Building an intelligent system for differential diagnosis of blood cancer through pattern recognition modules
Matvey V. Sedykh1, Julia A. Brodskaya1; 1Gagarin's State Technical University of Saratov, Saratov, Russia
Abstract
The paper presents an intelligent system for the differential diagnosis of blood cancer based on adapted pattern recognition algorithms. The solution uses such methods of pattern recognition as: KNN, naive Bayes, random forest, fully connected neural networks and a logical method implemented by an expert system. The proposed system is of particular value for screening workers with increased occupational risks and possible occupational pathology in hematology in this regard. The features of building hybrid intelligent systems combining machine learning methods and formalized expert knowledge are also considered. The developed system allows you to indicate the clinical signs of the disease in the selected class and receive probable diagnoses. The use of this development as a decision support system can improve the accuracy and reduce the time for disease diagnosis.
Speaker
Matvey Vladimirovich Sedykh
Gagarin's State Technical University of Saratov
Russia
Discussion
Ask question
