Differentiation of gastric schwannomas from gastrointestinal stromal tumors by CT using machine learning
- PDF / 2,032,639 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 41 Downloads / 199 Views
(0123456789().,-volV)(0123456789(). ,- volV)
HOLLOW ORGAN GI
Differentiation of gastric schwannomas from gastrointestinal stromal tumors by CT using machine learning Jian Wang1 • Zongyu Xie2 • Xiandi Zhu1 • Zhongfeng Niu3 • Hongli Ji4 • Linyang He4 • Qiuxiang Hu4 Cui Zhang1
•
Received: 1 August 2020 / Revised: 16 September 2020 / Accepted: 27 September 2020 Ó Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Objective To identify schwannomas from gastrointestinal stromal tumors (GISTs) by CT features using Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), and Gradient Boosting Decision Tree (GBDT). Methods This study enrolled 49 patients with schwannomas and 139 with GISTs proven by pathology. CT features with P \ 0.1 derived from univariate analysis were inputted to four models. Five machine learning (ML) versions, multivariate analysis, and radiologists’ subjective diagnostic performance were compared to evaluate diagnosis performance of all the traditional and advanced methods. Results The CT features with P \ 0.1 were as follows: (1) CT attenuation value of unenhancement phase (CTU), (2) portal venous enhancement (CTV), (3) degree of enhancement in the portal venous phase (DEPP), (4) CT attenuation value of portal venous phase minus arterial phase (CTV-CTA), (5) enhanced potentiality (EP), (6) location, (7) contour, (8) growth pattern, (9) necrosis, (10) surface ulceration, (11) enlarged lymph node (LN). LR (M1), RF, DT, and GBDT models contained all of the above 11 variables, while LR (M2) was developed using six most predictive variables derived from (M1). LR (M2) model with AUC of 0.967 in test dataset was thought to be optimal model in differentiating the two tumors. Location in gastric body, exophytic and mixed growth pattern, lack of necrosis and surface ulceration, enlarged lymph nodes, and larger EP were the most important CT features suggestive of schwannomas. Conclusion LR (M2) provided the optimal diagnostic potency among other ML versions, multivariate analysis, and radiologists’ performance on differentiation of schwannomas from GISTs. Keywords Schwannoma Gastrointestinal stromal tumor Tomography, X-ray computed Machine learning
Introduction Jian Wang and Zongyu Xie have contributed equally to this work.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00261-020-02797-9) contains supplementary material, which is available to authorized users. & Cui Zhang [email protected] Jian Wang [email protected] Zongyu Xie [email protected] Xiandi Zhu [email protected]
Gastric schwannomas, which only account for 1–2% of alimentary tract mesenchymal tumors, are benign neurogenic tumors that originate from Schwann cells of the subepithelial Auerbach’s plexusthe and are S-100 protein– positive spindle cell tumors in histology [1, 2]. Zhongfeng Niu [email protected] Hongli Ji [email protected] Linyang He [email protected] Qiuxiang Hu [email protected] Extended author information available on the last pag
Data Loading...
