Affordable Access

Access to the full text

Machine learning-based radiomic models to predict intensity-modulated radiation therapy response, Gleason score and stage in prostate cancer

Authors
  • Abdollahi, Hamid1
  • Mofid, Bahram2
  • Shiri, Isaac3, 4
  • Razzaghdoust, Abolfazl5
  • Saadipoor, Afshin2
  • Mahdavi, Arash6
  • Galandooz, Hassan Maleki7
  • Mahdavi, Seied Rabi1, 8
  • 1 Iran University of Medical Sciences, Department of Medical Physics, School of Medicine, Tehran, Iran , Tehran (Iran)
  • 2 Shahid Beheshti University of Medical Sciences, Shohada-e-Tajrish Medical Center, Tehran, Iran , Tehran (Iran)
  • 3 Iran University of Medical Sciences, Biomedical and Health Informatics, Rajaie Cardiovascular Medical and Research Center, Tehran, Iran , Tehran (Iran)
  • 4 Tehran University of Medical Sciences, Research Center for Molecular and Cellular Imaging, Tehran, Iran , Tehran (Iran)
  • 5 Shahid Beheshti University of Medical Sciences, Urology and Nephrology Research Center, Student Research Committee, Tehran, Iran , Tehran (Iran)
  • 6 Shahid Beheshti University of Medical Sciences, Department of Radiology, Modarres Hospital, Tehran, Iran , Tehran (Iran)
  • 7 Shahid Beheshti University, Faculty of Computer Science and Engineering, Image Processing and Distributed System Lab, Tehran, Iran , Tehran (Iran)
  • 8 Iran University of Medical Sciences, Radiation Biology Research Center, Tehran, Iran , Tehran (Iran)
Type
Published Article
Journal
La radiologia medica
Publisher
Springer Milan
Publication Date
Jan 03, 2019
Volume
124
Issue
6
Pages
555–567
Identifiers
DOI: 10.1007/s11547-018-0966-4
Source
Springer Nature
Keywords
License
Yellow

Abstract

Objective To develop different radiomic models based on the magnetic resonance imaging (MRI) radiomic features and machine learning methods to predict early intensity-modulated radiation therapy (IMRT) response, Gleason scores (GS) and prostate cancer (Pca) stages.MethodsThirty-three Pca patients were included. All patients underwent pre- and post-IMRT T2-weighted (T2 W) and apparent diffusing coefficient (ADC) MRI. IMRT response was calculated in terms of changes in the ADC value, and patients were divided as responders and non-responders. A wide range of radiomic features from different feature sets were extracted from all T2 W and ADC images. Univariate radiomic analysis was performed to find highly correlated radiomic features with IMRT response, and a paired t test was used to find significant features between responders and non-responders. To find high predictive radiomic models, tenfold cross-validation as the criterion for feature selection and classification was applied on the pre-, post- and delta IMRT radiomic features, and area under the curve (AUC) of receiver operating characteristics was calculated as model performance value.ResultsOf 33 patients, 15 patients (45%) were found as responders. Univariate analysis showed 20 highly correlated radiomic features with IMRT response (20 ADC and 20 T2). Two and fifteen T2 and ADC radiomic features were found as significant (P-value ≤ 0.05) features between responders and non-responders, respectively. Several cross-combined predictive radiomic models were obtained, and post-T2 radiomic models were found as high predictive models (AUC 0.632) followed by pre-ADC (AUC 0.626) and pre-T2 (AUC 0.61). For GS prediction, T2 W radiomic models were found as more predictive (mean AUC 0.739) rather than ADC models (mean AUC 0.70), while for stage prediction, ADC models had higher prediction performance (mean AUC 0.675).ConclusionsRadiomic models developed by MR image features and machine learning approaches are noninvasive and easy methods for personalized prostate cancer diagnosis and therapy.

Report this publication

Statistics

Seen <100 times