Sajad P Shayesteh1, Afsaneh Alikhassi2, Armaghan Fard Esfahani3, M Miraie4, Parham Geramifar3, Ahmad Bitarafan-Rajabi5, Peiman Haddad6. 1. Department of Physiology, Pharmacology and Medical Physics, Faculty of Medicine, Alborz University of Medical Sciences, Karaj, Iran. 2. Department of Radiology, Cancer Institute of Iran, Tehran University of Medical Sciences, Tehran, Iran. 3. Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran. 4. Cancer Research Centre & Radiation Oncology Department, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran. 5. Cardiovascular Intervention Research Center, Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran; Echocardiography Research Center, Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran. 6. Radiation Oncology Research Center, Cancer Institute, Tehran University of Medical Sciences, Tehran, Iran. Electronic address: haddad@tums.ac.ir.
Abstract
OBJECTIVES: The aim of this study was to investigate and validate the performance of individual and ensemble machine learning models (EMLMs) based on magnetic resonance imaging (MRI) to predict neo-adjuvant chemoradiation therapy (nCRT) response in rectal cancer patients. We also aimed to study the effect of Laplacian of Gaussian (LOG) filter on EMLMs predictive performance. METHODS: 98 rectal cancer patients were divided into a training (n = 53) and a validation set (n = 45). All patients underwent MRI a week before nCRT. Several features from intensity, shape and texture feature sets were extracted from MR images. SVM, Bayesian network, neural network and KNN classifiers were used individually and together for response prediction. Predictive performance was evaluated using the area under the receiver operator characteristic (ROC) curve (AUC). RESULTS: Patients' nCRT responses included 17 patients with Grade 0, 28 with Grade 1, 34 with Grade 2, and 19 with Grade 3 according to AJCC/CAP pathologic grading. In without preprocessing MR Image the best result was for Bayesian network classifier with AUC and accuracy of 75.2% and 80.9% respectively, which was confirmed in the validation set with an AUC and accuracy of 74% and 79% respectively. In EMLMs the best result was for 4 (SVM.NN.BN.KNN) classifier EMLM with AUC and accuracy of 97.8% and 92.8% in testing and 95% and 90% in validation set respectively. CONCLUSIONS: In conclusion, we observed that machine learning methods can used to predict nCRT response in patients with rectal cancer. Preprocessing LOG filters and EL models can improve the prediction process.
OBJECTIVES: The aim of this study was to investigate and validate the performance of individual and ensemble machine learning models (EMLMs) based on magnetic resonance imaging (MRI) to predict neo-adjuvant chemoradiation therapy (nCRT) response in rectal cancerpatients. We also aimed to study the effect of Laplacian of Gaussian (LOG) filter on EMLMs predictive performance. METHODS: 98 rectal cancerpatients were divided into a training (n = 53) and a validation set (n = 45). All patients underwent MRI a week before nCRT. Several features from intensity, shape and texture feature sets were extracted from MR images. SVM, Bayesian network, neural network and KNN classifiers were used individually and together for response prediction. Predictive performance was evaluated using the area under the receiver operator characteristic (ROC) curve (AUC). RESULTS:Patients' nCRT responses included 17 patients with Grade 0, 28 with Grade 1, 34 with Grade 2, and 19 with Grade 3 according to AJCC/CAP pathologic grading. In without preprocessing MR Image the best result was for Bayesian network classifier with AUC and accuracy of 75.2% and 80.9% respectively, which was confirmed in the validation set with an AUC and accuracy of 74% and 79% respectively. In EMLMs the best result was for 4 (SVM.NN.BN.KNN) classifier EMLM with AUC and accuracy of 97.8% and 92.8% in testing and 95% and 90% in validation set respectively. CONCLUSIONS: In conclusion, we observed that machine learning methods can used to predict nCRT response in patients with rectal cancer. Preprocessing LOG filters and EL models can improve the prediction process.
Authors: Niels W Schurink; Simon R van Kranen; Maaike Berbee; Wouter van Elmpt; Frans C H Bakers; Sander Roberti; Joost J M van Griethuysen; Lisa A Min; Max J Lahaye; Monique Maas; Geerard L Beets; Regina G H Beets-Tan; Doenja M J Lambregts Journal: Eur Radiol Date: 2021-02-10 Impact factor: 5.315
Authors: Barbara Juarez Amorim; Angel Torrado-Carvajal; Shadi A Esfahani; Sara S Marcos; Mark Vangel; Dan Stein; David Groshar; Onofrio A Catalano Journal: Mol Imaging Biol Date: 2020-10 Impact factor: 3.488
Authors: Joao Miranda; Gary Xia Vern Tan; Maria Clara Fernandes; Onur Yildirim; John A Sims; Jose de Arimateia Batista Araujo-Filho; Felipe Augusto de M Machado; Antonildes N Assuncao-Jr; Cesar Higa Nomura; Natally Horvat Journal: Clin Imaging Date: 2021-11-16 Impact factor: 2.420
Authors: Laure Fournier; Lena Costaridou; Luc Bidaut; Nicolas Michoux; Frederic E Lecouvet; Lioe-Fee de Geus-Oei; Ronald Boellaard; Daniela E Oprea-Lager; Nancy A Obuchowski; Anna Caroli; Wolfgang G Kunz; Edwin H Oei; James P B O'Connor; Marius E Mayerhoefer; Manuela Franca; Angel Alberich-Bayarri; Christophe M Deroose; Christian Loewe; Rashindra Manniesing; Caroline Caramella; Egesta Lopci; Nathalie Lassau; Anders Persson; Rik Achten; Karen Rosendahl; Olivier Clement; Elmar Kotter; Xavier Golay; Marion Smits; Marc Dewey; Daniel C Sullivan; Aad van der Lugt; Nandita M deSouza Journal: Eur Radiol Date: 2021-01-25 Impact factor: 5.315