Publications by authors named "Mihir Deshmukh"
Article Synopsis
- The study develops a new biological ageing marker called RetiPhenoAge using deep learning algorithms that analyze retinal images to predict phenotypic age, surpassing traditional chronological age evaluations.
- Researchers trained a convolutional neural network on retinal photographs from the UK Biobank to identify patterns linked to various health biomarkers and assess the marker’s effectiveness in predicting morbidity and mortality across three independent cohorts.
- The study also compares RetiPhenoAge with other ageing markers and investigates its relationship with systemic health conditions and genetic factors, employing various statistical models to evaluate risks associated with mortality and illness.
Aims: Computer-aided detection systems for retinal fluid could be beneficial for disease monitoring and management by chronic age-related macular degeneration (AMD) and diabetic retinopathy (DR) patients, to assist in disease prevention via early detection before the disease progresses to a "wet AMD" pathology or diabetic macular edema (DME), requiring treatment. We propose a proof-of-concept AI-based app to help predict fluid via a "fluid score", prevent fluid progression, and provide personalized, serial monitoring, in the context of predictive, preventive, and personalized medicine (PPPM) for patients at risk of retinal fluid complications.
Methods: The app comprises a convolutional neural network-Vision Transformer (CNN-ViT)-based segmentation deep learning (DL) network, trained on a small dataset of 100 training images (augmented to 992 images) from the Singapore Epidemiology of Eye Diseases (SEED) study, together with a CNN-based classification network trained on 8497 images, that can detect fluid vs.
Background: Granular dystrophy is the most common stromal dystrophy. To perform automated segmentation of corneal stromal deposits, we trained and tested a deep learning (DL) algorithm from patients with corneal stromal dystrophy and compared its performance with human segmentation.
Methods: In this retrospective cross-sectional study, we included slit-lamp photographs by sclerotic scatter from patients with corneal stromal dystrophy and real-world slit-lamp photographs via various techniques (diffuse illumination, tangential illumination, and sclerotic scatter).
Response to: Revisiting the Problem of Optic Nerve Detection in a Retinal Image Using Automated Machine Learning.
Asia Pac J Ophthalmol (Phila)
January 2022
Article Synopsis
- The study assesses the effectiveness of deep learning (DL) algorithms in detecting pterygium—an eye condition—using color photographs taken with slit-lamp and handheld cameras.
- Researchers analyzed a total of 2503 images from a major eye disease study, applied the algorithms, and validated their performance on internal and two external test image sets.
- Results showed high accuracy in detecting pterygium, with sensitivity and specificity rates suggesting these DL algorithms could serve as a practical screening tool for identifying serious cases of pterygium in clinical settings.
Background: The application of deep learning to retinal photographs has yielded promising results in predicting age, sex, blood pressure, and haematological parameters. However, the broader applicability of retinal photograph-based deep learning for predicting other systemic biomarkers and the generalisability of this approach to various populations remains unexplored.
Methods: With use of 236 257 retinal photographs from seven diverse Asian and European cohorts (two health screening centres in South Korea, the Beijing Eye Study, three cohorts in the Singapore Epidemiology of Eye Diseases study, and the UK Biobank), we evaluated the capacities of 47 deep-learning algorithms to predict 47 systemic biomarkers as outcome variables, including demographic factors (age and sex); body composition measurements; blood pressure; haematological parameters; lipid profiles; biochemical measures; biomarkers related to liver function, thyroid function, kidney function, and inflammation; and diabetes.