Jajodia, A.; Gupta, A.; Prosch, H.; Mayerhoefer, M.; Mitra, S.; Pasricha, S.; Mehta, A.; Puri, S.; Chaturvedi, A. Combination of Radiomics and Machine Learning with Diffusion-Weighted MR Imaging for Clinical Outcome Prognostication in Cervical Cancer. Tomography2021, 7, 344-357.
Jajodia, A.; Gupta, A.; Prosch, H.; Mayerhoefer, M.; Mitra, S.; Pasricha, S.; Mehta, A.; Puri, S.; Chaturvedi, A. Combination of Radiomics and Machine Learning with Diffusion-Weighted MR Imaging for Clinical Outcome Prognostication in Cervical Cancer. Tomography 2021, 7, 344-357.
Jajodia, A.; Gupta, A.; Prosch, H.; Mayerhoefer, M.; Mitra, S.; Pasricha, S.; Mehta, A.; Puri, S.; Chaturvedi, A. Combination of Radiomics and Machine Learning with Diffusion-Weighted MR Imaging for Clinical Outcome Prognostication in Cervical Cancer. Tomography2021, 7, 344-357.
Jajodia, A.; Gupta, A.; Prosch, H.; Mayerhoefer, M.; Mitra, S.; Pasricha, S.; Mehta, A.; Puri, S.; Chaturvedi, A. Combination of Radiomics and Machine Learning with Diffusion-Weighted MR Imaging for Clinical Outcome Prognostication in Cervical Cancer. Tomography 2021, 7, 344-357.
Abstract
Objectives: To explore the potential of Radiomics alone and in combination with diffusion-weighted derived quantitative parameter namely apparent diffusion co-efficient (ADC) using supervised classification algorithms in predicting outcomes and prognosis. Materials and Methods: Retrospective evaluation of the imaging was done for a study cohort of uterine cervical cancer, candidates for radical treatment with chemo radiation. ADC values were calculated from the darkest part of the tumor, both before (labeled preADC) and post treatment (labeled postADC) with chemo radiation. Post extraction of 851 Radiomics features and feature selection by taking the union of the features which had Pearson correlation >0.35 for recurrence, >0.49 for lymph node and >0.40 for metastasis, analysis was done to predict clinical outcomes. Results: The study enrolled 52 patients who presented with variable FIGO stages and age range 28–79 (Median = 53 years) with median follow-up of 26.5 months (range, 7–76 months). Disease recurrence occurred in 12 patients (23%). Metastasis occurred in 15 patients (28%). A model generated with 24 radiomics features and preADC using a monotone multi-layer perceptron neural network to predict the recurrence yields AUC of 0.80 and kappa value as 0.55 and shows that addition of radiomics features on ADC values improves the statistical metrics by 40% approximately for AUC and 223% approximately for Kappa. Similarly, neural network model for prediction of metastasis re-turns AUC of 0.84 and kappa value as 0.65 over performs by 25% for AUC and 140% for Kappa approximately. There was a significant input of GLSZM features (SALGLE and LGLZE) and GLDM features (SDLGLE and DE) correlation with clinical outcomes of recurrence and metastasis. Conclusions: The study is an effort to bridge the unmet need of translational predictive biomarkers in stratification of uterine cervical cancer patients based on prognosis.
Keywords
radiomics; diffusion-weighted; MRI; cervical cancer
Subject
Medicine and Pharmacology, Immunology and Allergy
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.