Abstract

The healthcare system in developing countries facing many challenges due to factors such as lack of doctors, medical equipment, overwhelmed hospitals, and increased number of refugees. The World Health Organization annually announces reports related to patients per doctor ratios, and according to reports even in many developed countries, it is low. The aim of this work was to develop a medical tool that will try to solve various issues and help assist patients as well as doctors. The tool is based on two machine learning algorithms for disease diagnosis which are rule-based method and decision tree algorithm. The tool also has several useful functionalities that help patients with their conditions. Using scikit-learn framework we were able to develop and integrate algorithms inside the tool. During the benchmarking study, the implemented machine learning algorithms achieved the following performance: an accuracy of 75% for the rule-based classifier, and 89% for the ID3 decision tree classifier.

Authors

Shamiluulu, Shahriar;  Boukar, Moussa Mahamat;  Yussupova, Zulfiya

Publons users who've claimed - I am an author
Contributors on Publons
  • 1 author