Abstract:Purpose Accurate quantification of gastrointestinal stromal tumors' (GISTs) risk stratification on multicenter endoscopic ultrasound (EUS) images plays a pivotal role in aiding the surgical decision-making process. This study focuses on automatically classifying higher-risk and lower-risk GISTs in the presence of a multicenter setting and limited data. Methods In this study, we retrospectively enrolled 914 patients with GISTs (1824 EUS images in total) from 18 hospitals in China. We propose a triple normalization-based deep learning framework with ultrasound-specific pretraining and meta attention, namely, TN-USMA model. The triple normalization module consists of the intensity normalization, size normalization, and spatial resolution normalization. First, the image intensity is standardized and same-size regions of interest (ROIs) and same-resolution tumor masks are generated in parallel. Then, the transfer learning strategy is utilized to mitigate the data scarcity problem. The same-size ROIs are fed into a deep architecture with ultrasound-specific pretrained weights, which are obtained from self-supervised learning using a large volume of unlabeled ultrasound images. Meanwhile, tumors' size features are calculated from the same-resolution masks individually. Afterward, the size features together with two demographic features are integrated to the model before the final classification layer using a meta attention mechanism to further enhance feature representations. The diagnostic performance of the proposed method was compared with one radiomics-based method and two state-of-the-art deep learning methods. Four evaluation metrics, namely, the accuracy, the area under the receiver operator curve, the sensitivity, and the specificity were used to evaluate the model performance. Results The proposed TN-USMA model achieves an overall accuracy of 0.834 (95% confidence interval [CI]: 0.772, 0.885), an area under the receiver operator curve of 0.881 (95% CI: 0.825, 0.924), a sensitivity of 0.844 (95% CI: 0.672, 0.947), and a specificity of 0.832 (95% CI: 0.762, 0.888). The AUC significantly outperforms other two deep learning approaches (p < 0.05, DeLong et al). Moreover, the performance is stable under different variations of multicenter dataset partitions. Conclusions The proposed TN-USMA model can successfully differentiate higher-risk GISTs from lower-risk ones. It is accurate, robust, generalizable, and efficient for potential clinical applications.

Semi-Supervised Gastrointestinal Stromal Tumor Detection Via Self-Training

Self- and Semi-supervised Learning for Gastroscopic Lesion Detection

Gastrointestinal Stromal Tumors Diagnosis on Multi-Center Endoscopic Ultrasound Images Using Multi-Scale Image Normalization and Transfer Learning

Semi-Supervised Learning with Ensemble Self-Training for Cancer Classification

Pattern Classification for Gastrointestinal Stromal Tumors by Integration of Radiomics and Deep Convolutional Features

Low-Shot Early Gastric Cancer Diagnostic Model Driven by Unsupervised Features

Accurate recognition of colorectal cancer with semi-supervised deep learning on pathological images

TN-USMA Net: Triple Normalization-Based Gastrointestinal Stromal Tumors Classification on Multicenter EUS Images with Ultrasound-Specific Pretraining and Meta Attention.

Deep Learning‐Based Segmentation and Risk Stratification for Gastrointestinal Stromal Tumors in Transabdominal Ultrasound Imaging

A General Global and Local Pre-Training Framework for 3D Medical Image Segmentation.

Self-Paced Co-Training and Foundation Model for Semi-Supervised Medical Image Segmentation

Early gastric cancer detection and lesion segmentation based on deep learning and gastroscopic images

Improving image classification of gastrointestinal endoscopy using curriculum self-supervised learning

Incremental Self-training for Semi-supervised Learning

A new semi-supervised self-training method for lung cancer prediction

Enhanced Three-Dimensional U-Net with Graph-Based Refining for Segmentation of Gastrointestinal Stromal Tumours

Self-supervised learning for gastritis detection with gastric X-ray images

Semi-Supervised Segmentation Framework for Gastrointestinal Lesion Diagnosis in Endoscopic Images

Use of a Convolutional Neural Network to Predict the Malignant Potential of Gastrointestinal Stromal Tumors in Transabdominal Ultrasound Images: Visualization of the Focus of the Prediction Model

Automatic Detection of Early Gastrointestinal Cancer Lesions Based on Optimal Feature Extraction from Gastroscopic Images

GuidedNet: Semi-Supervised Multi-Organ Segmentation via Labeled Data Guide Unlabeled Data