Discriminative ensemble meta-learning with co-regularization for rare fundus diseases diagnosis

Gao，Mengdi1,2,3,4,5; Jiang，Hongyang6; Zhu，Lei1,2,3,4,5; Jiang，Zhe1,2,3,4,5; Geng，Mufeng1,2,3,4,5; Ren，Qiushi1,2,3,4,5; Lu，Yanye1,3,4

2023-10-01

10.1016/j.media.2023.102884

Medical Image Analysis   Impact Factor & Quartile

1361-8415

1361-8423

Deep neural networks (DNNs) have been widely applied in the medical image community, contributing to automatic ophthalmic screening systems for some common diseases. However, the incidence of fundus diseases patterns exhibits a typical long-tailed distribution. In clinic, a small number of common fundus diseases have sufficient observed cases for large-scale analysis while most of the fundus diseases are infrequent. For these rare diseases with extremely low-data regimes, it is challenging to train DNNs to realize automatic diagnosis. In this work, we develop an automatic diagnosis system for rare fundus diseases, based on the meta-learning framework. The system incorporates a co-regularization loss and the ensemble-learning strategy into the meta-learning framework, fully leveraging the advantage of multi-scale hierarchical feature embedding. We initially conduct comparative experiments on our newly-constructed lightweight multi-disease fundus images dataset for the few-shot recognition task (namely, FundusData-FS). Moreover, we verify the cross-domain transferability from miniImageNet to FundusData-FS, and further confirm our method's good repeatability. Rigorous experiments demonstrate that our method can detect rare fundus diseases, and is superior to the state-of-the-art methods. These investigations demonstrate that the potential of our method for the real clinical practice is promising.

Deep learning Ensemble-learning Few-shot learning Meta-learning Rare conditions

SCI

English

Others

Natural Science Foundation of Beijing Municipality[Z210008];

Computer Science ; Engineering ; Radiology, Nuclear Medicine & Medical Imaging

Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications ; Engineering, Biomedical ; Radiology, Nuclear Medicine & Medical Imaging

WOS:001039840100001

ELSEVIER

COMPUTER SCIENCE

2-s2.0-85165123828

Scopus

1.Institute of Medical Technology,Peking University Health Science Center,Peking University,Beijing,100191,China
2.Department of Biomedical Engineering,College of Future Technology,Peking University,Beijing,100871,China
3.National Biomedical Imaging Center,Peking University,Beijing,100871,China
4.Institute of Biomedical Engineering,Peking University Shenzhen Graduate School,Shenzhen,518055,China
5.Institute of Biomedical Engineering,Shenzhen Bay Laboratory 5F,Shenzhen,518071,China
6.Department of Computer Science and Engineering,Southern University of Science and Technology,Shenzhen,518055,China

Gao，Mengdi,Jiang，Hongyang,Zhu，Lei,et al. Discriminative ensemble meta-learning with co-regularization for rare fundus diseases diagnosis[J]. Medical Image Analysis,2023,89.

Gao，Mengdi.,Jiang，Hongyang.,Zhu，Lei.,Jiang，Zhe.,Geng，Mufeng.,...&Lu，Yanye.(2023).Discriminative ensemble meta-learning with co-regularization for rare fundus diseases diagnosis.Medical Image Analysis,89.

Gao，Mengdi,et al."Discriminative ensemble meta-learning with co-regularization for rare fundus diseases diagnosis".Medical Image Analysis 89(2023).