Deep learning from “passive feeding” to “selective eating” of real-world data
Open Access
- 30 October 2020
- journal article
- research article
- Published by Springer Science and Business Media LLC in npj Digital Medicine
- Vol. 3 (1), 1-9
- https://doi.org/10.1038/s41746-020-00350-y
Abstract
Artificial intelligence (AI) based on deep learning has shown excellent diagnostic performance in detecting various diseases with good-quality clinical images. Recently, AI diagnostic systems developed from ultra-widefield fundus (UWF) images have become popular standard-of-care tools in screening for ocular fundus diseases. However, in real-world settings, these systems must base their diagnoses on images with uncontrolled quality (“passive feeding”), leading to uncertainty about their performance. Here, using 40,562 UWF images, we develop a deep learning–based image filtering system (DLIFS) for detecting and filtering out poor-quality images in an automated fashion such that only good-quality images are transferred to the subsequent AI diagnostic system (“selective eating”). In three independent datasets from different clinical institutions, the DLIFS performed well with sensitivities of 96.9%, 95.6% and 96.6%, and specificities of 96.6%, 97.9% and 98.8%, respectively. Furthermore, we show that the application of our DLIFS significantly improves the performance of established AI diagnostic systems in real-world settings. Our work demonstrates that “selective eating” of real-world data is necessary and needs to be considered in the development of image-based AI systems.Keywords
This publication has 39 references indexed in Scilit:
- ULTRA-WIDEFIELD FUNDUS IMAGINGRetina, 2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Quality evaluation of digital fundus images through combined measuresJournal of Medical Imaging, 2014
- Ultra-wide field imaging of retinopathy of prematurity (ROP) using Optomap-200TXBMJ Case Reports, 2013
- Validating Retinal Fundus Image Analysis Algorithms: Issues and a ProposalInvestigative Ophthalmology & Visual Science, 2013
- Assessment of Diabetic Retinopathy Using Nonmydriatic Ultra-Widefield Scanning Laser Ophthalmoscopy (Optomap) Compared With ETDRS 7-Field Stereo PhotographyDiabetes Care, 2012
- Nonmydriatic Ultrawide Field Retinal Imaging Compared With Dilated Standard 7-Field 35-mm Photography and Retinal Specialist Examination for Evaluation of Diabetic RetinopathyAmerican Journal of Ophthalmology, 2012
- An automated retinal image quality grading algorithm2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2011
- Ophthalmic imaging today: an ophthalmic photographer's viewpoint – a reviewClinical & Experimental Ophthalmology, 2009
- Nonmydriatic screening for diabetic retinopathy by ultra-widefield scanning laser ophthalmoscopy (Optomap)Albrecht von Graefes Archiv für Ophthalmologie, 2007