Medical Reports Digitizer
- Developed and implemented a Computer Vision pipeline to segment medical reports by identifying the document structure.
- Resulting document segments were then used to extract medical information using OCR.
- Entire pipeline is implemented on an Android Application using OpenCV and Tesserract OCR Libraries.
- Results obtained from OCR of user’s reports were then fed into a LSTM based RNN Network to predict user’s health and advice any changes to benefit their health.
