Authors: Chi-en Amy Tai, Alexander Wong

Published on: May 13, 2024

Impact Score: 7.8

Arxiv code: Arxiv:2405.07854

Summary

• What is new: The use of optimized synthetic correlated diffusion imaging (CDI$^s$) combined with diffusion-weighted imaging (DWI) for predicting pathologic complete response in breast cancer.
• Why this is important: Current methods for recommending neoadjuvant chemotherapy for breast cancer are subjective and fraught with biases and uncertainties.
• What the research proposes: Application of optimized CDI$^s$ to enhance pathologic complete response prediction using multiparametric MRI in a noninvasive manner.
• Results: Achieved a leave-one-out cross-validation accuracy of 93.28%, significantly improving by over 5.5% from previous methods.

Technical Details

Technological frameworks used: Multiparametric MRI fusion, Leave-one-out cross-validation

Models used: Volumetric deep radiomic features extraction

Data used: Synthetic correlated diffusion imaging (CDI$^s$), Diffusion-weighted imaging (DWI)

Potential Impact

Healthcare providers, medical imaging companies, pharmaceuticals involved in chemotherapy treatments

Want to implement this idea in a business?

We have generated a startup concept here: RadiomIQ.