Poster: NCD Screening Focus
Smartphone AI for Multi-NCD Screening in Rural India
Authors & Affiliations
- Umathurappan Chandrasekar, BE (Mech)
Principal IT Architect, Infraspace, Apple Valley, MN, USA - Kannaiayan Prakash, BE (CS)
Senior Engineering Manager, Huntington National Bank, Minneapolis, MN, USA - Radhika Gupta, BDS, MPH (Epidemiology)
Public Health Specialist, Infraspace, Apple Valley, MN, USA
Correspondence: Umathurappan Chandrasekar, 15224 Florist Circle, Apple Valley, MN 55124; uma@infraspace.net
Conflict of Interest & Funding: No conflicts of interest. Funding: This work received no external funding.
Abstract
Background: Diabetes projected 200M+ cases by 2025; 70% rural undiagnosis strains NPCDCS.
Objective: Validate smartphone AI for retinopathy/hypertension/oral cancer screening across 10 states.
Methods: Edge AI models (40k NPCDCS cases): fundus selfie CNNs, voice BP transformers, oral lesion detection. Multilingual apps co-designed with 200 ASHAs. UP/Bihar pilots (n=8,000).
Results: Retinopathy 95% sensitivity, hypertension 90%, oral lesions 92%. Transportability >88% across states. Fairness <3% caste/gender bias. Pilots: +40% early detection, 50% cost reduction.
Conclusion: Offline smartphone AI scales NCD screening equitably via participatory design and governance.
Table 1: Multi-NCD Screening Metrics
| Disease | Sensitivity | Specificity | State Transportability | Cost Reduction |
|---|---|---|---|---|
| Retinopathy | 95% | 92% | 89% | 50% |
| Hypertension | 90% | 88% | 88% | 50% |
| Oral Cancer | 92% | 90% | 88% | 50% |
View Posters
