A systematic review of machine learning-based tumor-infiltrating lymphocytes analysis in colorectal cancer: Overview of techniques, performance metrics, and clinical outcomes

• Published in: Computers in biology and medicine Vol. 173; p. 108306
• Main Authors: Kazemi, Azar, Rasouli-Saravani, Ashkan, Gharib, Masoumeh, Albuquerque, Tomé, Eslami, Saeid, Schüffler, Peter J.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.05.2024; Elsevier Limited
• Subjects: Artificial intelligence; Automation; Biomarkers; Cancer; Classification; Clinical outcomes; Colorectal cancer; Colorectal carcinoma; Decision making; Deep learning; Diagnosis; Human error; Identification; Immune response; Learning algorithms; Lymphocytes; Machine learning; Medical prognosis; Microenvironments; Minority & ethnic groups; Neural networks; Pathology; Patients; Performance measurement; Survival; Systematic review; Tumor-infiltrating lymphocytes; Tumors; Whole slide images; Whole slide images; Tumor-infiltrating lymphocytes; Survival; Machine learning; Colorectal cancer
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108306

The incidence of colorectal cancer (CRC), one of the deadliest cancers around the world, is increasing. Tissue microenvironment (TME) features such as tumor-infiltrating lymphocytes (TILs) can have a crucial impact on diagnosis or decision-making for treating patients with CRC. While clinical studies showed that TILs improve the host immune response, leading to a better prognosis, inter-observer agreement for quantifying TILs is not perfect. Incorporating machine learning (ML) based applications in clinical routine may promote diagnosis reliability. Recently, ML has shown potential for making progress in routine clinical procedures. We aim to systematically review the TILs analysis based on ML in CRC histological images. Deep learning (DL) and non-DL techniques can aid pathologists in identifying TILs, and automated TILs are associated with patient outcomes. However, a large multi-institutional CRC dataset with a diverse and multi-ethnic population is necessary to generalize ML methods. [Display omitted] •We present a review of the CRC TILs analysis by machine learning to bridge the gap that exists in the literature.•Deep learning techniques are promising in the domain.•Although there is disagreement in some studies, overall, a higher density of TILs is associated with favorable outcomes.•The TILs analysis domain faces challenges, including a lack of diverse and large datasets.•Further validation of promising models is needed by using large heterogeneous datasets.