Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics
<b>Objectives:</b> To establish and verify radiomics models based on multiparametric MRI for preoperatively identifying the microsatellite instability (MSI) status of rectal cancer (RC) by comparing different machine learning algorithms. <b>Methods:</b> This retrospective stu...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/2/269 |
| _version_ | 1797443816567865344 |
|---|---|
| author | Yang Zhang Jing Liu Cuiyun Wu Jiaxuan Peng Yuguo Wei Sijia Cui |
| author_facet | Yang Zhang Jing Liu Cuiyun Wu Jiaxuan Peng Yuguo Wei Sijia Cui |
| author_sort | Yang Zhang |
| collection | DOAJ |
| description | <b>Objectives:</b> To establish and verify radiomics models based on multiparametric MRI for preoperatively identifying the microsatellite instability (MSI) status of rectal cancer (RC) by comparing different machine learning algorithms. <b>Methods:</b> This retrospective study enrolled 383 (training set, 268; test set, 115) RC patients between January 2017 and June 2022. A total of 4148 radiomics features were extracted from multiparametric MRI, including T<sub>2</sub>-weighted imaging, T<sub>1</sub>-weighted imaging, apparent diffusion coefficient, and contrast-enhanced T<sub>1</sub>-weighted imaging. The analysis of variance, correlation test, univariate logistic analysis, and a gradient-boosting decision tree were used for the dimension reduction. Logistic regression, Bayes, support vector machine (SVM), K-nearest neighbor (KNN), and tree machine learning algorithms were used to build different radiomics models. The relative standard deviation (RSD) and bootstrap method were used to quantify the stability of these five algorithms. Then, predictive performances of different models were assessed using area under curves (AUCs). The performance of the best radiomics model was evaluated using calibration and discrimination. <b>Results:</b> Among these 383 patients, the prevalence of MSI was 14.62% (56/383). The RSD value of logistic regression algorithm was the lowest (4.64%), followed by Bayes (5.44%) and KNN (5.45%), which was significantly better than that of SVM (19.11%) and tree (11.94%) algorithms. The radiomics model based on logistic regression algorithm performed best, with AUCs of 0.827 and 0.739 in the training and test sets, respectively. <b>Conclusions:</b> We developed a radiomics model based on the logistic regression algorithm, which could potentially be used to facilitate the individualized prediction of MSI status in RC patients. |
| first_indexed | 2024-03-09T13:02:29Z |
| format | Article |
| id | doaj.art-2c45931f34f94f67bad31be00a43dd6d |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-09T13:02:29Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-2c45931f34f94f67bad31be00a43dd6d2023-11-30T21:52:22ZengMDPI AGDiagnostics2075-44182023-01-0113226910.3390/diagnostics13020269Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI RadiomicsYang Zhang0Jing Liu1Cuiyun Wu2Jiaxuan Peng3Yuguo Wei4Sijia Cui5Cancer Center, Department of Radiology, Zhejiang Provincial People’s Hospital, Affiliated People’s Hospital, Hangzhou Medical College, Hangzhou 310014, ChinaCancer Center, Department of Radiology, Zhejiang Provincial People’s Hospital, Affiliated People’s Hospital, Hangzhou Medical College, Hangzhou 310014, ChinaCancer Center, Department of Radiology, Zhejiang Provincial People’s Hospital, Affiliated People’s Hospital, Hangzhou Medical College, Hangzhou 310014, ChinaMedical College, Jinzhou Medical University, Jinzhou 121001, ChinaPrecision Health Institution, General Electric Healthcare, Hangzhou 310004, ChinaCancer Center, Department of Radiology, Zhejiang Provincial People’s Hospital, Affiliated People’s Hospital, Hangzhou Medical College, Hangzhou 310014, China<b>Objectives:</b> To establish and verify radiomics models based on multiparametric MRI for preoperatively identifying the microsatellite instability (MSI) status of rectal cancer (RC) by comparing different machine learning algorithms. <b>Methods:</b> This retrospective study enrolled 383 (training set, 268; test set, 115) RC patients between January 2017 and June 2022. A total of 4148 radiomics features were extracted from multiparametric MRI, including T<sub>2</sub>-weighted imaging, T<sub>1</sub>-weighted imaging, apparent diffusion coefficient, and contrast-enhanced T<sub>1</sub>-weighted imaging. The analysis of variance, correlation test, univariate logistic analysis, and a gradient-boosting decision tree were used for the dimension reduction. Logistic regression, Bayes, support vector machine (SVM), K-nearest neighbor (KNN), and tree machine learning algorithms were used to build different radiomics models. The relative standard deviation (RSD) and bootstrap method were used to quantify the stability of these five algorithms. Then, predictive performances of different models were assessed using area under curves (AUCs). The performance of the best radiomics model was evaluated using calibration and discrimination. <b>Results:</b> Among these 383 patients, the prevalence of MSI was 14.62% (56/383). The RSD value of logistic regression algorithm was the lowest (4.64%), followed by Bayes (5.44%) and KNN (5.45%), which was significantly better than that of SVM (19.11%) and tree (11.94%) algorithms. The radiomics model based on logistic regression algorithm performed best, with AUCs of 0.827 and 0.739 in the training and test sets, respectively. <b>Conclusions:</b> We developed a radiomics model based on the logistic regression algorithm, which could potentially be used to facilitate the individualized prediction of MSI status in RC patients.https://www.mdpi.com/2075-4418/13/2/269rectal cancermicrosatellite instabilityalgorithmradiomics |
| spellingShingle | Yang Zhang Jing Liu Cuiyun Wu Jiaxuan Peng Yuguo Wei Sijia Cui Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics Diagnostics rectal cancer microsatellite instability algorithm radiomics |
| title | Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics |
| title_full | Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics |
| title_fullStr | Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics |
| title_full_unstemmed | Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics |
| title_short | Preoperative Prediction of Microsatellite Instability in Rectal Cancer Using Five Machine Learning Algorithms Based on Multiparametric MRI Radiomics |
| title_sort | preoperative prediction of microsatellite instability in rectal cancer using five machine learning algorithms based on multiparametric mri radiomics |
| topic | rectal cancer microsatellite instability algorithm radiomics |
| url | https://www.mdpi.com/2075-4418/13/2/269 |
| work_keys_str_mv | AT yangzhang preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics AT jingliu preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics AT cuiyunwu preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics AT jiaxuanpeng preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics AT yuguowei preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics AT sijiacui preoperativepredictionofmicrosatelliteinstabilityinrectalcancerusingfivemachinelearningalgorithmsbasedonmultiparametricmriradiomics |