EUVIMEDEUVIMED entdecken
Crossref

Impact of harmonization on predicting complications in head and neck cancer after radiotherapy using MRI radiomics and machine learning techniques

Benyamin Khajetash, Ghasem Hajianfar, Amin Talebi, Seid Rabi Mahdavi, Beth Ghavidel, Farshid Arbabi Kalati, Seyed Hadi Molana, Yang Lei, Meysam Tavakoli

Medical Physics · 2025-06 · 52 (6) · 5091-5103

DOI: 10.1002/mp.17793 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical Physics, online erschienen am 31.03.2025. Verfasst von Benyamin Khajetash und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical physics, Iran University of Medical Sciences, Tehran, Iran. Hauptthemen laut MEDLINE: Head and Neck Neoplasms, Magnetic Resonance Imaging, Machine Learning, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 46 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Variations in medical images specific to individual scanners restrict the use of radiomics in both clinical practice and research. To create reproducible and generalizable radiomics‐based models for outcome prediction and assessment, data harmonization is essential.

Purpose

This study aims to investigate the impact of harmonization in performance of machine learning‐based radiomics model toward the prediction of radiotherapy‐induced toxicity (early and late sticky saliva and xerostomia) in head and neck cancer (HNC) patients after radiation therapy using and ‐weighted magnetic resonance (MR) images.

Methods

A total of 85 HNC patients who underwent radiotherapy was studied. Radiomic features were extracted from and ‐weighted MR images with standardized protocols. Data harmonization was performed using ComBat algorithm to reduce inter‐center variability. Besides imaging features, both dosimetric and demographic features were extracted and used in our model. Recursive feature elimination was employed as feature selection method to identify the most important variables. Ten classification algorithms, including eXtreme Gradient Boosting (XGBoost), multilayer perceptron (MLP), support vector machines (SVM), random forest (RF), k‐nearest neighbor (KNN), Naive Bayes (NB), logistic regression (LR), and decision tree (DT), boosted generalized linear model (GLMB), and stack learning (SL) were utilized and compared to develop predictive models. This evaluation comparisons were performed before and after harmonization to demonstrate its significance.

Results

Our results indicate that harmonization consistently enhances predictive performance across various complications and imaging modalities. In early and late sticky saliva prediction using ‐weighted images, the SVM and RF models achieved an impressive area under the curve (AUC) of 0.88 0.09 and 0.97 0.05 with harmonization versus 0.42 0.12 and 0.83 0.08 without harmonization, respectively. Similarly, in early and late xerostomia prediction, the model attained an AUC of 0.79 0.15 and 0.61 0.14 with harmonization and 0.55 0.17 and 0.46 0.14 without harmonization.

Conclusion

Our study highlights the importance of harmonization techniques in improving the performance of predictive models utilizing magnetic resonance imaging radiomics features. While harmonization consistently enhanced performance for sticky saliva and early xerostomia using ‐weighted features, the prediction of early and late xerostomia using ‐weighted features remains challenging. These findings try to develop accurate and reliable predictive models in medical imaging, that contribute to improve patient care and treatment outcomes.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Head and Neck NeoplasmsMagnetic Resonance ImagingMachine LearningImage Processing, Computer-AssistedRadiation InjuriesHumansMaleMiddle AgedFemaleAgedAdultXerostomiaRadiomics

Stichwörter der Autor:innen

harmonizationhead and neck cancerradiomics

Autor:innen und Institutionen

  • Department of Medical physics, Iran University of Medical Sciences, Tehran, Iran.
  • Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Science, Tehran, Iran.
  • Department of Medical Physics, School of Medicine, Iran University of Medical Sciences, Tehran, Iran.
  • Radiation Biology Research Center, Iran University of Medical Sciences, Tehran, Iran.
  • Department of Radiation Oncology, Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.
  • Department of Radiation Oncology, Roshana Cancer Institute, Tehran, Iran.

Zitieren

APA

Khajetash, B., Hajianfar, G., Talebi, A., Mahdavi, S. R., Ghavidel, B., Kalati, F. A., Molana, S. H., Lei, Y., & Tavakoli, M. (2025). Impact of harmonization on predicting complications in head and neck cancer after radiotherapy using MRI radiomics and machine learning techniques. Medical Physics, 52(6), 5091-5103. https://doi.org/10.1002/mp.17793

Vancouver

Khajetash B, Hajianfar G, Talebi A, Mahdavi SR, Ghavidel B, Kalati FA, et al. Impact of harmonization on predicting complications in head and neck cancer after radiotherapy using MRI radiomics and machine learning techniques. Medical Physics. 2025;52(6):5091-5103. doi:10.1002/mp.17793

BibTeX
@article{Khajetash2025,
  author = {Khajetash, Benyamin and Hajianfar, Ghasem and Talebi, Amin and Mahdavi, Seid Rabi and Ghavidel, Beth and Kalati, Farshid Arbabi and Molana, Seyed Hadi and Lei, Yang and Tavakoli, Meysam},
  title = {Impact of harmonization on predicting complications in head and neck cancer after radiotherapy using MRI radiomics and machine learning techniques},
  journal = {Medical Physics},
  year = {2025},
  volume = {52},
  number = {6},
  pages = {5091--5103},
  doi = {10.1002/mp.17793}
}
Kostenlos starten →