Challenges with Machine Learning for Microwave Breast Tumor detection

  • Maged A. Aldhaeebi
  • Saeed M. Bamatraf
  • Omar M. Ramahi

Abstract

In this paper, challenges of combining machine learning techniques with near-field microwave probes for breast tumor detection
is presented. The concept of using microwaves imaging (MI) modality for breast tumors detection is based on the electrical properties
contrast between normal and tumors breast tissues. MI utilized
microwave signals to illuminate the breast tissues using near field
probes placed at different locations surrounding the breast. The
backscattered microwaves signals are then received by the same
probes. Diagnosis breast tumor is done by estimating the variations in the response of the reflection coefficient of the probe. Machine learning techniques are applied to accentuate the variance
in the sensor’s responses for both healthy and tumorous cases.
The main challenge of using the machine learning technique with
near-field microwave probes for breast tumor detection is to find a
suitable combination of features and classifiers which discriminates
between the normal and abnormal breast.

Published
2020-01-02
How to Cite
Aldhaeebi, M., Bamatraf, S., & Ramahi, O. (2020). Challenges with Machine Learning for Microwave Breast Tumor detection. Journal of Computational Vision and Imaging Systems, 5(1), 1. Retrieved from https://openjournals.uwaterloo.ca/index.php/vsl/article/view/1658