Exploiting Phase Information from Microwave Imaging for Breast Density Classification: A Deep Learning Approach

View at publisher Summarise with VIRAT
Authors: Mehran Taghipour‐Gorjikolaie, Bilal Khalid, Banafsheh Khalesi, Navid Ghavami, Mario Badia, Mohammad Ghavami, Gianluigi Tiberi
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Dense breast tissue can cause problems mainly in breast cancer detection and risk assessment. It can reduce sensitivity of conventional imaging such as mammogram and may promote tumour development due to higher amounts of fibroglandular tissue. Moreover, it is a big challenge for AI-based models to detect breast cancer in dense breast samples, reducing the accuracy of such models. In this paper, phase information of microwave imaging has been used to develop an optimized deep learning-based model to classify breast density. Phase information is sensitive to variations in tissue composition and permittivity, making it a reliable indicator of breast density. Our obtained results show that a three-layer convolutional model can provide an accuracy of approximately 70% in both high- and low-density breast groups.

Fields of Study & Subjects

Microwave Imaging and Scattering Analysis Digital Radiography and Breast Imaging AI in cancer detection

Publication details

Journal / source
International Conference on Medical Microwave Imaging Applications (ICMWIA)
Pages
1–5
Publication date
3 February 2026
Language
EN