Title

AI-assisted diagnosis of anemia through peripheral smear image analysis: A cross-validation study

Authors

Ashita Nain^1,*, Sangeeta Gupta², Sylvester Noeldoss Lazarus³, Kawalinder Kaur Girgla⁴, Parth Jani⁵, Amrit Podder⁶, Sreemoyee Dutta⁷ & Ravi Babu Surisetti⁸

Affiliation

¹Department of Physiology, Lala Lajpat Rai Memorial Medical College, Meerut Uttar Pradesh, India; ²Department of Physiology, AIIMS Gorakhpur, Uttar Pradesh, India; ³Department of Pathology, American University of Barbados, Wildey, Barbados (Caricom); ⁴Department of Physiology, SGRD Institute of Medical Sciences and Research, Sri Amritsar, Punjab, India; ⁵Department of General Medicine, All India Institute of Medical Sciences, Rajkot, Gujarat, India; ⁶Department of Physiology, Teerthanker Mahaveer Medical College & Research Centre, Teerthanker Mahaveer University, Moradabad, Uttar Pradesh, India; ⁷Department of Periodontology, Kalinga Institute of Dental Sciences, Bhubaneswar, Odisha, India; ⁸Department of Zoology, Trans, Disciplinary Research Hub, Andhra University, Visakhapatnam Andhra Pradesh, India; *Corresponding author

Email

Ashita Nain - E-mail: drashitanain@gmail.com
Sangeeta Gupta - E-mail: drsangeeta77.65@rediffmail.com
Sylvester Noeldoss Lazarus - E-mail: dr.sylvestermd@gmail.com
Kawalinder Kaur Girgla - E-mail: girglaphysio@gmail.com
Parth Jani - E-mail: parthjani13@gmail.com
Amrit Podder - E-mail: amritpodder0@gmail.com
Sreemoyee Dutta - E-mail: mailsreemoyeedutta@gmail.com
Ravi Babu Surisetti - E-mail: ravibabusurisetti@gmail.com

Article Type

Research Article

Date

Received October 1, 2025; Revised October 31, 2025; Accepted October 31, 2025, Published October 31, 2025

Abstract

A deep semi-supervised learning model for automating anemia detection and classification from peripheral blood smear images is of interest. A convolutional neural network was trained on 3,200 images, with only 25% annotated by expert hematologists. The model achieved a classification accuracy of 93.4% and F1-scores above 90% for key anemia subtypes, demonstrating strong agreement with expert diagnoses (κ = 0.89). It significantly reduced diagnostic time and performed well in detecting microcytic and sickle cell anemia. This AI-based framework shows great potential for accurate anemia diagnosis, especially in resource-limited settings.

Keywords

Anemia diagnosis, deep learning, peripheral blood smear, semi-supervised learning, red blood cells, medical image analysis

Citation

Nain et al. Bioinformation 21(10): 3668-3672 (2025)

Edited by

Ritik Kashwani

ISSN

0973-2063

Publisher

Biomedical Informatics

License

This is an Open Access article which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. This is distributed under the terms of the Creative Commons Attribution License.