Have a personal or library account? Click to login
Deciphering Breast Cancer Complexity: A Study on the Predictive Power of MRI Texture Analysis for Tumor Characterization and Treatment Response Cover

Deciphering Breast Cancer Complexity: A Study on the Predictive Power of MRI Texture Analysis for Tumor Characterization and Treatment Response

Open Access
|Dec 2025

Figures & Tables

Table 1

Miller‑Payne scoring system for evaluating pathological response to neoadjuvant treatment [17].

GRADEDESCRIPTION
1Minimal or no cellular‑level changes, with unchanged overall cell density
2Up to 30% reduction in tumor cell density
3Reduction in tumor cell density ranging from 30 to 90%
4Greater than 90% decrease in tumor cell density, with cells discernible individually or in small clusters
5Complete absence of malignant cells within the tumor bed
jbsr-109-1-3913-g1.png
Figure 1

Region‑of‑interest (ROI) placement for texture analysis on T1A, T2A, and postcontrast T1A subtraction images of the same patient.

Table 2

Demographic, anatomical, and molecular characteristics of the cohort.

MEANSDCOUNT%
Age55.1913.31
Lesion volume1.150.21
Molecular characteristicsLuminal A1318.6
Luminal B2738.6
HER2‑overexpressed912.9
Triple (−)2130.0
GradeLow3245.7
High3854.3
Ki‑67Ki‑67−3245.7
Ki‑67+3854.3
p53Negative4361.4
Medium positive1217.1
Strong positive1521.4
Cerb‑2Cerb‑B2−5477.1
Cerb‑B2+1622.9
ERER−3042.9
ER+4057.1
PRPR−3448.6
PR+3651.4
Lymphovascular invasionNo4158.6
Yes2941.4
Lymph node metastasisNo3448.6
Yes3651.4
Neoadjuvant responseNo1152.4
Yes1047.6
No neoadjuvant therapy4970.0
Breast patternA811.4
B3245.7
C2028.6
D1014.3
Background contrast enhancement13245.7
21927.1
31724.3
422.9
Lesion locationUpper outer quadrant4057.1
Upper inner quadrant1622.9
Lower outer quadrant811.4
Lower inner quadrant34.3
Retro34.3
ShapeRound912.9
Oval11.4
Irregular6085.7
EdgeSharp34.3
Veiled00.0
Microlobulated3651.4
Indistinct11.4
Spiculated3042.9
SatelliteNo4057.1
yes3042.9
MultifocalNo5071.4
Yes2028.6
MulticentricNo6085.7
Yes1014.3
T1 signalIso4564.3
Hypo11.4
Hyper2434.3
T2 signalIso3651.4
Hypo1217.1
Hyper2231.4
Diffusion restrictionLess than parenchyma00.0
Equal to parenchyma00.0
More than parenchyma70100.0
Contrast patternHomogeneous45.7
Contrast enhancing septa00.0
Heterogeneous5984.3
No contrast enhancement00.0
Peripheral contrast enhancement710.0
Contrast enhancementSlow57.1
Medium34.3
Rapid6288.6
Contrast enhancement curveType 145.7
Type 25071.4
Type 31622.9
jbsr-109-1-3913-g2.png
Figure 2

Heatmap illustrating the distribution of the T1 radiomics feature set among target variables.

jbsr-109-1-3913-g3.png
Figure 3

Heatmap illustrating the distribution of the T2 radiomics feature set among target variables.

jbsr-109-1-3913-g4.png
Figure 4

Heatmap illustrating the distribution of the early T1 radiomics feature set among target variables.

jbsr-109-1-3913-g5.png
Figure 5

Heatmap illustrating the distribution of the late T1 radiomics feature set among target variables.

jbsr-109-1-3913-g6.png
Figure 6

Heatmap illustrating the distribution of the DWI radiomics feature set among target variables.

DOI: https://doi.org/10.5334/jbsr.3913 | Journal eISSN: 2514-8281
Language: English
Submitted on: Mar 2, 2025
|
Accepted on: Nov 20, 2025
|
Published on: Dec 23, 2025
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2025 Hamza Eren Güzel, Alı Murat Koç, Zehra Hılal Adibellı, Funda Taşli, Babak Saravi, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.