Aydin Eresen1,2, Kang Zhou1,3, Chong Sun1,4, Junjie Shangguan1, Bin Wang1,5, Liang Pan1,6, Su Hu1,7, Yongsheng Pang2, Zigeng Zhang2, Robert Minh Nhat Tran8, Ajeet Pal Bhatia2, Farouk Nouizi2,9, Nadine Abi-Jaoudeh2,9, Vahid Yaghmai2,9, Zhuoli Zhang1,2,9,10,11,12. 1. Department of Radiology, Feinberg School of Medicine, Northwestern University Chicago, IL, USA. 2. Department of Radiological Sciences, University of California Irvine Irvine, CA, USA. 3. Department of Radiology, Peking Union Medical College Hospital Beijing 100000, China. 4. Department of Orthopedics, Affiliated Hospital of Qingdao University Qingdao 266000, Shandong, China. 5. Department of General Surgery, Nanfang Hospital, Southern Medical University Guangzhou 510000, Guangdong, China. 6. Department of Radiology, Third Affiliated Hospital of Suzhou University Changzhou 213000, Jiangsu, China. 7. Department of Radiology, First Affiliated Hospital of Soochow University Suzhou 215000, Jiangsu, China. 8. School of Biological Sciences, University of California Irvine Irvine, CA, USA. 9. Chao Family Comprehensive Cancer Center, University of California Irvine Irvine, CA, USA. 10. Robert H. Lurie Comprehensive Cancer Center of Northwestern University Chicago, IL, USA. 11. Department of Biomedical Engineering, University of California Irvine Irvine, CA, USA. 12. Department of Pathology and Laboratory Medicine, University of California Irvine Irvine, CA, USA.
Abstract
OBJECTIVES: Accurate differentiation of temporary vs. permanent changes occurring following irreversible electroporation (IRE) holds immense importance for the early assessment of ablative treatment outcomes. Here, we investigated the benefits of advanced statistical learning models for an immediate evaluation of therapeutic outcomes by interpreting quantitative characteristics captured with conventional MRI. METHODS: The preclinical study integrated twenty-six rabbits with anatomical and perfusion MRI data acquired with a 3T clinical MRI scanner. T1w and T2w MRI data were quantitatively analyzed, and forty-six quantitative features were computed with four feature extraction methods. The candidate key features were determined by graph clustering following the filtering-based feature selection technique, RELIEFF algorithm. Kernel-based support vector machines (SVM) and random forest (RF) classifiers interpreting quantitative features of T1w, T2w, and combination (T1w+T2w) MRI were developed for replicating the underlying characteristics of the tissues to distinguish IRE ablation regions for immediate assessment of treatment response. Accuracy, sensitivity, specificity, and area under the receiver operating characteristics curve were used to evaluate classification performance. RESULTS: Following the analysis of quantitative variables, three features were integrated to develop a SVM classification model, while five features were utilized for generating RF classifiers. SVM classifiers demonstrated detection accuracy of 91.06%, 96.15%, and 98.04% for individual and combination MRI data, respectively. Besides, RF classifiers obtained slightly lower accuracy compared to SVM which were 95.06%, 89.40%, and 94.38% respectively. CONCLUSIONS: Quantitative models integrating structural characteristics of conventional T1w and T2w MRI data with statistical learning techniques identified IRE ablation regions allowing early assessment of treatment status. AJTR
OBJECTIVES: Accurate differentiation of temporary vs. permanent changes occurring following irreversible electroporation (IRE) holds immense importance for the early assessment of ablative treatment outcomes. Here, we investigated the benefits of advanced statistical learning models for an immediate evaluation of therapeutic outcomes by interpreting quantitative characteristics captured with conventional MRI. METHODS: The preclinical study integrated twenty-six rabbits with anatomical and perfusion MRI data acquired with a 3T clinical MRI scanner. T1w and T2w MRI data were quantitatively analyzed, and forty-six quantitative features were computed with four feature extraction methods. The candidate key features were determined by graph clustering following the filtering-based feature selection technique, RELIEFF algorithm. Kernel-based support vector machines (SVM) and random forest (RF) classifiers interpreting quantitative features of T1w, T2w, and combination (T1w+T2w) MRI were developed for replicating the underlying characteristics of the tissues to distinguish IRE ablation regions for immediate assessment of treatment response. Accuracy, sensitivity, specificity, and area under the receiver operating characteristics curve were used to evaluate classification performance. RESULTS: Following the analysis of quantitative variables, three features were integrated to develop a SVM classification model, while five features were utilized for generating RF classifiers. SVM classifiers demonstrated detection accuracy of 91.06%, 96.15%, and 98.04% for individual and combination MRI data, respectively. Besides, RF classifiers obtained slightly lower accuracy compared to SVM which were 95.06%, 89.40%, and 94.38% respectively. CONCLUSIONS: Quantitative models integrating structural characteristics of conventional T1w and T2w MRI data with statistical learning techniques identified IRE ablation regions allowing early assessment of treatment status. AJTR
Authors: Yang Guo; Yue Zhang; Rachel Klein; Grace M Nijm; Alan V Sahakian; Reed A Omary; Guang-Yu Yang; Andrew C Larson Journal: Cancer Res Date: 2010-02-02 Impact factor: 12.701
Authors: Olaguoke Akinwande; Shakeeb S Ahmad; Tracy Van Meter; Brittany Schulz; Robert C G Martin Journal: J Oncol Date: 2015-11-05 Impact factor: 4.375
Authors: Laurien G P H Vroomen; Hester J Scheffer; Marleen C A M Melenhorst; Marcus C de Jong; Janneke E van den Bergh; Cornelis van Kuijk; Foke van Delft; Geert Kazemier; Martijn R Meijerink Journal: Eur Radiol Date: 2016-09-22 Impact factor: 5.315