Using Machine Learning Models to Predict Genitourinary Involvement Among Gastrointestinal Stromal Tumour Patients

Abdulhameed Abiola Dere, Muhammad Alghafees

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Artificial intelligence, Gastrointestinal stromal tumors, Genitourinary oncology, Gastrointestinal oncology, Urology, Urologic oncology

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: Evaluation of ML algorithms for predicting genitourinary involvement in gastrointestinal stromal tumor patients at a Saudi Arabian research center found Random Forest achieving a 97.1% accuracy

Abstract:

Gastrointestinal stromal tumors (GISTs) can lead to involvement of other organs, including the genitourinary (GU) system. Machine learning may be a valuable tool in predicting GU involvement in GIST patients, and thus improving prognosis. This study aims to evaluate the use of machine learning algorithms to predict GU involvement among GIST patients in a specialist research center in Saudi Arabia. We analyzed data from all patients with histopathologically confirmed GIST at our facility from 2003 to 2020. Patient files were reviewed for the presence of renal cell carcinoma, adrenal tumors, or other genitourinary cancers. Three supervised machine learning algorithms were used: Logistic Regression, XGBoost Regressor, and Random Forests. A set of variables, including independent attributes, was entered into the models. A total of 170 patients were included in the study, with 58.8% (n=100) being male. The median age was 57 (range 9-91) years. The majority of GISTs were gastric (60%, n=102) with a spindle cell histology. The most common stage at diagnosis was T2 (27.6%, n=47) and N0 (20%, n=34). Six patients (3.5%) had GU involvement. The Random Forest model achieved the highest accuracy with 97.1%. Our study suggests that the Random Forest model is an effective tool for predicting GU involvement in GIST patients. Larger multicenter studies, utilizing more powerful algorithms such as deep learning and other artificial intelligence subsets, are necessary to further refine and improve these predictions.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 6379