Selected work

Representative Papers

DensiThAI, A Multi-View Deep Learning Framework for Breast Density Estimation using Infrared Images

Jan 29, 2026

Current breast density assessment relies on X-ray mammography, which involves ionizing radiation and lacks a safe, radiation-free alternative. This study proposes the first end-to-end deep learning framework based on multi-view infrared thermography to classify breast density without radiation, leveraging differences in surface temperature distributions. The model is trained under supervision using mammographic annotations as the gold standard. Evaluated on a multicenter dataset comprising 3,500 women, the five-view fusion model achieves an average AUROC of 0.73, with statistically significant performance differences across density categories (p ≪ 0.05). Moreover, the model demonstrates consistent performance across age groups, confirming the feasibility and potential of radiation-free breast density assessment through infrared thermography.

0 citationsRead paper

A Density-Informed Multimodal Artificial Intelligence Framework for Improving Breast Cancer Detection Across All Breast Densities

Oct 16, 2025

Mammography exhibits significantly reduced sensitivity in dense breasts, increasing the risk of missed diagnoses. To address this, we propose a breast density–driven multimodal AI framework that adaptively fuses mammography-based AI and thermography-based AI (Thermalytix), dynamically selecting the optimal imaging modality and model pathway based on individual breast density. Methodologically, the framework employs multi-view deep learning for mammographic analysis and introduces vasculo-thermal radiomics to model thermographic data, yielding an interpretable, lightweight decision system. Its key innovation lies in the first use of breast density as the central guiding variable for modality selection and model fusion—balancing high performance, broad applicability, and low-cost deployment. Experimental results demonstrate superior performance: 94.55% sensitivity and 79.93% specificity—significantly outperforming single-modality baselines—while maintaining robust detection rates across both dense and fatty breast types.

0 citationsRead paper

Recent publications

Latest Papers

DensiThAI, A Multi-View Deep Learning Framework for Breast Density Estimation using Infrared Images

Jan 29, 2026

Current breast density assessment relies on X-ray mammography, which involves ionizing radiation and lacks a safe, radiation-free alternative. This study proposes the first end-to-end deep learning framework based on multi-view infrared thermography to classify breast density without radiation, leveraging differences in surface temperature distributions. The model is trained under supervision using mammographic annotations as the gold standard. Evaluated on a multicenter dataset comprising 3,500 women, the five-view fusion model achieves an average AUROC of 0.73, with statistically significant performance differences across density categories (p ≪ 0.05). Moreover, the model demonstrates consistent performance across age groups, confirming the feasibility and potential of radiation-free breast density assessment through infrared thermography.

0 citationsRead paper

A Density-Informed Multimodal Artificial Intelligence Framework for Improving Breast Cancer Detection Across All Breast Densities

Oct 16, 2025

Mammography exhibits significantly reduced sensitivity in dense breasts, increasing the risk of missed diagnoses. To address this, we propose a breast density–driven multimodal AI framework that adaptively fuses mammography-based AI and thermography-based AI (Thermalytix), dynamically selecting the optimal imaging modality and model pathway based on individual breast density. Methodologically, the framework employs multi-view deep learning for mammographic analysis and introduces vasculo-thermal radiomics to model thermographic data, yielding an interpretable, lightweight decision system. Its key innovation lies in the first use of breast density as the central guiding variable for modality selection and model fusion—balancing high performance, broad applicability, and low-cost deployment. Experimental results demonstrate superior performance: 94.55% sensitivity and 79.93% specificity—significantly outperforming single-modality baselines—while maintaining robust detection rates across both dense and fatty breast types.

0 citationsRead paper