Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Improving Breast Cancer Grade Prediction with Multiparametric MRI Created Using Optimized Synthetic Correlated Diffusion Imaging

Chi-en Amy Tai, Alexander Wong

TL;DR

This work targets noninvasive breast cancer grade prediction, addressing the limitations of biopsy-based grading. It optimizes synthetic correlated diffusion imaging signals, CDI$^s$, via Nelder–Mead to maximize AUC for tumor delineation and fuses the optimized CDI$^s$ with diffusion-weighted imaging to form a multiparametric MRI. An end-to-end deep radiomic pipeline using a MONAI 34-layer volumetric residual CNN processes 224×224×25 brain? volumes (per patient) to predict grades, evaluated with leave-one-out cross-validation. Results show a LOOCV accuracy of 95.79% with optimized CDI$^s$, surpassing 87.70% for the unoptimized approach and achieving high sensitivity and specificity, indicating a promising noninvasive approach for breast cancer grading and highlighting the benefit of domain-specific CDI optimization.

Abstract

Breast cancer was diagnosed for over 7.8 million women between 2015 to 2020. Grading plays a vital role in breast cancer treatment planning. However, the current tumor grading method involves extracting tissue from patients, leading to stress, discomfort, and high medical costs. A recent paper leveraging volumetric deep radiomic features from synthetic correlated diffusion imaging (CDI$^s$) for breast cancer grade prediction showed immense promise for noninvasive methods for grading. Motivated by the impact of CDI$^s$ optimization for prostate cancer delineation, this paper examines using optimized CDI$^s$ to improve breast cancer grade prediction. We fuse the optimized CDI$^s$ signal with diffusion-weighted imaging (DWI) to create a multiparametric MRI for each patient. Using a larger patient cohort and training across all the layers of a pretrained MONAI model, we achieve a leave-one-out cross-validation accuracy of 95.79%, over 8% higher compared to that previously reported.

Improving Breast Cancer Grade Prediction with Multiparametric MRI Created Using Optimized Synthetic Correlated Diffusion Imaging

TL;DR

This work targets noninvasive breast cancer grade prediction, addressing the limitations of biopsy-based grading. It optimizes synthetic correlated diffusion imaging signals, CDI, via Nelder–Mead to maximize AUC for tumor delineation and fuses the optimized CDI with diffusion-weighted imaging to form a multiparametric MRI. An end-to-end deep radiomic pipeline using a MONAI 34-layer volumetric residual CNN processes 224×224×25 brain? volumes (per patient) to predict grades, evaluated with leave-one-out cross-validation. Results show a LOOCV accuracy of 95.79% with optimized CDI, surpassing 87.70% for the unoptimized approach and achieving high sensitivity and specificity, indicating a promising noninvasive approach for breast cancer grading and highlighting the benefit of domain-specific CDI optimization.

Abstract

Breast cancer was diagnosed for over 7.8 million women between 2015 to 2020. Grading plays a vital role in breast cancer treatment planning. However, the current tumor grading method involves extracting tissue from patients, leading to stress, discomfort, and high medical costs. A recent paper leveraging volumetric deep radiomic features from synthetic correlated diffusion imaging (CDI) for breast cancer grade prediction showed immense promise for noninvasive methods for grading. Motivated by the impact of CDI optimization for prostate cancer delineation, this paper examines using optimized CDI to improve breast cancer grade prediction. We fuse the optimized CDI signal with diffusion-weighted imaging (DWI) to create a multiparametric MRI for each patient. Using a larger patient cohort and training across all the layers of a pretrained MONAI model, we achieve a leave-one-out cross-validation accuracy of 95.79%, over 8% higher compared to that previously reported.
Paper Structure (3 sections, 2 figures, 2 tables)

This paper contains 3 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Results

Figures (2)

  • Figure 1: Clinical support workflow copied from tai2023enhancing that is used in this study.
  • Figure 2: An example slice illustrating visual differences between (a) Unoptimized CDIs, (b) Optimized CDIs, (c) the associated DWI, and (d) the associated tumour mask for a patient who has SBR Grade III (High). In this patient case, grade prediction was correct using the Optimized CDIs signal fused with DWI.