1
|
El Habib Daho M, Li Y, Zeghlache R, Boité HL, Deman P, Borderie L, Ren H, Mannivanan N, Lepicard C, Cochener B, Couturier A, Tadayoni R, Conze PH, Lamard M, Quellec G. DISCOVER: 2-D multiview summarization of Optical Coherence Tomography Angiography for automatic diabetic retinopathy diagnosis. Artif Intell Med 2024; 149:102803. [PMID: 38462293 DOI: 10.1016/j.artmed.2024.102803] [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: 08/24/2023] [Revised: 12/19/2023] [Accepted: 02/03/2024] [Indexed: 03/12/2024]
Abstract
Diabetic Retinopathy (DR), an ocular complication of diabetes, is a leading cause of blindness worldwide. Traditionally, DR is monitored using Color Fundus Photography (CFP), a widespread 2-D imaging modality. However, DR classifications based on CFP have poor predictive power, resulting in suboptimal DR management. Optical Coherence Tomography Angiography (OCTA) is a recent 3-D imaging modality offering enhanced structural and functional information (blood flow) with a wider field of view. This paper investigates automatic DR severity assessment using 3-D OCTA. A straightforward solution to this task is a 3-D neural network classifier. However, 3-D architectures have numerous parameters and typically require many training samples. A lighter solution consists in using 2-D neural network classifiers processing 2-D en-face (or frontal) projections and/or 2-D cross-sectional slices. Such an approach mimics the way ophthalmologists analyze OCTA acquisitions: (1) en-face flow maps are often used to detect avascular zones and neovascularization, and (2) cross-sectional slices are commonly analyzed to detect macular edemas, for instance. However, arbitrary data reduction or selection might result in information loss. Two complementary strategies are thus proposed to optimally summarize OCTA volumes with 2-D images: (1) a parametric en-face projection optimized through deep learning and (2) a cross-sectional slice selection process controlled through gradient-based attribution. The full summarization and DR classification pipeline is trained from end to end. The automatic 2-D summary can be displayed in a viewer or printed in a report to support the decision. We show that the proposed 2-D summarization and classification pipeline outperforms direct 3-D classification with the advantage of improved interpretability.
Collapse
Affiliation(s)
- Mostafa El Habib Daho
- Univ Bretagne Occidentale, Brest, F-29200, France; Inserm, UMR 1101, Brest, F-29200, France
| | - Yihao Li
- Univ Bretagne Occidentale, Brest, F-29200, France; Inserm, UMR 1101, Brest, F-29200, France
| | - Rachid Zeghlache
- Univ Bretagne Occidentale, Brest, F-29200, France; Inserm, UMR 1101, Brest, F-29200, France
| | - Hugo Le Boité
- Sorbonne University, Paris, F-75006, France; Service d'Ophtalmologie, Hôpital Lariboisière, APHP, Paris, F-75475, France
| | - Pierre Deman
- ADCIS, Saint-Contest, F-14280, France; Evolucare Technologies, Le Pecq, F-78230, France
| | | | - Hugang Ren
- Carl Zeiss Meditec, Dublin, CA 94568, USA
| | | | - Capucine Lepicard
- Service d'Ophtalmologie, Hôpital Lariboisière, APHP, Paris, F-75475, France
| | - Béatrice Cochener
- Univ Bretagne Occidentale, Brest, F-29200, France; Inserm, UMR 1101, Brest, F-29200, France; Service d'Ophtalmologie, CHRU Brest, Brest, F-29200, France
| | - Aude Couturier
- Service d'Ophtalmologie, Hôpital Lariboisière, APHP, Paris, F-75475, France
| | - Ramin Tadayoni
- Service d'Ophtalmologie, Hôpital Lariboisière, APHP, Paris, F-75475, France; Paris Cité University, Paris, F-75006, France
| | - Pierre-Henri Conze
- Inserm, UMR 1101, Brest, F-29200, France; IMT Atlantique, Brest, F-29200, France
| | - Mathieu Lamard
- Univ Bretagne Occidentale, Brest, F-29200, France; Inserm, UMR 1101, Brest, F-29200, France
| | | |
Collapse
|
2
|
Pigat L, Geisler BP, Sheikhalishahi S, Sander J, Kaspar M, Schmutz M, Rohr SO, Wild CM, Goss S, Zaghdoudi S, Hinske LC. Predicting Hypoxia Using Machine Learning: Systematic Review. JMIR Med Inform 2024; 12:e50642. [PMID: 38329094 PMCID: PMC10879670 DOI: 10.2196/50642] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/17/2023] [Revised: 11/02/2023] [Accepted: 11/05/2023] [Indexed: 02/09/2024] Open
Abstract
Background Hypoxia is an important risk factor and indicator for the declining health of inpatients. Predicting future hypoxic events using machine learning is a prospective area of study to facilitate time-critical interventions to counter patient health deterioration. Objective This systematic review aims to summarize and compare previous efforts to predict hypoxic events in the hospital setting using machine learning with respect to their methodology, predictive performance, and assessed population. Methods A systematic literature search was performed using Web of Science, Ovid with Embase and MEDLINE, and Google Scholar. Studies that investigated hypoxia or hypoxemia of hospitalized patients using machine learning models were considered. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. Results After screening, a total of 12 papers were eligible for analysis, from which 32 models were extracted. The included studies showed a variety of population, methodology, and outcome definition. Comparability was further limited due to unclear or high risk of bias for most studies (10/12, 83%). The overall predictive performance ranged from moderate to high. Based on classification metrics, deep learning models performed similar to or outperformed conventional machine learning models within the same studies. Models using only prior peripheral oxygen saturation as a clinical variable showed better performance than models based on multiple variables, with most of these studies (2/3, 67%) using a long short-term memory algorithm. Conclusions Machine learning models provide the potential to accurately predict the occurrence of hypoxic events based on retrospective data. The heterogeneity of the studies and limited generalizability of their results highlight the need for further validation studies to assess their predictive performance.
Collapse
Affiliation(s)
- Lena Pigat
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | | | | | - Julia Sander
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | - Mathias Kaspar
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | - Maximilian Schmutz
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
- Hematology and Oncology, University Hospital of Augsburg, Augsburg, Germany
| | - Sven Olaf Rohr
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | - Carl Mathis Wild
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
- Gynecology and Obstetrics, University Hospital of Augsburg, Augsburg, Germany
| | - Sebastian Goss
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | - Sarra Zaghdoudi
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
| | - Ludwig Christian Hinske
- Digital Medicine, University Hospital of Augsburg, Augsburg, Germany
- Department of Anaesthesiology, LMU University Hospital, LMU Munich, Munich, Germany
| |
Collapse
|