Clinical explainable differential diagnosis of polypoidal choroidal vasculopathy and age-related macular degeneration using deep learning

Da Ma; Meenakshi Kumar; Vikas Khetan; Parveen Sen; Muna Bhende; Shuo Chen; Timothy T L Yu; Sieun Lee; Eduardo V Navajas; Joanne A Matsubara; Myeong Jin Ju; Marinko V Sarunic; Rajiv Raman; Mirza Faisal Beg

doi:10.1016/j.compbiomed.2022.105319

Clinical explainable differential diagnosis of polypoidal choroidal vasculopathy and age-related macular degeneration using deep learning

Comput Biol Med. 2022 Apr:143:105319. doi: 10.1016/j.compbiomed.2022.105319. Epub 2022 Feb 14.

Authors

Affiliations

¹ Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC, USA; School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada. Electronic address: dma@wakehealth.edu.
² Shri Bhagwan Mahavir Vitreoretinal Service, Medical Research Foundation, Sankara Nethralaya, Chennai, India.
³ School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada.
⁴ School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada; Mental Health & Clinical Neurosciences, School of Medicine, University of Nottingham, Nottingham, United Kingdom.
⁵ Department of Ophthalmology & Visual Sciences, The University of British Columbia, Vancouver, BC, Canada; University of British Columbia Vancouver General Hospital, Eye Care Centre, Vancouver, BC, Canada.
⁶ Department of Ophthalmology & Visual Sciences, The University of British Columbia, Vancouver, BC, Canada; University of British Columbia Vancouver General Hospital, Eye Care Centre, Vancouver, BC, Canada; School of Biomedical Engineering, University of British Columbia, BC, Canada.
⁷ School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada; Institute of Ophthalmology, University College London, London, UK; Department of Medical Physics and Biomedical Engineering, University College London, United Kingdom.
⁸ Shri Bhagwan Mahavir Vitreoretinal Service, Medical Research Foundation, Sankara Nethralaya, Chennai, India. Electronic address: rajivpgraman@gmail.com.
⁹ School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada. Electronic address: faisal-lab@sfu.ca.

PMID: 35220077
DOI: 10.1016/j.compbiomed.2022.105319

Abstract

Background: This study aims to achieve an automatic differential diagnosis between two types of retinal pathologies with similar pathological features - Polypoidal choroidal vasculopathy (PCV) and wet age-related macular degeneration (AMD) from volumetric optical coherence tomography (OCT) images, and identify clinically-relevant pathological features, using an explainable deep-learning-based framework.

Methods: This is a retrospective study with data from a cross-sectional cohort. The OCT volume of 73 eyes from 59 patients was included in this study. Disease differentiation was achieved through single-B-scan-based classification followed by a volumetric probability prediction aggregation step. We compared different labeling strategies with and without identifying pathological B-scans within each OCT volume. Clinical interpretability was achieved through normalized aggregation of B-scan-based saliency maps followed by maximum-intensity-projection onto the en face plane. We derived the PCV score from the proposed differential diagnosis framework with different labeling strategies. The en face projection of saliency map was validated with the pathologies identified in Indocyanine green angiography (ICGA).

Results: Model trained with both labeling strategies achieved similar level differentiation power (>90%), with good correspondence between pathological features detected from the projected en face saliency map and ICGA.

Conclusions: This study demonstrated the potential clinical application of non-invasive differential diagnosis using AI-driven OCT-based analysis, with minimal requirement of labeling efforts, along with clinical explainability achieved through automatically detected disease-related pathologies.

Keywords: Age-related macular degeneration; Convolutional Neural Network; Deep learning; Differential diagnosis; Explainable AI model; Optical coherence tomography; Polypoidal choroidal vasculopathy.