Computer Sciences and Information Technology

Title :

Area of research :

Focus area :

Principal Investigator :

Timeline Start Year :

Timeline End Year :

Contact info :

Executive Summary :

The problem of understanding medical reports in low-resource languages is a significant issue, particularly for underprivileged and illiterate individuals in rural areas. The proposed solution aims to scan medical reports, understand them thoroughly using artificial intelligence, and translate them in local languages for underprivileged and non-medical people. The data used in the research includes medical reports, scans, and routine blood tests from open-source platforms.

There are currently no end-to-end solutions for automatic translation of medical reports in low-resource languages, and current solutions for tele-consultation are not widely used by people who do not understand English. The proposed project aims to develop a translator service for medical reports in low-resource languages using artificial intelligence and natural language processing, tailored to low-resource Indian languages like Bengali, Assamese, and Telugu.

The system will consist of three major components: a module for translating medical reports into low-resource languages, a module for image-to-text conversion of medical images, and a module for consultation with primary health workers. The model will be trained on parallel corpus, perform text extraction from images, and fine-tune the model for low-resource language translation.

Total Budget (INR):

Implementing Agency :

Funding Agency :