A Deep Learning-based System for Detecting Anemia from Eye Conjunctiva Images taken from a Smartphone

Abstract

Anemia is a severe health condition commonly prevalent among women of reproductive age and children below five years. Screening patients before the condition becomes critical and can save many lives. World Health Organization (WHO) has set the “Global nutrition target 2025-anemia,” aiming to reduce 50% of anemia cases among women of reproductive age. This target can be achieved through a time-efficient, cost-effective, and easy-to-use tool. Traditional testing methods require specific chemicals, machines, and equipment that are not available everywhere. It also requires the presence of nurses, laboratory workers, and doctors. These methods are costly, time-consuming, and produce biohazard waste, thus polluting the environment. We developed an Artificial Intelligence (AI)-based bot that can be used for screening people for anemia. The bot service is based on two models: a segmentation model to segment the Region of Interest (ROI) and a classification model to classify anemic cases from normal ones. To train the model, we have collected data from 160 anemic and 140 non-anemic persons. In this paper, we have explained the architecture of the models, all the training parameters, and their deployment on cloud services using the REAN chatbot service. We manage to reach an Intersection Over Union (IOU) score of 0.922 for the segmentation model; validation recall of 0.95 and validation accuracy of 0.9699 for the classification model. This system is easy to use and does not depend on the availability of comprehensive laboratory infrastructure or trained personnel and thus can enable screening of anemia in low-resource settings.

KEYWORDS:

• Anemia
• Chatbot
• Convolution neural networks
• Deep learning
• Image classification
• Image segmentation
• Non-invasive disease diagnosis
• Transfer learning
• U-Net

Acknowledgements

We want to thank all the clinical coordinators who helped collect the required data. A special thanks to Mr. Sri Vasireddy, founder and CEO of REAN Foundation, for providing the opportunity and motivation to create a social impact and funding the above research. We would also like to thank the Department of Biomedical Engineering, Indian Institute of Technology, Ropar for providing us the necessary support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Pallavi

Pallavi received the BTech degree in biotechnology from National Institute of Technology Jalandhar, India in 2019 and the MTech degree in biomedical engineering from Indian Institute of Technology Ropar, India in 2022. She is currently working as a senior AI engineer in Rean Foundation. Her areas of interest are computer vision and natural language processing. Email: [email protected]

Bijit Basumatary

Bijit Basumatary received the MTech degree in biomedical engineering from Indian Institute of Technology Ropar and later joined as PhD scholar in August, 2021. Earlier he received BTech degree in instrumentation engineering from Central Institute of Technology Kokrajhar, Assam in 2018. He is currently pursuing PhD in the Department of Biomedical Engineering, Indian Institute of Technology Ropar, Punjab. His areas of interest are functional electrical stimulation, biomedical instrumentation and electromyography.

Rahul Shukla

Rahul Shukla received the integrated MTech in biomedical engineering from the Indian Institute of Information Technology, Allahabad, India in 2018. He is currently pursuing PhD at the Department of Biomedical Engineering, Indian Institute of Technology, Ropar, India. His areas of interest are seizure detection devices, apps for seizure patients, the brain-computer interface, and machine learning for biological applications. Email: [email protected]

Rakesh Kumar

Rakesh Kumar received the MBBS degree from Govt Medical College, Sangrur Road, Patiala in 2010 and Master of Science (MS) in machine learning AI from Liverpool John Moores University and PG diploma in machine learning AI from International Institute of Information Technology Bangalore in 2022. He is currently working as medical officer at Rean Foundation. His area of interest is AI in healthcare. Email: [email protected]

Bodhisatwa Das

Bodhisatwa Das received his PhD from IIT Kharagpur and postdoctoral degree from Rutgers University. He received MTech degree in clinical engineering from Indian Institute of Technology, Madras. He is currently working as an assistant professor in Department of Biomedical Engineering of Indian Institute of Technology Ropar. His areas of interest are biomaterials, tissue engineering, nanomedicine and wound healing. Email: [email protected]

Ashish Kumar Sahani

Ashish Kumar Sahani received his PhD from IIT Madras and postdoctoral training from Harvard Medical School and University of Michigan. He is working as an assistant professor in Department of Biomedical Engineering of Indian Institute of Technology Ropar, where he is heading the Medical Devices Lab. His areas of interest are medical devices, instrumentation, machine learning and signal processing. Email: [email protected]