SIH26038: Explainable AI for Diabetic Retinopathy Screening in Rural India — SIH 2026

SIH26038: Explainable AI for Diabetic Retinopathy Screening in Rural India — SIH 2026

What they'll ask you

12

Generated from the brief's own wording and the competition's published rules — never from a guess about what this ministry prefers.

Specific benchmark datasets to use for validation?
The brief never answers this, so a panel will. Whatever you decide, say it the same way twice.
Exact hardware specifications of the portable fundus cameras being simulated?
The brief never answers this, so a panel will. Whatever you decide, say it the same way twice.
Has any part of this been shown at a previous event, hackathon or college project?
The guidelines are explicit: your solution must not have appeared in any previous event or programme, of any sort. A recycled project is what a team under time pressure reaches for.

9 more— the ones a panel gets to if there is time

The brief asks for image quality assessment. How would you build that?
Decoded from SIH26038 itself — Evaluates fundus images for focus and illumination, enhances borderline cases, and rejects ungradeable ones. The brief asks for it by name.
The brief asks for retinal structure segmentation. How would you build that?
Decoded from SIH26038 itself — Extracts clinical features such as the optic disc, blood vessels, microaneurysms, and hemorrhages. The brief asks for it by name.
The brief asks for dr severity grading. How would you build that?
Decoded from SIH26038 itself — Classifies the severity of diabetic retinopathy using the International Clinical DR scale. The brief asks for it by name.
The brief asks for explainability module. How would you build that?
Decoded from SIH26038 itself — Generates Grad-CAM attention maps and automated annotated reports for clinical validation. The brief asks for it by name.
The brief asks for simulink workflow simulation. How would you build that?
Decoded from SIH26038 itself — Models the telemedicine screening pipeline to simulate bandwidth constraints and optimize resource allocation. The brief asks for it by name.
Why not use what already exists? Name the closest thing to this that is already running.
A team that has not named the alternative themselves is answering this for the first time in the room.
Which single thing will you demonstrate end to end, start to finish, with nothing skipped?
Ours, not a rule: a narrow thing that fully works survives questioning better than a broad thing that half works. If nobody on the team can name it, that is the finding.
Show me this working: the jury will check whether your model achieves greater than 90 percent sensitivity and greater than 85 percent specificity for referable DR.
This is the evaluator read for your problem statement, decoded from the brief's own wording.
Show me this working: they will test your explainable Grad-CAM outputs for clinical usefulness and review your Simulink workflow simulation for resource optimization.
This is the evaluator read for your problem statement, decoded from the brief's own wording.

Take this to an AI

Pick what you are about to do and copy the prompt. It carries the organisers' own wording, the constraints they never spell out, and an instruction not to invent requirements they never set.

Who has this problem, what already exists, and what you would have to find out.

# Context: India's Smart India Hackathon 2026

I am on a student team entering SIH 2026. Problem statement SIH26038, set by MathWorks.
Its official title, word for word: "Explainable AI for Diabetic Retinopathy Screening in Rural India"
Theme: MedTech / BioTech / HealthTech. Category: software.

In plain terms: A MATLAB-based software pipeline that screens retinal images for diabetic retinopathy, providing automated grading, explainable heatmaps, and a Simulink resource model for rural healthcare workers.

## The organisers' full description, verbatim

Background: India has over 77 million diabetic adults - the second highest globally. Diabetic Retinopathy (DR) affects ~18% of this population and is a leading cause of preventable blindness. Early screening can prevent90% of vision loss, but India has only ~1 ophthalmologist per 100,000 rural population, making mass manual screening infeasible. Existing AI solutions function as black boxes, lack clinical validation rigor, and fail with variable image quality from portable fundus cameras in field conditions. A robust, explainable, and validated screening system is essential for deployment in primary healthcare centres across rural India. Description: Design a MATLAB-based retinal image analysis pipeline for automated DR screening addressing real-world deployment challenges: 1. Image Quality Assessment and Enhancement: Automatically evaluate fundus images for adequacy (focus, illumination, field of view). Apply adaptive enhancement (CLAHE, illumination normalization, denoising) for borderline images; reject ungradeable ones with recapture feedback. 2. Retinal Structure Segmentation: Extract clinically relevant structures - optic disc/fovea localization, vessel segmentation, microaneurysm detection, exudate segmentation, hemorrhage classification, and neovascularization detection. 3. DR Severity Grading: Classify using the International Clinical DR severity scale (Levels 0-4, from no DR to proliferative DR) with clinically acceptable sensitivity (>90%) and specificity (>85%) for referable DR (Level 2+). 4. Explainability Module: Implement Grad-CAM attention maps, lesion-level evidence correlated with clinical criteria, calibrated confidence scores, and automated annotated reports - enabling ophthalmologist validation in under 30 seconds for a human-in-theloop workflow. 5. Simulink Workflow Simulation: Model the telemedicine screening pipeline in Simulink - image acquisition rates, bandwidth constraints, processing throughput, and review capacity - to optimize resource allocation for district-level programs serving 100,000+ patients annually. This problem demands clinical validation rigor, sub-pixel microaneurysm detection, and clinically meaningful explainability • Tools: Image Processing Toolbox, Computer Vision Toolbox, Deep Learning Toolbox, Medical Imaging Toolbox, Simulink, Statistics and Machine Learning Toolbox Expected Solution: A working prototype demonstrating: DR classification with >90% sensitivity and >85% specificity for referable DR; explainable Grad-CAM outputs rated as clinically useful; a Simulink model optimizing screening resource allocation; and validation against published benchmarks showing the integrated pipeline outperforms any single technique approach.

## What the brief never says

- specific benchmark datasets to use for validation
- exact hardware specifications of the portable fundus cameras being simulated
## Constraints that are not negotiable

- The judged build is a prototype demonstrated in roughly 36 hours by students, not a product.
- A jury evaluates it against this problem statement's own wording, not against general merit.
- We submit on sih.gov.in; the problem statement number and title above must match exactly.
- Your solution must not have appeared in any previous event or programme, of any sort.
- The organisers attached a dataset, so build against theirs rather than inventing one.

## What I want from you

1. Who actually has this problem in India, and what they do about it today.
2. What already exists — government systems, open-source projects, published research — that we should not rebuild.
3. For each item under 'what the brief never says', what the realistic answer is likely to be and how I could verify it.
4. The two or three things about this domain that a student team would get wrong without local knowledge.
## How to answer

- Everything above between quotation marks is the organisers' own wording. Treat it as the requirement.
- Do NOT invent requirements the brief does not state. If something matters and is not stated, say so and ask me, rather than assuming.
- Where you are inferring rather than reading, mark it as your inference.
- Do not guess what this ministry wants from its name. Its wording above is the only evidence.

---
Problem statement facts assembled by Wynit from MathWorks's published brief for SIH 2026.
The plain-English summary, the list of parts and the evaluator read are our interpretation of that text; the quoted title and the full description are theirs.

The government's original description (342 words)

Background: India has over 77 million diabetic adults - the second highest globally. Diabetic Retinopathy (DR) affects ~18% of this population and is a leading cause of preventable blindness. Early screening can prevent90% of vision loss, but India has only ~1 ophthalmologist per 100,000 rural population, making mass manual screening infeasible. Existing AI solutions function as black boxes, lack clinical validation rigor, and fail with variable image quality from portable fundus cameras in field conditions. A robust, explainable, and validated screening system is essential for deployment in primary healthcare centres across rural India. Description: Design a MATLAB-based retinal image analysis pipeline for automated DR screening addressing real-world deployment challenges: 1. Image Quality Assessment and Enhancement: Automatically evaluate fundus images for adequacy (focus, illumination, field of view). Apply adaptive enhancement (CLAHE, illumination normalization, denoising) for borderline images; reject ungradeable ones with recapture feedback. 2. Retinal Structure Segmentation: Extract clinically relevant structures - optic disc/fovea localization, vessel segmentation, microaneurysm detection, exudate segmentation, hemorrhage classification, and neovascularization detection. 3. DR Severity Grading: Classify using the International Clinical DR severity scale (Levels 0-4, from no DR to proliferative DR) with clinically acceptable sensitivity (>90%) and specificity (>85%) for referable DR (Level 2+). 4. Explainability Module: Implement Grad-CAM attention maps, lesion-level evidence correlated with clinical criteria, calibrated confidence scores, and automated annotated reports - enabling ophthalmologist validation in under 30 seconds for a human-in-theloop workflow. 5. Simulink Workflow Simulation: Model the telemedicine screening pipeline in Simulink - image acquisition rates, bandwidth constraints, processing throughput, and review capacity - to optimize resource allocation for district-level programs serving 100,000+ patients annually. This problem demands clinical validation rigor, sub-pixel microaneurysm detection, and clinically meaningful explainability • Tools: Image Processing Toolbox, Computer Vision Toolbox, Deep Learning Toolbox, Medical Imaging Toolbox, Simulink, Statistics and Machine Learning Toolbox Expected Solution: A working prototype demonstrating: DR classification with >90% sensitivity and >85% specificity for referable DR; explainable Grad-CAM outputs rated as clinically useful; a Simulink model optimizing screening resource allocation; and validation against published benchmarks showing the integrated pipeline outperforms any single technique approach.