang, Ulla, et al.
(författare)
-
Exploring the Role of Complexity in Health Care Technology Bottom-Up Innovations : Multiple-Case Study Using the Nonadoption, Abandonment, Scale-Up, Spread, and Sustainability Complexity Assessment Tool
- 2024
-
Ingår i: JMIR Human Factors. - : JMIR Publications. - 2292-9495. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: New digital technology presents new challenges to health care on multiple levels. There are calls for further research that considers the complex factors related to digital innovations in complex health care settings to bridge the gap when moving from linear, logistic research to embracing and testing the concept of complexity. The nonadoption, abandonment, scale-up, spread, and sustainability (NASSS) framework was developed to help study complexity in digital innovations.OBJECTIVE: This study aims to investigate the role of complexity in the development and deployment of innovations by retrospectively assessing challenges to 4 digital health care innovations initiated from the bottom up.METHODS: A multicase retrospective, deductive, and explorative analysis using the NASSS complexity assessment tool LONG was conducted. In total, 4 bottom-up innovations developed in Region Västra Götaland in Sweden were explored and compared to identify unique and shared complexity-related challenges.RESULTS: The analysis resulted in joint insights and individual learning. Overall, the complexity was mostly found outside the actual innovation; more specifically, it related to the organization's readiness to integrate new innovations, how to manage and maintain innovations, and how to finance them. The NASSS framework sheds light on various perspectives that can either facilitate or hinder the adoption, scale-up, and spread of technological innovations. In the domain of condition or diagnosis, a well-informed understanding of the complexity related to the condition or illness (diabetes, cancer, bipolar disorders, and schizophrenia disorders) is of great importance for the innovation. The value proposition needs to be clearly described early to enable an understanding of costs and outcomes. The questions in the NASSS complexity assessment tool LONG were sometimes difficult to comprehend, not only from a language perspective but also due to a lack of understanding of the surrounding organization's system and its setting.CONCLUSIONS: Even when bottom-up innovations arise within the same support organization, the complexity can vary based on the developmental phase and the unique characteristics of each project. Identifying, defining, and understanding complexity may not solve the issues but substantially improves the prospects for successful deployment. Successful innovation within complex organizations necessitates an adaptive leadership and structures to surmount cultural resistance and organizational impediments. A rigid, linear, and stepwise approach risks disregarding interconnected variables and dependencies, leading to suboptimal outcomes. Success lies in embracing the complexity with its uncertainty, nurturing creativity, and adopting a nonlinear methodology that accommodates the iterative nature of innovation processes within complex organizations.
|
|
| 8. |
- Pfeiffer, Christoph, 1989, et al.
(författare)
-
On-scalp MEG sensor localization using magnetic dipole-like coils: A method for highly accurate co-registration
- 2020
-
Ingår i: Neuroimage. - : Elsevier BV. - 1053-8119 .- 1095-9572. ; 212
-
Tidskriftsartikel (refereegranskat)abstract
- Source modelling in magnetoencephalography (MEG) requires precise co-registration of the sensor array and the anatomical structure of the measured individual's head. In conventional MEG, the positions and orientations of the sensors relative to each other are fixed and known beforehand, requiring only localization of the head relative to the sensor array. Since the sensors in on-scalp MEG are positioned on the scalp, locations of the individual sensors depend on the subject's head shape and size. The positions and orientations of on-scalp sensors must therefore be measured a every recording. This can be achieved by inverting conventional head localization, localizing the sensors relative to the head - rather than the other way around. In this study we present a practical method for localizing sensors using magnetic dipole-like coils attached to the subject's head. We implement and evaluate the method in a set of on-scalp MEG recordings using a 7-channel on-scalp MEG system based on high critical temperature superconducting quantum interference devices (high-T-c SQUIDs). The method allows individually localizing the sensor positions, orientations, and responsivities with high accuracy using only a short averaging time (<= 2 mm, < 3 degrees and < 3%, respectively, with 1-s averaging), enabling continuous sensor localization. Calibrating and jointly localizing the sensor array can further improve the accuracy of position and orientation (< 1 mm and < 1 degrees, respectively, with 1-s coil recordings). We demonstrate source localization of on-scalp recorded somatosensory evoked activity based on coregistration with our method. Equivalent current dipole fits of the evoked responses corresponded well (within 4.2 mm) with those based on a commercial, whole-head MEG system.
|
|
| 9. |
|
|
| 10. |
- Insulander Björk, Klara, 1982, et al.
(författare)
-
Experimental determination of concentration factors of Mn, Zn and I in the phytoplankton species Phaeodactylum Tricornutum
- 2023
-
Ingår i: Journal of Environmental Radioactivity. - : Elsevier BV. - 0265-931X .- 1879-1700. ; 261
-
Tidskriftsartikel (refereegranskat)abstract
- Anthropogenic radionuclides released into the environment cause a radiation dose to wildlife and humans which must be quantified, both to assess the effect of normal releases, and to predict the consequences of a larger, unplanned release. To estimate the spread of the radioactive elements, the ecosystem around release points is modelled, and element uptake is usually quantified by concentration factors (CF), which relates the concentration of an element in an organism to the concentration of the same element in a medium under equilibrium conditions. In this work, we experimentally determine some phytoplankton CF that are needed for improved modelling of the marine ecosystems around nuclear facilities and release points. CFs that require better determination have been identified through literature search. Sensitivity studies, using the currently used ecosystem modelling software PREDO, show that for most studied groups, the dose committed by the respective radionuclides is almost proportional to the corresponding phytoplankton CFs. In the present work, CFs are determined through laboratory experiments with cultured phytoplankton and radionuclides of the concerned elements, assessing the element uptake by the phytoplankton through detection of the emitted radiation. The three CF assessed in this work were those for manganese, zinc and iodine in phytoplankton. Conservative estimates of these CF based on the present data are 40 000 L/kg for manganese, 50 000 L/kg for zinc and 180 L/kg for iodine with the phytoplankton masses referring to their dry weight.
|
|
| 11. |
- Hagberg, Eva, et al.
(författare)
-
Semi-supervised learning with natural language processing for right ventricle classification in echocardiography—a scalable approach
- 2022
-
Ingår i: Computers in Biology and Medicine. - : Elsevier BV. - 0010-4825 .- 1879-0534. ; 143
-
Tidskriftsartikel (refereegranskat)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
|
|
| 12. |
- Ortiz Catalan, Max Jair, 1982, et al.
(författare)
-
Chronic Use of a Sensitized Bionic Hand Does Not Remap the Sense of Touch
- 2020
-
Ingår i: Cell Reports. - : Elsevier BV. - 2211-1247. ; 33:12
-
Tidskriftsartikel (refereegranskat)abstract
- Electrical stimulation of tactile nerve fibers can be used to restore touch through a bionic hand. Ortiz-Catalan et al. show that a mismatch between the location of the sensor on the bionic hand and the tactile experience is not resolved after long-term prosthesis use.
|
|
| 13. |
- Fan, Xuelong, et al.
(författare)
-
Effects of sensor types and angular velocity computational methods in field measurements of occupational upper arm and trunk postures and movements
- 2021
-
Ingår i: Sensors. - : MDPI AG. - 1424-8220. ; 21:16
-
Tidskriftsartikel (refereegranskat)abstract
- Accelerometer-based inclinometers have dominated kinematic measurements in previous field studies, while the use of inertial measurement units that additionally include gyroscopes is rapidly increasing. Recent laboratory studies suggest that these two sensor types and the two commonly used angular velocity computational methods may produce substantially different results. The aim of this study was, therefore, to evaluate the effects of sensor types and angular velocity computational methods on the measures of work postures and movements in a real occupational setting. Half-workday recordings of arm and trunk postures, and movements from 38 warehouse workers were compared using two sensor types: accelerometers versus accelerometers with gyroscopes—and using two angular velocity computational methods, i.e., inclination velocity versus generalized velocity. The results showed an overall small difference (<2° and value independent) for posture percentiles between the two sensor types, but substantial differences in movement percentiles both between the sensor types and between the angular computational methods. For example, the group mean of the 50th percentiles were for accelerometers: 71°/s (generalized velocity) and 33°/s (inclination velocity)—and for accelerometers with gyroscopes: 31°/s (generalized velocity) and 16°/s (inclination velocity). The significant effects of sensor types and angular computational methods on angular velocity measures in field work are important in inter-study comparisons and in comparisons to recommended threshold limit values.
|
|
| 14. |
- Ali, Muhaddisa Barat, 1986, et al.
(författare)
-
A novel federated deep learning scheme for glioma and its subtype classification
- 2023
-
Ingår i: Frontiers in Neuroscience. - 1662-4548 .- 1662-453X. ; 17
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Deep learning (DL) has shown promising results in molecular-based classification of glioma subtypes from MR images. DL requires a large number of training data for achieving good generalization performance. Since brain tumor datasets are usually small in size, combination of such datasets from different hospitals are needed. Data privacy issue from hospitals often poses a constraint on such a practice. Federated learning (FL) has gained much attention lately as it trains a central DL model without requiring data sharing from different hospitals. Method: We propose a novel 3D FL scheme for glioma and its molecular subtype classification. In the scheme, a slice-based DL classifier, EtFedDyn, is exploited which is an extension of FedDyn, with the key differences on using focal loss cost function to tackle severe class imbalances in the datasets, and on multi-stream network to exploit MRIs in different modalities. By combining EtFedDyn with domain mapping as the pre-processing and 3D scan-based post-processing, the proposed scheme makes 3D brain scan-based classification on datasets from different dataset owners. To examine whether the FL scheme could replace the central learning (CL) one, we then compare the classification performance between the proposed FL and the corresponding CL schemes. Furthermore, detailed empirical-based analysis were also conducted to exam the effect of using domain mapping, 3D scan-based post-processing, different cost functions and different FL schemes. Results: Experiments were done on two case studies: classification of glioma subtypes (IDH mutation and wild-type on TCGA and US datasets in case A) and glioma grades (high/low grade glioma HGG and LGG on MICCAI dataset in case B). The proposed FL scheme has obtained good performance on the test sets (85.46%, 75.56%) for IDH subtypes and (89.28%, 90.72%) for glioma LGG/HGG all averaged on five runs. Comparing with the corresponding CL scheme, the drop in test accuracy from the proposed FL scheme is small (−1.17%, −0.83%), indicating its good potential to replace the CL scheme. Furthermore, the empirically tests have shown that an increased classification test accuracy by applying: domain mapping (0.4%, 1.85%) in case A; focal loss function (1.66%, 3.25%) in case A and (1.19%, 1.85%) in case B; 3D post-processing (2.11%, 2.23%) in case A and (1.81%, 2.39%) in case B and EtFedDyn over FedAvg classifier (1.05%, 1.55%) in case A and (1.23%, 1.81%) in case B with fast convergence, which all contributed to the improvement of overall performance in the proposed FL scheme. Conclusion: The proposed FL scheme is shown to be effective in predicting glioma and its subtypes by using MR images from test sets, with great potential of replacing the conventional CL approaches for training deep networks. This could help hospitals to maintain their data privacy, while using a federated trained classifier with nearly similar performance as that from a centrally trained one. Further detailed experiments have shown that different parts in the proposed 3D FL scheme, such as domain mapping (make datasets more uniform) and post-processing (scan-based classification), are essential.
|
|
| 15. |
- Andersen, L. M., et al.
(författare)
-
On-scalp MEG SQUIDs are sensitive to early somatosensory activity unseen by conventional MEG
- 2020
-
Ingår i: NeuroImage. - : Elsevier BV. - 1053-8119 .- 1095-9572. ; 221
-
Tidskriftsartikel (refereegranskat)abstract
- Magnetoencephalography (MEG) has a unique capacity to resolve the spatio-temporal development of brain activity from non-invasive measurements. Conventional MEG, however, relies on sensors that sample from a distance (20–40 mm) to the head due to thermal insulation requirements (the MEG sensors function at 4 K in a helmet). A gain in signal strength and spatial resolution may be achieved if sensors are moved closer to the head. Here, we report a study comparing measurements from a seven-channel on-scalp SQUID MEG system to those from a conventional (in-helmet) SQUID MEG system. We compared the spatio-temporal resolution between on-scalp and conventional MEG by comparing the discrimination accuracy for neural activity patterns resulting from stimulating five different phalanges of the right hand. Because of proximity and sensor density differences between on-scalp and conventional MEG, we hypothesized that on-scalp MEG would allow for a more high-resolved assessment of these activity patterns, and therefore also a better classification performance in discriminating between neural activations from the different phalanges. We observed that on-scalp MEG provided better classification performance during an early post-stimulus period (10–20 ms). This corresponded to the electroencephalographic (EEG) component P16/N16 and was an unexpected observation as this component is usually not observed in conventional MEG. This finding shows that on-scalp MEG enables a richer registration of the cortical signal, indicating a sensitivity to what are potentially sources in the thalamo-cortical radiation. We had originally expected that on-scalp MEG would provide better classification accuracy based on activity in proximity to the P60m component compared to conventional MEG. This component indeed allowed for the best classification performance for both MEG systems (60–75%, chance 50%). However, we did not find that on-scalp MEG allowed for better classification than conventional MEG at this latency. We suggest that this absence of differences is due to the limited sensor coverage in the recording, in combination with our strategy for positioning the on-scalp MEG sensors. We show how the current sensor coverage may have limited our chances to register the necessary between-phalange source field dissimilarities for fair hypothesis testing, an approach we otherwise believe to be useful for future benchmarking measurements. © 2020 The Authors
|
|
| 16. |
- Böhler, Christian, et al.
(författare)
-
Multilayer Arrays for Neurotechnology Applications (MANTA): Chronically Stable Thin-Film Intracortical Implants
- 2023
-
Ingår i: Advanced Science. - : John Wiley & Sons. - 2198-3844. ; 10:14
-
Tidskriftsartikel (refereegranskat)abstract
- Flexible implantable neurointerfaces show great promise in addressing one of the major challenges of implantable neurotechnology, namely the loss of signal connected to unfavorable probe tissue interaction. The authors here show how multilayer polyimide probes allow high-density intracortical recordings to be combined with a reliable long-term stable tissue interface, thereby progressing toward chronic stability of implantable neurotechnology. The probes could record 10–60 single units over 5 months with a consistent peak-to-peak voltage at dimensions that ensure robust handling and insulation longevity. Probes that remain in intimate contact with the signaling tissue over months to years are a game changer for neuroscience and, importantly, open up for broader clinical translation of systems relying on neurotechnology to interface the human brain.
|
|
| 17. |
- Koriakina, Nadezhda, 1991-, et al.
(författare)
-
Deep multiple instance learning versus conventional deep single instance learning for interpretable oral cancer detection
- 2024
-
Ingår i: PLOS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 19:4 April
-
Tidskriftsartikel (refereegranskat)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.
|
|
| 18. |
- Frishammar, Johan, et al.
(författare)
-
Digital health platforms for the elderly? Key adoption and usage barriers and ways to address them
- 2023
-
Ingår i: Technological forecasting & social change. - : Elsevier. - 0040-1625 .- 1873-5509. ; 189
-
Tidskriftsartikel (refereegranskat)abstract
- Digital healthcare platforms (DHPs) represent a relatively new phenomenon that could provide a valuable complement to physical primary care – for example, by reducing costs, improving access to healthcare, and allowing patient monitoring. However, such platforms are mainly used today by the younger generations, which creates a “digital divide” between the younger and the elderly. This article aims to identify: i) the perceived key barriers that inhibit adoption and usage of DHPs by the elderly, and ii) what DHP providers can do to facilitate increased adoption and usage by the elderly. The article draws on qualitative interviews with elderly and complementary process data from a major Swedish DHP. We find that the elderly perceives two key barriers to initial adoption of DHPs: i) negative attitudes and technology anxiety and ii) one key barrier affecting both adoption and usage – lack of trust. The analysis also identifies multiple development suggestions for DHP improvement to better accommodate the needs of the elderly, including suggestions for application development and tailored education activities. We provide an integrated framework outlining the key barriers perceived and ways to address them. In so doing, we contribute to the literature on mHealth and to the literature on platforms in healthcare.
|
|
| 19. |
- Ge, Chenjie, 1991, et al.
(författare)
-
Deep semi-supervised learning for brain tumor classification
- 2020
-
Ingår i: BMC Medical Imaging. - : Springer Science and Business Media LLC. - 1471-2342. ; 20:1
-
Tidskriftsartikel (refereegranskat)abstract
- Background: This paper addresses issues of brain tumor, glioma, classification from four modalities of Magnetic Resonance Image (MRI) scans (i.e., T1 weighted MRI, T1 weighted MRI with contrast-enhanced, T2 weighted MRI and FLAIR). Currently, many available glioma datasets often contain some unlabeled brain scans, and many datasets are moderate in size. Methods: We propose to exploit deep semi-supervised learning to make full use of the unlabeled data. Deep CNN features were incorporated into a new graph-based semi-supervised learning framework for learning the labels of the unlabeled data, where a new 3D-2D consistent constraint is added to make consistent classifications for the 2D slices from the same 3D brain scan. A deep-learning classifier is then trained to classify different glioma types using both labeled and unlabeled data with estimated labels. To alleviate the overfitting caused by moderate-size datasets, synthetic MRIs generated by Generative Adversarial Networks (GANs) are added in the training of CNNs. Results: The proposed scheme has been tested on two glioma datasets, TCGA dataset for IDH-mutation prediction (molecular-based glioma subtype classification) and MICCAI dataset for glioma grading. Our results have shown good performance (with test accuracies 86.53% on TCGA dataset and 90.70% on MICCAI dataset). Conclusions: The proposed scheme is effective for glioma IDH-mutation prediction and glioma grading, and its performance is comparable to the state-of-the-art.
|
|
| 20. |
- Jacobsson, Martin, 1976-, et al.
(författare)
-
Deep Learning-Based Early Prediction of Intraoperative Hypotension
- 2021
-
Konferensbidrag (refereegranskat)abstract
- This work focuses on predicting near-term onset of hypotension prior to onset using convolutional neural networks. Based solely on the arterial blood pressure curve, our initial attempt can predict an onset with 60% sensitivity and 80% specificity 5-15 minutes before onset.Clinical relevance Hypotension is common during large surgery. By identifying and treating hypotensive episodes early, preferably even before onset, hypotension and its associate post- surgery complications are reduced. Even a prediction with 80% sensitivity/specificity is valuable for the anesthesiologist.
|
|
| 21. |
- Simistira Liwicki, Foteini, et al.
(författare)
-
Bimodal electroencephalography-functional magnetic resonance imaging dataset for inner-speech recognition
- 2023
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 10
-
Tidskriftsartikel (refereegranskat)abstract
- The recognition of inner speech, which could give a ‘voice’ to patients that have no ability to speak or move, is a challenge for brain-computer interfaces (BCIs). A shortcoming of the available datasets is that they do not combine modalities to increase the performance of inner speech recognition. Multimodal datasets of brain data enable the fusion of neuroimaging modalities with complimentary properties, such as the high spatial resolution of functional magnetic resonance imaging (fMRI) and the temporal resolution of electroencephalography (EEG), and therefore are promising for decoding inner speech. This paper presents the first publicly available bimodal dataset containing EEG and fMRI data acquired nonsimultaneously during inner-speech production. Data were obtained from four healthy, right-handed participants during an inner-speech task with words in either a social or numerical category. Each of the 8-word stimuli were assessed with 40 trials, resulting in 320 trials in each modality for each participant. The aim of this work is to provide a publicly available bimodal dataset on inner speech, contributing towards speech prostheses.
|
|
| 22. |
- Larsson, Julia, et al.
(författare)
-
Optimizing study design in LPS challenge studies for quantifying drug induced inhibition of TNF? response: Did we miss the prime time?
- 2022
-
Ingår i: European Journal of Pharmaceutical Sciences. - : Elsevier BV. - 0928-0987 .- 1879-0720. ; 176
-
Tidskriftsartikel (refereegranskat)abstract
- In this work we evaluate the study design of LPS challenge experiments used for quantification of drug induced inhibition of TNF alpha response and provide general guidelines of how to improve the study design. Analysis of model simulated data, using a recently published TNF alpha turnover model, as well as the optimal design tool PopED have been used to find the optimal values of three key study design variables - time delay between drug and LPS administration, LPS dose, and sampling time points - that in turn could make the resulting TNF alpha response data more informative. Our findings suggest that the current rule of thumb for choosing the time delay should be reconsidered, and that the placement of the measurements after maximal TNF alpha response are crucial for the quality of the experiment. Furthermore, a literature study summarizing a wide range of published LPS challenge studies is provided, giving a broader perspective of how LPS challenge studies are usually conducted both in a preclinical and clinical setting.
|
|
| 23. |
- Lind, Carl, et al.
(författare)
-
Reducing postural load in order picking through a smart workwear system using real-time vibrotactile feedback
- 2020
-
Ingår i: Applied Ergonomics. - : Elsevier. - 0003-6870 .- 1872-9126. ; 89
-
Tidskriftsartikel (refereegranskat)abstract
- Vibrotactile feedback training may be one possible method for interventions that target at learning better work techniques and improving postures in manual handling. This study aimed to evaluate the short term effect of real-time vibrotactile feedback on postural exposure using a smart workwear system for work postures intervention in simulated industrial order picking. Fifteen workers at an industrial manufacturing plant performed order-picking tasks, in which the vibrotactile feedback was used for postural training at work. The system recorded the trunk and upper arm postures. Questionnaires and semi-structured interviews were conducted about the users’ experience of the system. The results showed reduced time in trunk inclination ≥20°, ≥30° and ≥45° and dominant upper arm elevation ≥30° and ≥45° when the workers received feedback, and for trunk inclination ≥20°, ≥30° and ≥45° and dominant upper arm elevation ≥30°, after feedback withdrawal. The workers perceived the system as useable, comfortable, and supportive for learning. The system has the potential of contributing to improved postures in order picking through an automated short-term training program. © 2020 Elsevier Ltd
|
|
| 24. |
- Pozzoli, Susanna, et al.
(författare)
-
Domain expertise–agnostic feature selection for the analysis of breast cancer data*
- 2020
-
Ingår i: Artificial Intelligence in Medicine. - : Elsevier B.V.. - 0933-3657 .- 1873-2860. ; 108
-
Tidskriftsartikel (refereegranskat)abstract
- Progress in proteomics has enabled biologists to accurately measure the amount of protein in a tumor. This work is based on a breast cancer data set, result of the proteomics analysis of a cohort of tumors carried out at Karolinska Institutet. While evidence suggests that an anomaly in the protein content is related to the cancerous nature of tumors, the proteins that could be markers of cancer types and subtypes and the underlying interactions are not completely known. This work sheds light on the potential of the application of unsupervised learning in the analysis of the aforementioned data sets, namely in the detection of distinctive proteins for the identification of the cancer subtypes, in the absence of domain expertise. In the analyzed data set, the number of samples, or tumors, is significantly lower than the number of features, or proteins; consequently, the input data can be thought of as high-dimensional data. The use of high-dimensional data has already become widespread, and a great deal of effort has been put into high-dimensional data analysis by means of feature selection, but it is still largely based on prior specialist knowledge, which in this case is not complete. There is a growing need for unsupervised feature selection, which raises the issue of how to generate promising subsets of features among all the possible combinations, as well as how to evaluate the quality of these subsets in the absence of specialist knowledge. We hereby propose a new wrapper method for the generation and evaluation of subsets of features via spectral clustering and modularity, respectively. We conduct experiments to test the effectiveness of the new method in the analysis of the breast cancer data, in a domain expertise–agnostic context. Furthermore, we show that we can successfully augment our method by incorporating an external source of data on known protein complexes. Our approach reveals a large number of subsets of features that are better at clustering the samples than the state-of-the-art classification in terms of modularity and shows a potential to be useful for future proteomics research.
|
|
| 25. |
- de Dios, Eddie, et al.
(författare)
-
Introduction to Deep Learning in Clinical Neuroscience
- 2022
-
Ingår i: Acta Neurochirurgica, Supplement. - Cham : Springer International Publishing. - 2197-8395 .- 0065-1419. ; 134, s. 79-89
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)abstract
- The use of deep learning (DL) is rapidly increasing in clinical neuroscience. The term denotes models with multiple sequential layers of learning algorithms, architecturally similar to neural networks of the brain. We provide examples of DL in analyzing MRI data and discuss potential applications and methodological caveats. Important aspects are data pre-processing, volumetric segmentation, and specific task-performing DL methods, such as CNNs and AEs. Additionally, GAN-expansion and domain mapping are useful DL techniques for generating artificial data and combining several smaller datasets. We present results of DL-based segmentation and accuracy in predicting glioma subtypes based on MRI features. Dice scores range from 0.77 to 0.89. In mixed glioma cohorts, IDH mutation can be predicted with a sensitivity of 0.98 and specificity of 0.97. Results in test cohorts have shown improvements of 5–7% in accuracy, following GAN-expansion of data and domain mapping of smaller datasets. The provided DL examples are promising, although not yet in clinical practice. DL has demonstrated usefulness in data augmentation and for overcoming data variability. DL methods should be further studied, developed, and validated for broader clinical use. Ultimately, DL models can serve as effective decision support systems, and are especially well-suited for time-consuming, detail-focused, and data-ample tasks.
|
|
