Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors
PurposeTo assess the performance of deep neural network (DNN) and machine learning based radiomics on 3D computed tomography (CT) and clinical characteristics to predict benign or malignant sacral tumors.Materials and methodsThis single-center retrospective analysis included 459 patients with pathol...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-10-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fonc.2020.564725/full |
| _version_ | 1828469429876293632 |
|---|---|
| author | Ping Yin Ning Mao Hao Chen Chao Sun Sicong Wang Xia Liu Nan Hong |
| author_facet | Ping Yin Ning Mao Hao Chen Chao Sun Sicong Wang Xia Liu Nan Hong |
| author_sort | Ping Yin |
| collection | DOAJ |
| description | PurposeTo assess the performance of deep neural network (DNN) and machine learning based radiomics on 3D computed tomography (CT) and clinical characteristics to predict benign or malignant sacral tumors.Materials and methodsThis single-center retrospective analysis included 459 patients with pathologically proven sacral tumors. After semi-automatic segmentation, 1,316 hand-crafted radiomics features of each patient were extracted. All models were built on training set (321 patients) and tested on validation set (138 patients). A DNN model and four machine learning classifiers (logistic regression [LR], random forest [RF], support vector machine [SVM] and k-nearest neighbor [KNN]) based on CT features and clinical characteristics were built, respectively. The area under the receiver operating characteristic curve (AUC) and accuracy (ACC) were used to evaluate different models.ResultsIn total, 459 patients (255 males, 204 females; mean age of 42.1 ± 17.8 years, range 4–82 years) were enrolled in this study, including 206 cases of benign tumor and 253 cases of malignant tumor. The sex, age and tumor size had significant differences between the benign tumors and malignant tumors (χ2sex = 10.854, Zage = −6.616, Zsize = 2.843, P < 0.05). The radscore, sex, and age were important indicators for differentiating benign and malignant sacral tumors (odds ratio [OR]1 = 2.492, OR2 = 2.236, OR3 = 1.037, P < 0.01). Among the four clinical-radiomics models (RMs), clinical-LR had the best performance in the validation set (AUC = 0.84, ACC = 0.81). The clinical-DNN model also achieved a high performance (an AUC of 0.83 and an ACC of 0.76 in the validation set) in identifying benign and malignant sacral tumors.ConclusionsBoth the clinical-LR and clinical-DNN models would have a high impact on assisting radiologists in their clinical diagnosis of sacral tumors. |
| first_indexed | 2024-12-11T04:39:33Z |
| format | Article |
| id | doaj.art-043d3820357d4e2980e340da7db82f31 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-12-11T04:39:33Z |
| publishDate | 2020-10-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-043d3820357d4e2980e340da7db82f312022-12-22T01:20:39ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2020-10-011010.3389/fonc.2020.564725564725Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral TumorsPing Yin0Ning Mao1Hao Chen2Chao Sun3Sicong Wang4Xia Liu5Nan Hong6Department of Radiology, Peking University People’s Hospital, Beijing, Beijing Municipality, ChinaDepartment of Radiology, Yantai Yuhuangding Hospital, Qingdao University, Yantai, ChinaDepartment of Radiology, Peking University People’s Hospital, Beijing, Beijing Municipality, ChinaDepartment of Radiology, Peking University People’s Hospital, Beijing, Beijing Municipality, ChinaPharmaceutical Diagnostics, GE Healthcare, Shanghai, ChinaDepartment of Radiology, Peking University People’s Hospital, Beijing, Beijing Municipality, ChinaDepartment of Radiology, Peking University People’s Hospital, Beijing, Beijing Municipality, ChinaPurposeTo assess the performance of deep neural network (DNN) and machine learning based radiomics on 3D computed tomography (CT) and clinical characteristics to predict benign or malignant sacral tumors.Materials and methodsThis single-center retrospective analysis included 459 patients with pathologically proven sacral tumors. After semi-automatic segmentation, 1,316 hand-crafted radiomics features of each patient were extracted. All models were built on training set (321 patients) and tested on validation set (138 patients). A DNN model and four machine learning classifiers (logistic regression [LR], random forest [RF], support vector machine [SVM] and k-nearest neighbor [KNN]) based on CT features and clinical characteristics were built, respectively. The area under the receiver operating characteristic curve (AUC) and accuracy (ACC) were used to evaluate different models.ResultsIn total, 459 patients (255 males, 204 females; mean age of 42.1 ± 17.8 years, range 4–82 years) were enrolled in this study, including 206 cases of benign tumor and 253 cases of malignant tumor. The sex, age and tumor size had significant differences between the benign tumors and malignant tumors (χ2sex = 10.854, Zage = −6.616, Zsize = 2.843, P < 0.05). The radscore, sex, and age were important indicators for differentiating benign and malignant sacral tumors (odds ratio [OR]1 = 2.492, OR2 = 2.236, OR3 = 1.037, P < 0.01). Among the four clinical-radiomics models (RMs), clinical-LR had the best performance in the validation set (AUC = 0.84, ACC = 0.81). The clinical-DNN model also achieved a high performance (an AUC of 0.83 and an ACC of 0.76 in the validation set) in identifying benign and malignant sacral tumors.ConclusionsBoth the clinical-LR and clinical-DNN models would have a high impact on assisting radiologists in their clinical diagnosis of sacral tumors.https://www.frontiersin.org/article/10.3389/fonc.2020.564725/fulldeep learningradiomicssacral tumorsmachine learningcomputed tomography |
| spellingShingle | Ping Yin Ning Mao Hao Chen Chao Sun Sicong Wang Xia Liu Nan Hong Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors Frontiers in Oncology deep learning radiomics sacral tumors machine learning computed tomography |
| title | Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors |
| title_full | Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors |
| title_fullStr | Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors |
| title_full_unstemmed | Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors |
| title_short | Machine and Deep Learning Based Radiomics Models for Preoperative Prediction of Benign and Malignant Sacral Tumors |
| title_sort | machine and deep learning based radiomics models for preoperative prediction of benign and malignant sacral tumors |
| topic | deep learning radiomics sacral tumors machine learning computed tomography |
| url | https://www.frontiersin.org/article/10.3389/fonc.2020.564725/full |
| work_keys_str_mv | AT pingyin machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT ningmao machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT haochen machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT chaosun machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT sicongwang machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT xialiu machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors AT nanhong machineanddeeplearningbasedradiomicsmodelsforpreoperativepredictionofbenignandmalignantsacraltumors |