1. Automatic detection of multiple pathologies in fundus photographs using spin-off learning
- Author
-
Quellec, Gwenolé, Lamard, Mathieu, Conze, Pierre-Henri, Massin, Pascale, Cochener, Béatrice, Laboratoire de Traitement de l'Information Medicale (LaTIM), Université de Brest (UBO)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre Hospitalier Régional Universitaire de Brest (CHRU Brest)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-Institut Brestois Santé Agro Matière (IBSAM), Université de Brest (UBO), Département lmage et Traitement Information (IMT Atlantique - ITI), IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT), Service d'ophthalmologie [CHU Lariboisière], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Hôpital Lariboisière-Fernand-Widal [APHP], Assistance publique - Hôpitaux de Paris (AP-HP) (AP-HP)-Université Paris Diderot - Paris 7 (UPD7), Centre Hospitalier Régional Universitaire de Brest (CHRU Brest), CCSD, Accord Elsevier, Institut National de la Santé et de la Recherche Médicale (INSERM), Université de Brest (UBO)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre Hospitalier Régional Universitaire de Brest (CHRU Brest)-IMT Atlantique Bretagne-Pays de la Loire (IMT Atlantique), and IMT Atlantique Bretagne-Pays de la Loire (IMT Atlantique)
- Subjects
[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV] ,Few-shot learning ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,Rare conditions ,[INFO.INFO-IM]Computer Science [cs]/Medical Imaging ,[INFO.INFO-IM] Computer Science [cs]/Medical Imaging ,Deep learning ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,Diabetic retinopathy screening ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing - Abstract
International audience; In the last decades, large datasets of fundus photographs have been collected in diabetic retinopathy (DR) screening networks. Through deep learning, these datasets were used to train automatic detectors for DR and a few other frequent pathologies, with the goal to automate screening. One challenge limits the adoption of such systems so far: automatic detectors ignore rare conditions that ophthalmologists currently detect, such as papilledema or anterior ischemic optic neuropathy. The reason is that standard deep learning requires too many examples of these conditions. However, this limitation can be addressed with few-shot learning, a machine learning paradigm where a classifier has to generalize to a new category not seen in training, given only a few examples of this category. This paper presents a new few-shot learning framework that extends convolutional neural networks (CNNs), trained for frequent conditions, with an unsupervised probabilistic model for rare condition detection. It is based on the observation that CNNs often perceive photographs containing the same anomalies as similar, even though these CNNs were trained to detect unrelated conditions. This observation was based on the t-SNE visualization tool, which we decided to incorporate in our probabilistic model. Experiments on a dataset of 164,660 screening examinations from the OPHDIAT screening network show that 37 conditions, out of 41, can be detected with an area under the ROC curve (AUC) greater than 0.8 (average AUC: 0.938). In particular, this framework significantly outperforms other frameworks for detecting rare conditions, including multitask learning, transfer learning and Siamese networks, another few-shot learning solution. We expect these richer predictions to trigger the adoption of automated eye pathology screening, which will revolutionize clinical practice in ophthalmology.
- Published
- 2020