APPLICATION OF NEURAL NETWORKS IN DIAGNOSTICS OF BREAST CANCER ACCORDING TO THE DATA OF MICROWAVE RADIO THERMOMETRY
In recent years, the method of microwave radiothermometry has been increasingly used in the diagnosis of breast cancer. However, the analysis and interpretation of thermometric data is quite a difficult task. In this paper, we propose a method for developing a neural network designed for the diagnos...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
North-Caucasus Federal University
2022-08-01
|
| Series: | Современная наука и инновации |
| Subjects: | |
| Online Access: | https://msi.elpub.ru/jour/article/view/146 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In recent years, the method of microwave radiothermometry has been increasingly used in the diagnosis of breast cancer. However, the analysis and interpretation of thermometric data is quite a difficult task. In this paper, we propose a method for developing a neural network designed for the diagnosis of breast cancer according to microwave radiothermometry. The effectiveness of the combined use of the characteristic space based on medical knowledge and the results of anamnesis is proved. One of the modern and effective methods of functional diagnosis is the method of microwave radiothermometry, which in recent years has found application in a number of areas of medicine, including the diagnosis of breast cancer. However, the analysis of microwave radiothermometry data is a very complex task, which prevents the widespread use of this method in screening. This problem can be solved by creating an effective expert system based on the use of mathematical and computer modeling methods, the capabilities of modern information technologies and, above all, machine learning algorithms. This article is devoted to one of the aspects of this problem. Specifically, the possibilities of using neural networks in the diagnosis of breast cancer based on microwave radiothermometry and history results are discussed. The parameters proposed in this study, as well as the combined use of modeling functions and anamnesis results in the first layer, allowed to significantly increase the efficiency of the classifier in comparison with previous results. |
|---|---|
| ISSN: | 2307-910X |