Cervical Cancer Screening: Artificial Intelligence Algorithm for Automatic Diagnostic Support

Mirugwe A and Ashaba C

Published on: 1970-01-01

Abstract

Introduction: Cervical cancer is the fourth most prevalent  cancer among women worldwide and is a significant contributor to cancer- related deaths, with an estimated 300,000 women losing their lives to the disease annually. Most of these fatalities occur in low and middle- income countries (LMICs), such as Uganda, where access to screening and treatment options is limited. Early detection of cervical cancer is crucial to improve the chances of survival for patients. Currently, cervi- cal cancer screening is typically performed through Pap smears, which involve manual examination of cervical samples for abnormalities by med- ical experts. This process is costly, time-consuming, and prone to errors, leading to inaccurate diagnoses. Therefore, it is essential to find more effective and efficient alternative methods for cervical cancer screening to improve access in LMICs and alleviate the burden of cervical cancer.

Objective: The purpose  of  this study   is  to develop    an automated     pre-cervical cancer  screening algorithm  to  detect   precancerous cervical lesions.

Methodology: We developed a cancer screening algorithm using a 21- layer deep-learning convolution neural network trained on a dataset of 2300 images collected from local sources and some obtained from Kaggle.

Results: The best-performing      classifier      had      an      AUC      of the   accuracy   of   91.37%,   a   precision   of   88.80%,   a   recall   of 94.69%,    an    F1    score    of    91.65%,    and    an    AUC    of    96.0%.

Conclusion: The development and implementation of automated pre- cervical cancer screening  algorithms  have  the  potential  to  revolution- ize  cervical  cancer  detection  and  contribute  significantly  to  reducing the  burden  of  the  disease,  particularly  in  resource-limited  settings.