Texture Analysis of ADC Maps for Predicting Treatment Response in Locally Advanced Rectal Cancer
Journal of Pioneering Medical Sciences · 05.07.2026 · 15 (6) · 101-109
DOI: 10.47310/jpms2026150614 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Pioneering Medical Sciences, online erschienen am 05.07.2026. Verfasst von Tavri Hasan-Fahmi Rashid und 1 weiteren Autor:innen.
Abstract (Originalsprache)
This study evaluated ADC texture parameters before and after neoadjuvant Chemoradiotherapy (nCRT). It assessed predictive models combining magnetic resonance imaging tumor regression grade with ADC radiomics for the pathological Complete Response (pCR).
Methods
In this prospective single-center study, 34 histologically confirmed LARC patients underwent MRI with DWI/ADC before and after nCRT. First-order texture parameters extracted from ADC maps. Associations between imaging features (TRG, ADC values, texture metrics) and pathological responses were evaluated using ROC analysis.
Results
Patients achieving pCR showed higher baseline mean ADC (1.094 Vs 0.958 ×10⁻³ mm²/s, p<0.001) and greater post-treatment increases (0.288±0.116 Vs 0.156±0.178×10⁻³ mm²/s, p = 0.047). The pCR patients demonstrate greater reductions in skewness (-1.18 Vs -0.41, p = 0.020) and kurtosis (-1.02 Vs -0.60, p = 0.020) and lower post-treatment skewness values (p = 0.007). Excellent accuracy was shown by the MRI TRG alone (AUC = 0.973; sensitivity 100%; specificity 92%). When combining TRG with mean ADC, performance improved (AUC = 0.993); the ADC radiomics model achieved perfect discrimination (AUC = 1.000, with 100% sensitivity and specificity). These perfect metrics should be interpreted cautiously, given the small sample size (n = 34, pCR = 9) and risk of overfitting. The Δ mean ADC threshold ≥ 0.215×10⁻³ mm²/s provided 67% sensitivity and 72% specificity for pCR prediction.
Conclusion
The ADC radiomic features-particularly mean ADC changes and post-treatment skewness-demonstrate high accuracy for predicting pCR in LARC. Limitations include the single-center design and modest cohort size, which require external validation in larger, multicenter studies before clinical translation.
Zitieren
APA
Rashid, T. H., & Rashid, S. A. (2026). Texture Analysis of ADC Maps for Predicting Treatment Response in Locally Advanced Rectal Cancer. Journal of Pioneering Medical Sciences, 15(6), 101-109. https://doi.org/10.47310/jpms2026150614
Vancouver
Rashid TH, Rashid SA. Texture Analysis of ADC Maps for Predicting Treatment Response in Locally Advanced Rectal Cancer. Journal of Pioneering Medical Sciences. 2026;15(6):101-109. doi:10.47310/jpms2026150614
BibTeX
@article{Rashid2026,
author = {Rashid, Tavri Hasan-Fahmi and Rashid, Sameeah Abdulrahman},
title = {Texture Analysis of ADC Maps for Predicting Treatment Response in Locally Advanced Rectal Cancer},
journal = {Journal of Pioneering Medical Sciences},
year = {2026},
volume = {15},
number = {6},
pages = {101--109},
doi = {10.47310/jpms2026150614}
}
EUVIMED