EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Machine Learning-Based Model for the Prediction of Xerostomia after Head and Neck Radiotherapy

Fazlı Yağız Yedekçi, Melek Tuğçe Yılmaz, Sezin Yüce Sarı, Mustafa Cengiz, Gökhan Özyiğit, Gözde Yazıcı

Advances in Medical Pyhsics and Applied Sciences · 2026

DOI: 10.63187/ampas.44 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Medical Pyhsics and Applied Sciences, 2026. Verfasst von Fazlı Yağız Yedekçi und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A Machine Learning-Based Model for the Prediction of Xerostomia after Head and Neck Radiotherapy Abstract Aim: The aim of this study is to create an ML model to predict xerostomia in patients undergoing RT.

Methods

Data from patients with LAHNC treated in our department between 2012 and 2017 were used in this retrospective study. The patients were selected with an aim of a minimum follow-up period of longer than two years. RIX was assessed with the XI which includes 11 questions. The ML method of random forest (RF) was used for the prediction of RIX. We developed the model using dosimetric parameters as input and RIX scores as output.

Results

The model worked with an accuracy rate of 96.6% to generate the importance of the survey questions on the total RIX score. The question “I have difficulties swallowing certain foods” was the most effectual and “I suck sweets or cough lollies to relieve dry mouth” was the least effectual question, respectively. Also, our ML model predicted the RIX scores with an accuracy of 81%.

Conclusion

The questions in the XI have different impacts on the total score. Our ML model showed that the XI items could be reduced from 11 to nine, costing an only 0.3% decrease in accuracy. The ML model offers to delineate the parotid stem cells and record the doses to these cells for a more accurate assessment of RIX. Keywords: Machine learning, Xerostomia, head and neck radiotherapy

Zitieren

APA

Yedekçi, F. Y., Yılmaz, M. T., Sarı, S. Y., Cengiz, M., Özyiğit, G., & Yazıcı, G. (2026). A Machine Learning-Based Model for the Prediction of Xerostomia after Head and Neck Radiotherapy. Advances in Medical Pyhsics and Applied Sciences. https://doi.org/10.63187/ampas.44

Vancouver

Yedekçi FY, Yılmaz MT, Sarı SY, Cengiz M, Özyiğit G, Yazıcı G. A Machine Learning-Based Model for the Prediction of Xerostomia after Head and Neck Radiotherapy. Advances in Medical Pyhsics and Applied Sciences. 2026. doi:10.63187/ampas.44

BibTeX
@article{Yedeki2026,
  author = {Yedekçi, Fazlı Yağız and Yılmaz, Melek Tuğçe and Sarı, Sezin Yüce and Cengiz, Mustafa and Özyiğit, Gökhan and Yazıcı, Gözde},
  title = {A Machine Learning-Based Model for the Prediction of Xerostomia after Head and Neck Radiotherapy},
  journal = {Advances in Medical Pyhsics and Applied Sciences},
  year = {2026},
  doi = {10.63187/ampas.44}
}
Kostenlos starten →