1
|
Rashidi G, Bounias D, Bujotzek M, Mora AM, Neher P, Maier-Hein KH. The potential of federated learning for self-configuring medical object detection in heterogeneous data distributions. Sci Rep 2024; 14:23844. [PMID: 39394440 PMCID: PMC11470020 DOI: 10.1038/s41598-024-74577-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/22/2024] [Accepted: 09/26/2024] [Indexed: 10/13/2024] Open
Abstract
Medical Object Detection (MOD) is a clinically relevant image processing method that locates structures of interest in radiological image data at object-level using bounding boxes. High-performing MOD models necessitate large datasets accurately reflecting the feature distribution of the corresponding problem domain. However, strict privacy regulations protecting patient data often hinder data consolidation, negatively affecting the performance and generalization of MOD models. Federated Learning (FL) offers a solution by enabling model training while the data remain at its original source institution. While existing FL solutions for medical image classification and segmentation demonstrate promising performance, FL for MOD remains largely unexplored. Motivated by this lack of technical solutions, we present an open-source, self-configuring and task-agnostic federated MOD framework. It integrates the FL framework Flower with nnDetection, a state-of-the-art MOD framework and provides several FL aggregation strategies. Furthermore, we evaluate model performance by creating simulated Independent Identically Distributed (IID) and non-IID scenarios, utilizing the publicly available datasets. Additionally, a detailed analysis of the distributions and characteristics of these datasets offers insights into how they can impact performance. Our framework's implementation demonstrates the feasibility of federated self-configuring MOD in non-IID scenarios and facilitates the development of MOD models trained on large distributed databases.
Collapse
Affiliation(s)
- Gabriel Rashidi
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany
- Faculty of Mathematics and Computer Science, Heidelberg University, Heidelberg, 69120, Germany
| | - Dimitrios Bounias
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany.
- Medical Faculty Heidelberg, Heidelberg University, Heidelberg, 69120, Germany.
| | - Markus Bujotzek
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany
- Medical Faculty Heidelberg, Heidelberg University, Heidelberg, 69120, Germany
| | - Andrés Martínez Mora
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany
- Medical Faculty Heidelberg, Heidelberg University, Heidelberg, 69120, Germany
| | - Peter Neher
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany
- German Cancer Consortium (DKTK), Partner Site Heidelberg, Im Neuenheimer Feld 280, Heidelberg, 69120, Germany
- Pattern Analysis and Learning Group, Heidelberg University Hospital, Heidelberg, 69120, Germany
- National Center for Tumor Diseases (NCT), Heidelberg University Hospital (UKHD) and German Cancer Research Center (DKFZ), Im Neuenheimer Feld 460, Heidelberg, 69120, Germany
| | - Klaus H Maier-Hein
- German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Heidelberg, 69120, Germany
- Faculty of Mathematics and Computer Science, Heidelberg University, Heidelberg, 69120, Germany
- Medical Faculty Heidelberg, Heidelberg University, Heidelberg, 69120, Germany
- German Cancer Consortium (DKTK), Partner Site Heidelberg, Im Neuenheimer Feld 280, Heidelberg, 69120, Germany
- Pattern Analysis and Learning Group, Heidelberg University Hospital, Heidelberg, 69120, Germany
- National Center for Tumor Diseases (NCT), Heidelberg University Hospital (UKHD) and German Cancer Research Center (DKFZ), Im Neuenheimer Feld 460, Heidelberg, 69120, Germany
| |
Collapse
|
2
|
Mahmutoglu MA, Rastogi A, Schell M, Foltyn-Dumitru M, Baumgartner M, Maier-Hein KH, Deike-Hofmann K, Radbruch A, Bendszus M, Brugnara G, Vollmuth P. Deep learning-based defacing tool for CT angiography: CTA-DEFACE. Eur Radiol Exp 2024; 8:111. [PMID: 39382818 PMCID: PMC11465008 DOI: 10.1186/s41747-024-00510-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/23/2024] [Accepted: 09/05/2024] [Indexed: 10/10/2024] Open
Abstract
The growing use of artificial neural network (ANN) tools for computed tomography angiography (CTA) data analysis underscores the necessity for elevated data protection measures. We aimed to establish an automated defacing pipeline for CTA data. In this retrospective study, CTA data from multi-institutional cohorts were utilized to annotate facemasks (n = 100) and train an ANN model, subsequently tested on an external institution's dataset (n = 50) and compared to a publicly available defacing algorithm. Face detection (MTCNN) and verification (FaceNet) networks were applied to measure the similarity between the original and defaced CTA images. Dice similarity coefficient (DSC), face detection probability, and face similarity measures were calculated to evaluate model performance. The CTA-DEFACE model effectively segmented soft face tissue in CTA data achieving a DSC of 0.94 ± 0.02 (mean ± standard deviation) on the test set. Our model was benchmarked against a publicly available defacing algorithm. After applying face detection and verification networks, our model showed substantially reduced face detection probability (p < 0.001) and similarity to the original CTA image (p < 0.001). The CTA-DEFACE model enabled robust and precise defacing of CTA data. The trained network is publicly accessible at www.github.com/neuroAI-HD/CTA-DEFACE . RELEVANCE STATEMENT: The ANN model CTA-DEFACE, developed for automatic defacing of CT angiography images, achieves significantly lower face detection probabilities and greater dissimilarity from the original images compared to a publicly available model. The algorithm has been externally validated and is publicly accessible. KEY POINTS: The developed ANN model (CTA-DEFACE) automatically generates facemasks for CT angiography images. CTA-DEFACE offers superior deidentification capabilities compared to a publicly available model. By means of graphics processing unit optimization, our model ensures rapid processing of medical images. Our model underwent external validation, underscoring its reliability for real-world application.
Collapse
Affiliation(s)
- Mustafa Ahmed Mahmutoglu
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany.
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany.
| | - Aditya Rastogi
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany
| | - Marianne Schell
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany
| | - Martha Foltyn-Dumitru
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany
| | - Michael Baumgartner
- Division for Medical Image Computing, German Cancer Research Center, Heidelberg, Germany
- Helmholtz Imaging, Heidelberg, Germany
- Faculty of Mathematics and Computer Science, Heidelberg University, Heidelberg, Germany
| | | | - Katerina Deike-Hofmann
- Department of Neuroradiology, Bonn University Hospital, Bonn, Germany
- Clinical Neuroimaging Group, German Center for Neurodegenerative Diseases, DZNE, Bonn, Germany
| | - Alexander Radbruch
- Department of Neuroradiology, Bonn University Hospital, Bonn, Germany
- Clinical Neuroimaging Group, German Center for Neurodegenerative Diseases, DZNE, Bonn, Germany
| | - Martin Bendszus
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
| | - Gianluca Brugnara
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany
| | - Philipp Vollmuth
- Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Germany
- Division for Computational Neuroimaging, Heidelberg University Hospital, Heidelberg, Germany
| |
Collapse
|
3
|
Qian Y, Li N, Li Y, Tao C, Liu Z, Zhang G, Yang F, Zhang H, Gao Y. Association between uric acid and the risk of hemorrhagic transformation in patients with acute ischemic stroke: a systematic review and meta-analysis. Front Neurol 2024; 15:1378912. [PMID: 39119562 PMCID: PMC11306017 DOI: 10.3389/fneur.2024.1378912] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/30/2024] [Accepted: 07/16/2024] [Indexed: 08/10/2024] Open
Abstract
Background The relationship between hemorrhagic transformation (HT) and uric acid (UA) remains controversial. This study aimed to investigate the relationship between UA concentrations and the risk of HT following acute ischemic stroke (AIS). Methods Electronic databases were searched for studies on HT and UA from inception to October 31, 2023. Two researchers independently reviewed the studies for inclusion. STATA Software 16.0 was used to compute the standardized mean difference (SMD) and 95% confidence interval (CI) for the pooled and post-outlier outcomes. Heterogeneity was evaluated using the I2 statistic and the Galbraith plot. Additionally, sensitivity analysis was performed. Lastly, Begg's funnel plot and Egger's test were used to assess publication bias. Results A total of 11 studies involving 4,608 patients were included in the meta-analysis. The pooled SMD forest plot (SMD = -0.313, 95% CI = -0.586--0.039, p = 0.025) displayed that low UA concentrations were linked to a higher risk of HT in post-AIS patients. However, heterogeneity (I2 = 89.8%, p < 0.001) was high among the studies. Six papers fell outside the Galbraith plot regression line, and there exclusive resulted in the absence of heterogeneity (I2 = 52.1%, p = 0.080). Meanwhile, repeated SMD analysis (SMD = -0.517, 95% CI = -0.748--0.285, p = 0.000) demonstrated that the HT group had lower UA concentrations. Finally, Begg's funnel plot and Egger's test indicated the absence of publication bias in our meta-analysis. Conclusion This meta-analysis illustrated a substantial connection between UA concentrations and HT, with lower UA concentrations independently linked with a higher risk of HT post-AIS. These results lay a theoretical reference for future studies.Systematic review registration:https://www.crd.york.ac.uk/PROSPERO/CRD42023485539.
Collapse
Affiliation(s)
- Ying Qian
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
- Institute for Brain Disorders, Beijing University of Chinese Medicine, Beijing, China
| | - Na Li
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
| | - Yuanyuan Li
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
| | - Chenxi Tao
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
| | - Zhenhong Liu
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
- Institute for Brain Disorders, Beijing University of Chinese Medicine, Beijing, China
| | - Guoxia Zhang
- Guang'anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China
| | - Fan Yang
- Guang'anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China
| | - Hongrui Zhang
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
- Institute for Brain Disorders, Beijing University of Chinese Medicine, Beijing, China
| | - Yonghong Gao
- Key Laboratory of Chinese Internal Medicine of Ministry of Education and Beijing, Dongzhimen Hospital, Beijing University of Chinese Medicine, Beijing, China
- Institute for Brain Disorders, Beijing University of Chinese Medicine, Beijing, China
| |
Collapse
|
4
|
Kuo DP, Chen YC, Li YT, Cheng SJ, Hsieh KLC, Kuo PC, Ou CY, Chen CY. Estimating the volume of penumbra in rodents using DTI and stack-based ensemble machine learning framework. Eur Radiol Exp 2024; 8:59. [PMID: 38744784 PMCID: PMC11093947 DOI: 10.1186/s41747-024-00455-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/20/2023] [Accepted: 03/05/2024] [Indexed: 05/16/2024] Open
Abstract
BACKGROUND This study investigates the potential of diffusion tensor imaging (DTI) in identifying penumbral volume (PV) compared to the standard gadolinium-required perfusion-diffusion mismatch (PDM), utilizing a stack-based ensemble machine learning (ML) approach with enhanced explainability. METHODS Sixteen male rats were subjected to middle cerebral artery occlusion. The penumbra was identified using PDM at 30 and 90 min after occlusion. We used 11 DTI-derived metrics and 14 distance-based features to train five voxel-wise ML models. The model predictions were integrated using stack-based ensemble techniques. ML-estimated and PDM-defined PVs were compared to evaluate model performance through volume similarity assessment, the Pearson correlation analysis, and Bland-Altman analysis. Feature importance was determined for explainability. RESULTS In the test rats, the ML-estimated median PV was 106.4 mL (interquartile range 44.6-157.3 mL), whereas the PDM-defined median PV was 102.0 mL (52.1-144.9 mL). These PVs had a volume similarity of 0.88 (0.79-0.96), a Pearson correlation coefficient of 0.93 (p < 0.001), and a Bland-Altman bias of 2.5 mL (2.4% of the mean PDM-defined PV), with 95% limits of agreement ranging from -44.9 to 49.9 mL. Among the features used for PV prediction, the mean diffusivity was the most important feature. CONCLUSIONS Our study confirmed that PV can be estimated using DTI metrics with a stack-based ensemble ML approach, yielding results comparable to the volume defined by the standard PDM. The model explainability enhanced its clinical relevance. Human studies are warranted to validate our findings. RELEVANCE STATEMENT The proposed DTI-based ML model can estimate PV without the need for contrast agent administration, offering a valuable option for patients with kidney dysfunction. It also can serve as an alternative if perfusion map interpretation fails in the clinical setting. KEY POINTS • Penumbral volume can be estimated by DTI combined with stack-based ensemble ML. • Mean diffusivity was the most important feature used for predicting penumbral volume. • The proposed approach can be beneficial for patients with kidney dysfunction.
Collapse
Affiliation(s)
- Duen-Pang Kuo
- Department of Medical Imaging, Taipei Medical University Hospital, No.250, Wu Hsing Street, Taipei, Taiwan
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
- Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
| | - Yung-Chieh Chen
- Department of Medical Imaging, Taipei Medical University Hospital, No.250, Wu Hsing Street, Taipei, Taiwan.
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan.
- Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.
| | - Yi-Tien Li
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
- Research Center for Neuroscience, Taipei Medical University, Taipei, Taiwan
- Ph.D. Program in Medical Neuroscience, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan
| | - Sho-Jen Cheng
- Department of Medical Imaging, Taipei Medical University Hospital, No.250, Wu Hsing Street, Taipei, Taiwan
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
| | - Kevin Li-Chun Hsieh
- Department of Medical Imaging, Taipei Medical University Hospital, No.250, Wu Hsing Street, Taipei, Taiwan
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
- Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
| | - Po-Chih Kuo
- Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan
| | - Chen-Yin Ou
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
| | - Cheng-Yu Chen
- Department of Medical Imaging, Taipei Medical University Hospital, No.250, Wu Hsing Street, Taipei, Taiwan
- Translational Imaging Research Center, Taipei Medical University Hospital, Taipei, Taiwan
- Research Center for Artificial Intelligence in Medicine, Taipei Medical University, Taipei, Taiwan
- Department of Radiology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
- Department of Radiology, National Defense Medical Center, Taipei, Taiwan
| |
Collapse
|
5
|
Warman R, Warman PI, Warman A, Bueso T, Ota R, Windisch T, Neves G. A deep learning method to identify and localize large-vessel occlusions from cerebral digital subtraction angiography. J Neuroimaging 2024; 34:366-375. [PMID: 38506407 DOI: 10.1111/jon.13193] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/09/2023] [Revised: 01/25/2024] [Accepted: 01/27/2024] [Indexed: 03/21/2024] Open
Abstract
BACKGROUND AND PURPOSE An essential step during endovascular thrombectomy is identifying the occluded arterial vessel on a cerebral digital subtraction angiogram (DSA). We developed an algorithm that can detect and localize the position of occlusions in cerebral DSA. METHODS We retrospectively collected cerebral DSAs from a single institution between 2018 and 2020 from 188 patients, 86 of whom suffered occlusions of the M1 and proximal M2 segments. We trained an ensemble of deep-learning models on fewer than 60 large-vessel occlusion (LVO)-positive patients. We evaluated the model on an independent test set and evaluated the truth of its predicted localizations using Intersection over Union and expert review. RESULTS On an independent test set of 166 cerebral DSA frames with an LVO prevalence of 0.19, the model achieved a specificity of 0.95 (95% confidence interval [CI]: 0.90, 0.99), a precision of 0.7450 (95% CI: 0.64, 0.88), and a sensitivity of 0.76 (95% CI: 0.66, 0.91). The model correctly localized the LVO in at least one frame in 13 of the 14 LVO-positive patients in the test set. The model achieved a precision of 0.67 (95% CI: 0.52, 0.79), recall of 0.69 (95% CI: 0.46, 0.81), and a mean average precision of 0.75 (95% CI: 0.56, 0.91). CONCLUSION This work demonstrates that a deep learning strategy using a limited dataset can generate effective representations used to identify LVOs. Generating an expanded and more complete dataset of LVOs with obstructed LVOs is likely the best way to improve the model's ability to localize LVOs.
Collapse
Affiliation(s)
- Roshan Warman
- Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
| | - Pranav I Warman
- Duke University School of Medicine, Durham, North Carolina, USA
| | - Anmol Warman
- Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
| | - Tulio Bueso
- Department of Neurology, Texas Tech University Medical Sciences Center, Lubbock, Texas, USA
| | - Riichi Ota
- Department of Neurology, Texas Tech University Medical Sciences Center, Lubbock, Texas, USA
| | - Thomas Windisch
- Department of Neurology, Texas Tech University Medical Sciences Center, Lubbock, Texas, USA
- Covenant Health, Lubbock, Texas, USA
| | - Gabriel Neves
- Department of Neurology, Section of Neurocritical Care, Washington University School of Medicine in St. Louis, St. Louis, Missouri, USA
| |
Collapse
|
6
|
Voigtlaender S, Pawelczyk J, Geiger M, Vaios EJ, Karschnia P, Cudkowicz M, Dietrich J, Haraldsen IRJH, Feigin V, Owolabi M, White TL, Świeboda P, Farahany N, Natarajan V, Winter SF. Artificial intelligence in neurology: opportunities, challenges, and policy implications. J Neurol 2024; 271:2258-2273. [PMID: 38367046 DOI: 10.1007/s00415-024-12220-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/20/2023] [Revised: 01/20/2024] [Accepted: 01/22/2024] [Indexed: 02/19/2024]
Abstract
Neurological conditions are the leading cause of disability and mortality combined, demanding innovative, scalable, and sustainable solutions. Brain health has become a global priority with adoption of the World Health Organization's Intersectoral Global Action Plan in 2022. Simultaneously, rapid advancements in artificial intelligence (AI) are revolutionizing neurological research and practice. This scoping review of 66 original articles explores the value of AI in neurology and brain health, systematizing the landscape for emergent clinical opportunities and future trends across the care trajectory: prevention, risk stratification, early detection, diagnosis, management, and rehabilitation. AI's potential to advance personalized precision neurology and global brain health directives hinges on resolving core challenges across four pillars-models, data, feasibility/equity, and regulation/innovation-through concerted pursuit of targeted recommendations. Paramount actions include swift, ethical, equity-focused integration of novel technologies into clinical workflows, mitigating data-related issues, counteracting digital inequity gaps, and establishing robust governance frameworks balancing safety and innovation.
Collapse
Affiliation(s)
- Sebastian Voigtlaender
- Systems Neuroscience Division, Max-Planck-Institute for Biological Cybernetics, Tübingen, Germany
- Virtual Diagnostics Team, QuantCo Inc., Cambridge, MA, USA
| | - Johannes Pawelczyk
- Faculty of Medicine, Ruprecht-Karls-University, Heidelberg, Germany
- Graduate Center of Medicine and Health, Technical University Munich, Munich, Germany
| | - Mario Geiger
- Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA
- NVIDIA, Zurich, Switzerland
| | - Eugene J Vaios
- Department of Radiation Oncology, Duke University Medical Center, Durham, NC, USA
| | - Philipp Karschnia
- Department of Neurosurgery, Ludwig-Maximilians-University and University Hospital Munich, Munich, Germany
- Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
| | - Merit Cudkowicz
- Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
| | - Jorg Dietrich
- Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA
| | - Ira R J Hebold Haraldsen
- Department of Neurology, Division of Clinical Neuroscience, Oslo University Hospital, Oslo, Norway
| | - Valery Feigin
- National Institute for Stroke and Applied Neurosciences, Auckland University of Technology, Auckland, New Zealand
| | - Mayowa Owolabi
- Center for Genomics and Precision Medicine, College of Medicine, University of Ibadan, Ibadan, Nigeria
- Neurology Unit, Department of Medicine, University of Ibadan, Ibadan, Nigeria
- Blossom Specialist Medical Center, Ibadan, Nigeria
- Lebanese American University of Beirut, Beirut, Lebanon
| | - Tara L White
- Department of Behavioral and Social Sciences, Brown University, Providence, RI, USA
| | | | | | | | - Sebastian F Winter
- Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA.
| |
Collapse
|
7
|
Hill DLG. AI in imaging: the regulatory landscape. Br J Radiol 2024; 97:483-491. [PMID: 38366148 PMCID: PMC11027239 DOI: 10.1093/bjr/tqae002] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/11/2023] [Revised: 12/03/2023] [Accepted: 12/26/2023] [Indexed: 02/18/2024] Open
Abstract
Artificial intelligence (AI) methods have been applied to medical imaging for several decades, but in the last few years, the number of publications and the number of AI-enabled medical devices coming on the market have significantly increased. While some AI-enabled approaches are proving very valuable, systematic reviews of the AI imaging field identify significant weaknesses in a significant proportion of the literature. Medical device regulators have recently become more proactive in publishing guidance documents and recognizing standards that will require that the development and validation of AI-enabled medical devices need to be more rigorous than required for tradition "rule-based" software. In particular, developers are required to better identify and mitigate risks (such as bias) that arise in AI-enabled devices, and to ensure that the devices are validated in a realistic clinical setting to ensure their output is clinically meaningful. While this evolving regulatory landscape will mean that device developers will take longer to bring novel AI-based medical imaging devices to market, such additional rigour is necessary to address existing weaknesses in the field and ensure that patients and healthcare professionals can trust AI-enabled devices. There would also be benefits in the academic community taking into account this regulatory framework, to improve the quality of the literature and make it easier for academically developed AI tools to make the transition to medical devices that impact healthcare.
Collapse
|