Radiological and Biological Dictionary of Radiomics Features: Addressing Understandable AI Issues in Personalized Breast Cancer; Dictionary Version BM1.0

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gorji, Arman, Sanati, Nima, Pouria, Amir Hossein, Mehrnia, Somayeh Sadat, Hacihaliloglu, Ilker, Rahmim, Arman, Salmanpour, Mohammad R.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916855514398720
author Gorji, Arman
Sanati, Nima
Pouria, Amir Hossein
Mehrnia, Somayeh Sadat
Hacihaliloglu, Ilker
Rahmim, Arman
Salmanpour, Mohammad R.
author_facet Gorji, Arman
Sanati, Nima
Pouria, Amir Hossein
Mehrnia, Somayeh Sadat
Hacihaliloglu, Ilker
Rahmim, Arman
Salmanpour, Mohammad R.
contents Radiomics-based AI models show promise for breast cancer diagnosis but often lack interpretability, limiting clinical adoption. This study addresses the gap between radiomic features (RF) and the standardized BI-RADS lexicon by proposing a dual-dictionary framework. First, a Clinically-Informed Feature Interpretation Dictionary (CIFID) was created by mapping 56 RFs to BI-RADS descriptors (shape, margin, internal enhancement) through literature and expert review. The framework was applied to classify triple-negative breast cancer (TNBC) versus non-TNBC using dynamic contrast-enhanced MRI from a multi-institutional cohort of 1,549 patients. We trained 27 machine learning classifiers with 27 feature selection methods. SHapley Additive exPlanations (SHAP) were used to interpret predictions and generate a complementary Data-Driven Feature Interpretation Dictionary (DDFID) for 52 additional RFs. The best model, combining Variance Inflation Factor (VIF) selection with Extra Trees Classifier, achieved an average cross-validation accuracy of 0.83. Key predictive RFs aligned with clinical knowledge: higher Sphericity (round/oval shape) and lower Busyness (more homogeneous enhancement) were associated with TNBC. The framework confirmed known imaging biomarkers and uncovered novel, interpretable associations. This dual-dictionary approach (BM1.0) enhances AI model transparency and supports the integration of RFs into routine breast cancer diagnosis and personalized care.
format Preprint
id arxiv_https___arxiv_org_abs_2507_16041
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Radiological and Biological Dictionary of Radiomics Features: Addressing Understandable AI Issues in Personalized Breast Cancer; Dictionary Version BM1.0
Gorji, Arman
Sanati, Nima
Pouria, Amir Hossein
Mehrnia, Somayeh Sadat
Hacihaliloglu, Ilker
Rahmim, Arman
Salmanpour, Mohammad R.
Computational Physics
Machine Learning
F.2.2, I.2.7
Radiomics-based AI models show promise for breast cancer diagnosis but often lack interpretability, limiting clinical adoption. This study addresses the gap between radiomic features (RF) and the standardized BI-RADS lexicon by proposing a dual-dictionary framework. First, a Clinically-Informed Feature Interpretation Dictionary (CIFID) was created by mapping 56 RFs to BI-RADS descriptors (shape, margin, internal enhancement) through literature and expert review. The framework was applied to classify triple-negative breast cancer (TNBC) versus non-TNBC using dynamic contrast-enhanced MRI from a multi-institutional cohort of 1,549 patients. We trained 27 machine learning classifiers with 27 feature selection methods. SHapley Additive exPlanations (SHAP) were used to interpret predictions and generate a complementary Data-Driven Feature Interpretation Dictionary (DDFID) for 52 additional RFs. The best model, combining Variance Inflation Factor (VIF) selection with Extra Trees Classifier, achieved an average cross-validation accuracy of 0.83. Key predictive RFs aligned with clinical knowledge: higher Sphericity (round/oval shape) and lower Busyness (more homogeneous enhancement) were associated with TNBC. The framework confirmed known imaging biomarkers and uncovered novel, interpretable associations. This dual-dictionary approach (BM1.0) enhances AI model transparency and supports the integration of RFs into routine breast cancer diagnosis and personalized care.
title Radiological and Biological Dictionary of Radiomics Features: Addressing Understandable AI Issues in Personalized Breast Cancer; Dictionary Version BM1.0
topic Computational Physics
Machine Learning
F.2.2, I.2.7
url https://arxiv.org/abs/2507.16041