Page 131 - AIH-2-3

Artificial Intelligence in Health

ORIGINAL RESEARCH ARTICLE
Screening and early detection of cervical

intraepithelial neoplasia and cervicitis using a
hemoglobin absorption map-derived machine

learning algorithm

Phebe George 1 , Rekha Upadhya 2 , Rinoy Suvarnadas 1 ,
1
Niranjana Sampthalia 3 , and Subhash Narayanan *
1 Research and Development Division, Sascan Meditech Pvt. Ltd., TIMed, Sree Chitra Tirunal
Institute for Medical Sciences and Technology, Thiruvananthapuram, Kerala, India
2 Department of Obstetrics and Gynaecology, Kasturba Medical College, Manipal Academy of Higher
Education, Manipal, Karnataka, India
3 Department of Biomedical Engineering, Manipal Institute of Technology, Manipal Academy of
Higher Education, Manipal, Karnataka, India

Abstract

Early and non-invasive detection of cervical malignancy holds great clinical significance.
Diffuse reflectance (DR) spectroscopy has the capability to map tissue transformation
*Corresponding author: at the biochemical, morphological, and cellular levels. We have developed a non-
Subhash Narayanan invasive, multimodal imaging system to map changes in tissue autofluorescence using
(subhash@sascan.in) DR for the screening and early detection of cervical cancer and cervical inflammation
Citation: George P, Upadhya R, (cervicitis). The developed multispectral imaging device consists of light-emitting
Suvarnadas R, Sampthalia N, diodes (LED) emitting at 375, 545, 575, and 610 nm wavelengths, along with a
Narayanan S. Screening and early
detection of cervical intraepithelial 5-megapixel monochrome camera for image acquisition. Camera operation and image
neoplasia and cervicitis using a analysis are controlled using proprietary software installed on a Windows tablet. The
hemoglobin absorption map-derived 375 nm LED-excited autofluorescence, and the elastically backscattered light at 545,
machine learning algorithm. Artif 575, and 610 nm originating from the cervix tissue are captured by the camera and
Intell Health. 2025;2(3):125-137.
doi: 10.36922/aih.8527 processed to assess tissue abnormalities. A machine learning (ML) algorithm based on
DR image intensity ratio values was developed for tissue classification. It was observed
Received: January 14, 2025
that the R610/R545 image ratio could discriminate malignant cervical sites from
Revised: March 6, 2025 normal tissues, achieving a sensitivity of 100% and specificity of 93%. In comparison,
Accepted: April 10, 2025 cervicitis could be discriminated from normal tissues using the R610/R575 ratio, with
a sensitivity of 91.6% and specificity of 94.4%. The study demonstrates the potential
Published online: May 2, 2025
of DR imaging in conjunction with ML algorithm to non-invasively screen and detect
Copyright: © 2025 Author(s). cervical intraepithelial neoplasia and cervicitis in real time. As compared to the existing
This is an Open-Access article
distributed under the terms of the practice of Pap smear and colposcopy-directed biopsy, which are subjective and require
Creative Commons Attribution a waiting period for results, objective screening using CerviScan would help reduce
License, permitting distribution, patient anxiety, unnecessary biopsies, and treatment costs. With increased patient
and reproduction in any medium,
provided the original work is screening, the accuracy of the ML algorithm would improve. When integrated into a
properly cited. cloud server, the system could address the needs of multiple users in a field setting.
Publisher’s Note: AccScience
Publishing remains neutral with Keywords: Cervical intraepithelial neoplasia; Cervical inflammation; Diffuse reflectance
regard to jurisdictional claims in
published maps and institutional image intensity ratio; Machine learning algorithm
affiliations.
Volume 2 Issue 3 (2025) 125 doi: 10.36922/aih.8527