Scholarlink

With the increase in the number of diagnosed cases of colon cancer and the high associated mortality rate, it becomes imperative to offer a rapid and effective diagnosis in order to improve patient care and optimize the chances of treatment. In this context, with the challenges in the field of oncology, the need of the application of Artificial Intelligence (AI) through techniques such as Machine Learning (ML) to offer solutions that will help doctors in their work by minimizing time and effort, within the framework of what is called precision medicine. This work will present a system for predicting the type of cancer according to the clinical and demographic factors of patients using ML models. The evaluation of five ML models (Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbors (KNNs), XGBoost and Decision Tree Classifier (DTC)) indicated that DTC is the best model. And to facilitate the prediction operation, the model identifier is integrated into a web interface that allows users to fill out forms with the requested factors so that the system gives its prediction on the type of cancer only on these factors without the need to do other medical examinations that take time and delay the treatment time.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based System for Colorectal Cancer Prediction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ibtissam Samnane,
  • Rachid Ed-Daoudi,
  • Samir Bouali,
  • M’barek El Haloui,
  • Badia Ettaki

摘要

With the increase in the number of diagnosed cases of colon cancer and the high associated mortality rate, it becomes imperative to offer a rapid and effective diagnosis in order to improve patient care and optimize the chances of treatment. In this context, with the challenges in the field of oncology, the need of the application of Artificial Intelligence (AI) through techniques such as Machine Learning (ML) to offer solutions that will help doctors in their work by minimizing time and effort, within the framework of what is called precision medicine. This work will present a system for predicting the type of cancer according to the clinical and demographic factors of patients using ML models. The evaluation of five ML models (Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbors (KNNs), XGBoost and Decision Tree Classifier (DTC)) indicated that DTC is the best model. And to facilitate the prediction operation, the model identifier is integrated into a web interface that allows users to fill out forms with the requested factors so that the system gives its prediction on the type of cancer only on these factors without the need to do other medical examinations that take time and delay the treatment time.