TY - JOUR
T1 - Deep learning to distinguish benign from malignant renal lesions based on routine MR imaging
AU - Xi, Ianto Lin
AU - Zhao, Yijun
AU - Wang, Robin
AU - Chang, Marcello
AU - Purkayastha, Subhanik
AU - Chang, Ken
AU - Huang, Raymond Y.
AU - Silva, Alvin C.
AU - Valliéres, Martin
AU - Habibollahi, Peiman
AU - Fan, Yong
AU - Zou, Beiji
AU - Gade, Terence P.
AU - Zhang, Paul J.
AU - Soulen, Michael C.
AU - Zhang, Zishu
AU - Bai, Harrison X.
AU - Stavropoulos, S. William
N1 - Publisher Copyright:
© 2020 American Association for Cancer Research.
PY - 2020/4/15
Y1 - 2020/4/15
N2 - Purpose: With increasing incidence of renal mass, it is important to make a pretreatment differentiation between benign renal mass and malignant tumor. We aimed to develop a deep learning model that distinguishes benign renal tumors from renal cell carcinoma (RCC) by applying a residual convolutional neural network (ResNet) on routine MR imaging. Experimental Design: PreoperativeMRimages (T2-weighted and T1-postcontrast sequences) of 1,162 renal lesions definitely diagnosed on pathology or imaging in a multicenter cohort were divided into training, validation, and test sets (70:20:10 split). An ensemble model based on ResNet was built combining clinical variables and T1C and T2WI MR images using a bagging classifier to predict renal tumor pathology. Final model performance was compared with expert interpretation and the most optimized radiomics model. Results: Among the 1,162 renal lesions, 655 were malignant and 507 were benign. Compared with a baseline zero rule algorithm, the ensemble deep learningmodel had a statistically significant higher test accuracy (0.70 vs. 0.56, P = 0.004). Compared with all experts averaged, the ensemble deep learningmodel had higher test accuracy (0.70 vs. 0.60, P = 0.053), sensitivity (0.92 vs. 0.80, P = 0.017), and specificity (0.41 vs. 0.35, P = 0.450). Compared with the radiomics model, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.62, P = 0.081), sensitivity (0.92 vs. 0.79, P = 0.012), and specificity (0.41 vs. 0.39, P = 0.770). Conclusions: Deep learning can noninvasively distinguish benign renal tumors from RCC using conventional MR imaging in a multiinstitutional dataset with good accuracy, sensitivity, and specificity comparable with experts and radiomics.
AB - Purpose: With increasing incidence of renal mass, it is important to make a pretreatment differentiation between benign renal mass and malignant tumor. We aimed to develop a deep learning model that distinguishes benign renal tumors from renal cell carcinoma (RCC) by applying a residual convolutional neural network (ResNet) on routine MR imaging. Experimental Design: PreoperativeMRimages (T2-weighted and T1-postcontrast sequences) of 1,162 renal lesions definitely diagnosed on pathology or imaging in a multicenter cohort were divided into training, validation, and test sets (70:20:10 split). An ensemble model based on ResNet was built combining clinical variables and T1C and T2WI MR images using a bagging classifier to predict renal tumor pathology. Final model performance was compared with expert interpretation and the most optimized radiomics model. Results: Among the 1,162 renal lesions, 655 were malignant and 507 were benign. Compared with a baseline zero rule algorithm, the ensemble deep learningmodel had a statistically significant higher test accuracy (0.70 vs. 0.56, P = 0.004). Compared with all experts averaged, the ensemble deep learningmodel had higher test accuracy (0.70 vs. 0.60, P = 0.053), sensitivity (0.92 vs. 0.80, P = 0.017), and specificity (0.41 vs. 0.35, P = 0.450). Compared with the radiomics model, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.62, P = 0.081), sensitivity (0.92 vs. 0.79, P = 0.012), and specificity (0.41 vs. 0.39, P = 0.770). Conclusions: Deep learning can noninvasively distinguish benign renal tumors from RCC using conventional MR imaging in a multiinstitutional dataset with good accuracy, sensitivity, and specificity comparable with experts and radiomics.
UR - http://www.scopus.com/inward/record.url?scp=85082407753&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85082407753&partnerID=8YFLogxK
U2 - 10.1158/1078-0432.CCR-19-0374
DO - 10.1158/1078-0432.CCR-19-0374
M3 - Article
C2 - 31937619
AN - SCOPUS:85082407753
SN - 1078-0432
VL - 26
SP - 1944
EP - 1952
JO - Clinical Cancer Research
JF - Clinical Cancer Research
IS - 8
ER -