Review Paper:
Assessing the
Current Status of Artificial Intelligence Applications for Predicting Microsatellite
Instability in Colon Cancer
Alsughayyir Jawaher
Res. J. Biotech.; Vol. 20(11); 402-416;
doi: https://doi.org/10.25303/2011rjbt4020416; (2025)
Abstract
Effective treatment decisions for colorectal cancer (CRC) depend on the histological
classification and microsatellite instability (MSI) status of the patient’s biopsy.
In recent years, artificial intelligence (AI) has emerged as a valuable tool in
the diagnostic process, offering efficiency, reducing the need for extensive manpower
and maintaining accuracy. This review explores recent advancements in AI technology
and its effectiveness in identifying prognostic biomarkers related to CRC and aims
to inform clinicians and gastroenterologists about novel patient management strategies.
A narrative non-systematic review of existing literature on using AI for detecting
deficient mismatch repair (dMMR)/MSI in CRC diagnosis was performed. Searches were
conducted in the PubMed database using a combination of keywords such as colorectal
cancer diagnosis, artificial intelligence and deep learning, focusing on publications
from 2019 onward.
The reviewed articles exhibited varying outcomes, with each utilizing the CNN model
under differing conditions like cohort types and sizes and convolution or filter
numbers, highlighting specific strengths and limitations for each model. AI-driven
predictive analytics offered researchers superior insights into genomics and proteomics
data, elevating patient characterization precision and streamlining pathology workflows.
