Dataset Information

Deep-Learning-Based Pre-Diagnosis Assessment Module for Retinal Photographs: A Multicenter Study.

ABSTRACT:

Purpose

Artificial intelligence (AI) deep learning (DL) has been shown to have significant potential for eye disease detection and screening on retinal photographs in different clinical settings, particular in primary care. However, an automated pre-diagnosis image assessment is essential to streamline the application of the developed AI-DL algorithms. In this study, we developed and validated a DL-based pre-diagnosis assessment module for retinal photographs, targeting image quality (gradable vs. ungradable), field of view (macula-centered vs. optic-disc-centered), and laterality of the eye (right vs. left).

Methods

A total of 21,348 retinal photographs from 1914 subjects from various clinical settings in Hong Kong, Singapore, and the United Kingdom were used for training, internal validation, and external testing for the DL module, developed by two DL-based algorithms (EfficientNet-B0 and MobileNet-V2).

Results

For image-quality assessment, the pre-diagnosis module achieved area under the receiver operating characteristic curve (AUROC) values of 0.975, 0.999, and 0.987 in the internal validation dataset and the two external testing datasets, respectively. For field-of-view assessment, the module had an AUROC value of 1.000 in all of the datasets. For laterality-of-the-eye assessment, the module had AUROC values of 1.000, 0.999, and 0.985 in the internal validation dataset and the two external testing datasets, respectively.

Conclusions

Our study showed that this three-in-one DL module for assessing image quality, field of view, and laterality of the eye of retinal photographs achieved excellent performance and generalizability across different centers and ethnicities.

Translational relevance

The proposed DL-based pre-diagnosis module realized accurate and automated assessments of image quality, field of view, and laterality of the eye of retinal photographs, which could be further integrated into AI-based models to improve operational flow for enhancing disease screening and diagnosis.

SUBMITTER: Yuen V

PROVIDER: S-EPMC8444486 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundDeep learning has been widely used for glaucoma diagnosis. However, there is no clinically validated algorithm for glaucoma incidence and progression prediction. This study aims to develop a clinically feasible deep-learning system for predicting and stratifying the risk of glaucoma onset and progression based on color fundus photographs (CFPs), with clinical validation of performance in external population cohorts.MethodsWe established data sets of CFPs and visual fields collected from longitudinal cohorts. The mean follow-up duration was 3 to 5 years across the data sets. Artificial intelligence (AI) models were developed to predict future glaucoma incidence and progression based on the CFPs of 17,497 eyes in 9346 patients. The area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity of the AI models were calculated with reference to the labels provided by experienced ophthalmologists. Incidence and progression of glaucoma were determined based on longitudinal CFP images or visual fields, respectively.ResultsThe AI model to predict glaucoma incidence achieved an AUROC of 0.90 (0.81-0.99) in the validation set and demonstrated good generalizability, with AUROCs of 0.89 (0.83-0.95) and 0.88 (0.79-0.97) in external test sets 1 and 2, respectively. The AI model to predict glaucoma progression achieved an AUROC of 0.91 (0.88-0.94) in the validation set, and also demonstrated outstanding predictive performance with AUROCs of 0.87 (0.81-0.92) and 0.88 (0.83-0.94) in external test sets 1 and 2, respectively.ConclusionOur study demonstrates the feasibility of deep-learning algorithms in the early detection and prediction of glaucoma progression.FUNDINGNational Natural Science Foundation of China (NSFC); the High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University; the Science and Technology Program of Guangzhou, China (2021), the Science and Technology Development Fund (FDCT) of Macau, and FDCT-NSFC.

Project description:International guidelines recommend deciding the treatment of colorectal lesions based on the estimated histology by endoscopic optical diagnosis. However, the theoretical and practical knowledge on optical diagnosis is not widely expanded The mail goal of this randomised controlled trial is to compare the pooled sensitivity of optical diagnosis for predicting deep submucosal invasion in large non-pedunculated polyps > 20 mm assessed in routine colonoscopies of gastroenterologists attending a e-learning module (intervention group) vs gastroenterologists who do not (control group) The main questions the study aims to answer are: * Is the pooled sensitivity of optical diagnosis for predicting deep submucosal invasion in large non-pedunculated polyps assessed in routine colonoscopies increased in those gastroenterologists participating in the e-learning module? * Is the pooled diagnostic accuracy of optical diagnosis for predicting deep sm invasion in large non-pedunculated polyps ≥ 20 mm assessed in routine colonoscopies increased in those gastroenterologists participating in the e-learning module? * In lesions with submucosal invasion, is the en bloc and complete resection rate (R0) increased in those gastroenterologists participating in the e-learning module? * In lesions referred to surgery, is the pooled benign polyps rate decreased in those gastroenterologists participating in the e-learning module? * In lesions treated with advanced en bloc procedures (ESD, TAMIS, fullthickness resection), is the pooled rate of histology with high-grade dysplasia, intramucosal cancer or submucosal invasion increased in those gastroenterologists participating in the e-learning module? * In lesions treated with piecemeal endoscopic resection, is the pooled rate of histology with high-grade dysplasia, intramucosal cancer or submucosal invasion decreased in those gastroenterologists participating in the e-learning module? * Is the diagnostic accuracy for predicting deep submucosal invasion in a test with pictures increased after participating in the e-learning module? The participants (or subjects of study) are gastroenterologists. They will be randomised to do the e-learning course (intervention group) or not (control group). Researchers will compare clinical outcomes of gastroenterologists participating in the e-learning module vs gastroenterologists not participating in the e-learning module to see if: * the pooled sensitivity of optical diagnosis for predicting deep submucosal invasion in large non-pedunculated polyps > 20 mm assessed in routine colonoscopies is increased. * the pooled diagnostic accuracy of optical diagnosis for predicting deep sm invasion in large non-pedunculated polyps > 20 mm is increased. * the en bloc and complete resection rate (R0) is increased in lesions with submucosal invasion. * the pooled benign polyps rate decreased in lesions referred to surgery. * the pooled rate of histology with high-grade dysplasia, intramucosal cancer or submucosal invasion increased in lesions treated with advanced en bloc procedures (ESD, TAMIS, fullthickness resection). * the pooled rate of histology with high-grade dysplasia, intramucosal cancer or submucosal invasion decreased in lesions treated with piecemeal endoscopic resection. * the diagnostic accuracy for predicting deep submucosal invasion in a test with pictures after participating is increased.

Project description:PurposeTo develop artificial intelligence (AI)-based deep learning (DL) models for automatically detecting the ischemia type and the non-perfusion area (NPA) from color fundus photographs (CFPs) of patients with branch retinal vein occlusion (BRVO).MethodsThis was a retrospective analysis of 274 CFPs from patients diagnosed with BRVO. All DL models were trained using a deep convolutional neural network (CNN) based on 45 degree CFPs covering the fovea and the optic disk. We first trained a DL algorithm to identify BRVO patients with or without the necessity of retinal photocoagulation from 219 CFPs and validated the algorithm on 55 CFPs. Next, we trained another DL algorithm to segment NPA from 104 CFPs and validated it on 29 CFPs, in which the NPA was manually delineated by 3 experienced ophthalmologists according to fundus fluorescein angiography. Both DL models have been cross-validated 5-fold. The recall, precision, accuracy, and area under the curve (AUC) were used to evaluate the DL models in comparison with three types of independent ophthalmologists of different seniority.ResultsIn the first DL model, the recall, precision, accuracy, and area under the curve (AUC) were 0.75 ± 0.08, 0.80 ± 0.07, 0.79 ± 0.02, and 0.82 ± 0.03, respectively, for predicting the necessity of laser photocoagulation for BRVO CFPs. The second DL model was able to segment NPA in CFPs of BRVO with an AUC of 0.96 ± 0.02. The recall, precision, and accuracy for segmenting NPA was 0.74 ± 0.05, 0.87 ± 0.02, and 0.89 ± 0.02, respectively. The performance of the second DL model was nearly comparable with the senior doctors and significantly better than the residents.ConclusionThese results indicate that the DL models can directly identify and segment retinal NPA from the CFPs of patients with BRVO, which can further guide laser photocoagulation. Further research is needed to identify NPA of the peripheral retina in BRVO, or other diseases, such as diabetic retinopathy.