Comparison of quantitative parameters and radiomic features as inputs into machine learning models to predict the Gleason score of prostate cancer lesions

Ying-Hwey Nai; Dennis Lai Hong Cheong; Sharmili Roy; Trina Kok; Mary C Stephenson; Josh Schaefferkoetter; John J Totman; Maurizio Conti; Lars Eriksson; Edward G Robins; Ziting Wang; Wynne Yuru Chua; Bertrand Wei Leng Ang; Arvind Kumar Singha; Thomas Paulraj Thamboo; Edmund Chiong; Anthonin Reilhac

doi:10.1016/j.mri.2023.03.009

Comparison of quantitative parameters and radiomic features as inputs into machine learning models to predict the Gleason score of prostate cancer lesions

Magn Reson Imaging. 2023 Jul:100:64-72. doi: 10.1016/j.mri.2023.03.009. Epub 2023 Mar 16.

Authors

Ying-Hwey Nai¹, Dennis Lai Hong Cheong², Sharmili Roy², Trina Kok², Mary C Stephenson³, Josh Schaefferkoetter⁴, John J Totman⁵, Maurizio Conti⁶, Lars Eriksson⁶, Edward G Robins⁷, Ziting Wang⁸, Wynne Yuru Chua⁹, Bertrand Wei Leng Ang⁹, Arvind Kumar Singha⁹, Thomas Paulraj Thamboo¹⁰, Edmund Chiong¹¹, Anthonin Reilhac²

Affiliations

¹ Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore. Electronic address: yinghweynai@yahoo.com.
² Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
³ Centre for Translational MR, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
⁴ Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Siemens Medical Solutions USA, Inc., Molecular Imaging, Knoxville, TN, USA.
⁵ Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Fatima College of Health Sciences, Abu Dhabi, United Arab Emirates.
⁶ Siemens Medical Solutions USA, Inc., Molecular Imaging, Knoxville, TN, USA.
⁷ Clinical Imaging Research Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Singapore BioImaging Consortium, Agency for Science, Technology and Research (A*Star), Singapore.
⁸ Department of Urology, National University Hospital, Singapore.
⁹ Department of Diagnostic Imaging, National University Hospital, Singapore.
¹⁰ Department of Pathology, National University Health System, Singapore.
¹¹ Department of Diagnostic Imaging, National University Hospital, Singapore; Department of Surgery, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.

PMID: 36933775
DOI: 10.1016/j.mri.2023.03.009

Abstract

Introduction: The classification of prostate cancer (PCa) lesions using Prostate Imaging Reporting and Data System (PI-RADS) suffers from poor inter-reader agreement. This study compared quantitative parameters or radiomic features from multiparametric magnetic resonance imaging (mpMRI) or positron emission tomography (PET), as inputs into machine learning (ML) to predict the Gleason scores (GS) of detected lesions for improved PCa lesion classification.

Methods: 20 biopsy-confirmed PCa subjects underwent imaging before radical prostatectomy. A pathologist assigned GS from tumour tissue. Two radiologists and one nuclear medicine physician delineated the lesions on the mpMR and PET images, yielding 45 lesion inputs. Seven quantitative parameters were extracted from the lesions, namely T2-weighted (T2w) image intensity, apparent diffusion coefficient (ADC), transfer constant (K^TRANS), efflux rate constant (K_ep), and extracellular volume ratio (V_e) from mpMR images, and SUV_mean and SUV_max from PET images. Eight radiomic features were selected out of 109 radiomic features from T2w, ADC and PET images. Quantitative parameters or radiomic features, with risk factors of age, prostate-specific antigen (PSA), PSA density and volume, of 45 different lesion inputs were input in different combinations into four ML models - Decision Tree (DT), Support Vector Machine (SVM), k-Nearest-Neighbour (kNN), Ensembles model (EM).

Results: SUV_max yielded the highest accuracy in discriminating detected lesions. Among the 4 ML models, kNN yielded the highest accuracies of 0.929 using either quantitative parameters or radiomic features with risk factors as input.

Conclusions: ML models' performance is dependent on the input combinations and risk factors further improve ML classification accuracy.

Keywords: Gleason score (GS); Machine learning (ML); Multiparametric magnetic resonance imaging (mpMRI); Positron emission tomography (PET); Quantitative parameters; Radiomics.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Humans
Machine Learning
Magnetic Resonance Imaging / methods
Male
Neoplasm Grading
Prostate-Specific Antigen
Prostatic Neoplasms* / pathology
Retrospective Studies

Substances

Prostate-Specific Antigen