AmritaSync: A Deep Learning based Anaemia Tracker for Non-Invasive Detection

The World Health Organization identifies Anaemia as a health hazard condition affecting a quarter of the total world population, necessitating automated, quick, and reliable detection. To address this, an application was designed to capture eye and fingernail images using a mobile phone and classify...

Full description

Saved in:
Bibliographic Details
Main Authors: P, Bharath Anuj, Chalicham, Divyanth, Purna, Amrutham Varshit, Venkata Siddhartha, Alla, Jeyakumar, G., Janci Rani, P.R, Kolandapalayam Shanmugam, Selvanayaki, Thangavel, Senthil Kumar
Format: Conference Proceeding
Language:English
Subjects:
Anaemia
extraction
Hazards
localization
Location awareness
Medical services
Mobile handsets
Organizations
Pipelines
prepossessing pipeline
recommendation system
Sociology
World Health Organization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The World Health Organization identifies Anaemia as a health hazard condition affecting a quarter of the total world population, necessitating automated, quick, and reliable detection. To address this, an application was designed to capture eye and fingernail images using a mobile phone and classify them as anaemic or non-anaemic with accuracy as high as 91. A prepossessing pipeline was also presented for accurate localization and extraction of the eye and fingernail region from the input image. The proposed system has the potential to provide fast and efficient Anaemia diagnosis, with a recommendation system for patients to select a diet or consult a doctor. Additionally, the application shows the exact region filtered by the CNN algorithm.
ISSN:2767-7788
DOI:10.1109/ICICT60155.2024.10544545