Differentiating Gastrointestinal Stromal Tumors from Leiomyomas Using a Neural Network Trained on Endoscopic Ultrasonography Images

Differentiating Gastrointestinal Stromal Tumors From Leiomyomas of Upper Digestive Tract Using Convolutional Neural Network Model by Endoscopic Ultrasonography

BACKGROUND

Gastrointestinal stromal tumors (GISTs) and leiomyomas are the most common submucosal tumors of the upper digestive tract, and the diagnosis of the tumors is essential for their treatment and prognosis. However, the ability of endoscopic ultrasonography (EUS) which could correctly identify the tumor types is limited and closely related to the knowledge, operational level, and experience of the endoscopists. Therefore, the convolutional neural network (CNN) is used to assist endoscopists in determining GISTs or leiomyomas with EUS.

MATERIALS AND METHODS

A model based on CNN was constructed according to GoogLeNet architecture to distinguish GISTs or leiomyomas. All EUS images collected from this study were randomly sampled and divided into training set (n=411) and testing set (n=103) in a ratio of 4:1. The CNN model was trained by EUS images from the training set, and the testing set was utilized to evaluate the performance of the CNN model. In addition, there were some comparisons between endoscopists and CNN models.

RESULTS

It was shown that the sensitivity and specificity in identifying leiomyoma were 95.92%, 94.44%, sensitivity and specificity in identifying GIST were 94.44%, 95.92%, and accuracy in total was 95.15% of the CNN model. It indicates that the diagnostic accuracy of the CNN model is equivalent to skilled endoscopists, or even higher than them.

CONCLUSION

While identifying GIST or leiomyoma, the performance of CNN model was robust, which is highlighting its promising role in supporting less-experienced endoscopists and reducing interobserver agreement.

Current status of artificial intelligence analysis for the treatment of pancreaticobiliary diseases using endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography

Pancreatic and biliary diseases encompass a range of conditions requiring accurate diagnosis for appropriate treatment strategies. This diagnosis relies heavily on imaging techniques like endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography. Artificial intelligence (AI), including machine learning and deep learning, is becoming integral in medical imaging and diagnostics, such as the detection of colorectal polyps. AI shows great potential in diagnosing pancreatobiliary diseases. Unlike machine learning, which requires feature extraction and selection, deep learning can utilize images directly as input. Accurate evaluation of AI performance is a complex task due to varied terminologies, evaluation methods, and development stages. Essential aspects of AI evaluation involve defining the AI's purpose, choosing appropriate gold standards, deciding on the validation phase, and selecting reliable validation methods. AI, particularly deep learning, is increasingly employed in endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography diagnostics, achieving high accuracy levels in detecting and classifying various pancreatobiliary diseases. The AI often performs better than doctors, even in tasks like differentiating benign from malignant pancreatic tumors, cysts, and subepithelial lesions, identifying gallbladder lesions, assessing endoscopic retrograde cholangiopancreatography difficulty, and evaluating the biliary strictures. The potential for AI in diagnosing pancreatobiliary diseases, especially where other modalities have limitations, is considerable. However, a crucial constraint is the need for extensive, high-quality annotated data for AI training. Future advances in AI, such as large language models, promise further applications in the medical field.

Convolutional neural network applied to preoperative venous-phase CT images predicts risk category in patients with gastric gastrointestinal stromal tumors

OBJECTIVE

The risk category of gastric gastrointestinal stromal tumors (GISTs) are closely related to the surgical method, the scope of resection, and the need for preoperative chemotherapy. We aimed to develop and validate convolutional neural network (CNN) models based on preoperative venous-phase CT images to predict the risk category of gastric GISTs.

METHOD

A total of 425 patients pathologically diagnosed with gastric GISTs at the authors' medical centers between January 2012 and July 2021 were split into a training set (154, 84, and 59 with very low/low, intermediate, and high-risk, respectively) and a validation set (67, 35, and 26, respectively). Three CNN models were constructed by obtaining the upper and lower 1, 4, and 7 layers of the maximum tumour mask slice based on venous-phase CT Images and models of CNN_layer3, CNN_layer9, and CNN_layer15 established, respectively. The area under the receiver operating characteristics curve (AUROC) and the Obuchowski index were calculated to compare the diagnostic performance of the CNN models.

RESULTS

In the validation set, CNN_layer3, CNN_layer9, and CNN_layer15 had AUROCs of 0.89, 0.90, and 0.90, respectively, for low-risk gastric GISTs; 0.82, 0.83, and 0.83 for intermediate-risk gastric GISTs; and 0.86, 0.86, and 0.85 for high-risk gastric GISTs. In the validation dataset, CNN_layer3 (Obuchowski index, 0.871) provided similar performance than CNN_layer9 and CNN_layer15 (Obuchowski index, 0.875 and 0.873, respectively) in prediction of the gastric GIST risk category (All P >.05).

CONCLUSIONS

The CNN based on preoperative venous-phase CT images showed good performance for predicting the risk category of gastric GISTs.

Data pyramid structure for optimizing EUS-based GISTs diagnosis in multi-center analysis with missing label

This study introduces the Data Pyramid Structure (DPS) to address data sparsity and missing labels in medical image analysis. The DPS optimizes multi-task learning and enables sustainable expansion of multi-center data analysis. Specifically, It facilitates attribute prediction and malignant tumor diagnosis tasks by implementing a segmentation and aggregation strategy on data with absent attribute labels. To leverage multi-center data, we propose the Unified Ensemble Learning Framework (UELF) and the Unified Federated Learning Framework (UFLF), which incorporate strategies for data transfer and incremental learning in scenarios with missing labels. The proposed method was evaluated on a challenging EUS patient dataset from five centers, achieving promising diagnostic performance. The average accuracy was 0.984 with an AUC of 0.927 for multi-center analysis, surpassing state-of-the-art approaches. The interpretability of the predictions further highlights the potential clinical relevance of our method.

Artificial Intelligence in the Prediction of Gastrointestinal Stromal Tumors on Endoscopic Ultrasonography Images: Development, Validation and Comparison with Endosonographers

Background/Aims

The accuracy of endosonographers in diagnosing gastric subepithelial lesions (SELs) using endoscopic ultrasonography (EUS) is influenced by experience and subjectivity. Artificial intelligence (AI) has achieved remarkable development in this field. This study aimed to develop an AI-based EUS diagnostic model for the diagnosis of SELs, and evaluated its efficacy with external validation.

Methods

We developed the EUS-AI model with ResNeSt50 using EUS images from two hospitals to predict the histopathology of the gastric SELs originating from muscularis propria. The diagnostic performance of the model was also validated using EUS images obtained from four other hospitals.

Results

A total of 2,057 images from 367 patients (375 SELs) were chosen to build the models, and 914 images from 106 patients (108 SELs) were chosen for external validation. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the model for differentiating gastrointestinal stromal tumors (GISTs) and non-GISTs in the external validation sets by images were 82.01%, 68.22%, 86.77%, 59.86%, and 78.12%, respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in the external validation set by tumors were 83.75%, 71.43%, 89.33%, 60.61%, and 80.56%, respectively. The EUS-AI model showed better performance (especially specificity) than some endosonographers. The model helped improve the sensitivity, specificity, and accuracy of certain endosonographers.

Conclusions

We developed an EUS-AI model to classify gastric SELs originating from muscularis propria into GISTs and non-GISTs with good accuracy. The model may help improve the diagnostic performance of endosonographers. Further work is required to develop a multi-modal EUS-AI system.

General Multiscenario Ultrasound Image Tumor Diagnosis Method Based on Unsupervised Domain Adaptation

OBJECTIVE

The utilization of computer-aided diagnosis (CAD) in breast ultrasound image classification has been limited by small sample sizes and domain shift. Current ultrasound classification methods perform inadequately when exposed to cross-domain scenarios, as they struggle with data sets from unobserved domains. In the medical field, there are situations in which all images must share the same networks as they capture the same symptom of the same participant, implying that they share identical structural content. Nevertheless, most domain adaptation methods are not suitable for medical images as they overlook the common features among the images.

METHODS

To overcome these challenges, we propose a novel diverse-domain 2-D feature selection network (FSN), which uses the similarities among medical images and extracts features with a reconstruction network with shared weights. Additionally, it penalizes the feature domain distance through two adversarial learning modules that align the feature space and select common features. Our experiments illustrate that the proposed method is robust and can be applied to ultrasound images of various diseases.

RESULTS

Compared with the latest domain adaptive methods, 2-D FSN markedly enhances the accuracy of classification of breast, thyroid and endoscopic ultrasound images, achieving accuracies of 82.4%, 96.4% and 89.7%, respectively. Furthermore, the model was evaluated on an unsupervised domain adaptation task using ultrasound images from multiple sources and achieved an average accuracy of 77.3% across widely varying domains.

CONCLUSION

In general, 2-D FSN improves the classification ability of the model on multidomain ultrasound data sets through the learning of common features and the combination of multimodule intelligence. The algorithm has good clinical guidance value.

Endoscopic ultrasound artificial intelligence-assisted for prediction of gastrointestinal stromal tumors diagnosis: A systematic review and meta-analysis

BACKGROUND

Subepithelial lesions (SELs) are gastrointestinal tumors with heterogeneous malignant potential. Endoscopic ultrasonography (EUS) is the leading method for evaluation, but without histopathological analysis, precise differentiation of SEL risk is limited. Artificial intelligence (AI) is a promising aid for the diagnosis of gastrointestinal lesions in the absence of histopathology.

AIM

To determine the diagnostic accuracy of AI-assisted EUS in diagnosing SELs, especially lesions originating from the muscularis propria layer.

METHODS

Electronic databases including PubMed, EMBASE, and Cochrane Library were searched. Patients of any sex and > 18 years, with SELs assessed by EUS AI-assisted, with previous histopathological diagnosis, and presented sufficient data values which were extracted to construct a 2 × 2 table. The reference standard was histopathology. The primary outcome was the accuracy of AI for gastrointestinal stromal tumor (GIST). Secondary outcomes were AI-assisted EUS diagnosis for GIST vs gastrointestinal leiomyoma (GIL), the diagnostic performance of experienced endoscopists for GIST, and GIST vs GIL. Pooled sensitivity, specificity, positive, and negative predictive values were calculated. The corresponding summary receiver operating characteristic curve and post-test probability were also analyzed.

RESULTS

Eight retrospective studies with a total of 2355 patients and 44154 images were included in this meta-analysis. The AI-assisted EUS for GIST diagnosis showed a sensitivity of 92% [95% confidence interval (CI): 0.89-0.95; P < 0.01), specificity of 80% (95%CI: 0.75-0.85; P < 0.01), and area under the curve (AUC) of 0.949. For diagnosis of GIST vs GIL by AI-assisted EUS, specificity was 90% (95%CI: 0.88-0.95; P = 0.02) and AUC of 0.966. The experienced endoscopists' values were sensitivity of 72% (95%CI: 0.67-0.76; P < 0.01), specificity of 70% (95%CI: 0.64-0.76; P < 0.01), and AUC of 0.777 for GIST. Evaluating GIST vs GIL, the experts achieved a sensitivity of 73% (95%CI: 0.65-0.80; P < 0.01) and an AUC of 0.819.

CONCLUSION

AI-assisted EUS has high diagnostic accuracy for fourth-layer SELs, especially for GIST, demonstrating superiority compared to experienced endoscopists' and improving their diagnostic performance in the absence of invasive procedures.

Artificial intelligence in endoscopic ultrasonography: risk stratification of gastric gastrointestinal stromal tumors

Background

Previous studies have identified useful endoscopic ultrasonography (EUS) features to predict the malignant potential of gastrointestinal stromal tumors (GISTs). However, the results of the studies were not consistent. Artificial intelligence (AI) has shown promising results in medicine.

Objectives

We aimed to build a risk stratification EUS-AI model to predict the malignancy potential of GISTs.

Design

This was a retrospective study with external validation.

Methods

We developed two models using EUS images from two hospitals to predict the GIST risk category. Model 1 was the four-category risk EUS-AI model, and Model 2 was the two-category risk EUS-AI model. The diagnostic performance of the models was validated with external cohorts.

Results

A total of 1320 images (880 were very low-risk, 269 were low-risk, 68 were intermediate-risk, and 103 were high-risk) were finally chosen for building the models and test sets, and a total of 656 images (211 were very low-risk, 266 were low-risk, 88 were intermediate-risk, and 91 were high-risk) were chosen for external validation. The overall accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the four-category risk EUS-AI model in the external validation sets by tumor were 74.50%, 55.00%, 79.05%, 53.49%, and 81.63%, respectively. The accuracy, sensitivity, specificity, PPV, and NPV for the two-category risk EUS-AI model for the prediction of very low-risk GISTs in the external validation sets by tumor were 86.25%, 94.44%, 79.55%, 79.07%, and 94.59%, respectively.

Conclusion

We developed a EUS-AI model for the risk stratification of GISTs with promising results, which may complement current clinical practice in the management of GISTs.

Registration

The study has been registered in the Chinese Clinical Trial Registry (No. ChiCTR2100051191).

Artificial intelligence-assisted endoscopic ultrasound in the diagnosis of gastrointestinal stromal tumors: a meta-analysis

BACKGROUND AND AIMS

Endoscopic ultrasonography (EUS) is useful for the diagnosis of gastrointestinal stromal tumors (GISTs), but is limited by subjective interpretation. Studies on artificial intelligence (AI)-assisted diagnosis are under development. Here, we used a meta-analysis to evaluate the diagnostic performance of AI in the diagnosis of GISTs using EUS images.

METHODS

PubMed, Ovid Medline, Embase, Web of science, and the Cochrane Library databases were searched for studies based on the EUS using AI for the diagnosis of GISTs, and a meta-analysis was performed to examine the accuracy.

RESULTS

Overall, 7 studies were included in our meta-analysis. A total of 2431 patients containing more than 36,186 images were used as the overall dataset, of which 480 patients were used for the final testing. The pooled sensitivity, specificity, positive, and negative likelihood ratio (LR) of AI-assisted EUS for differentiating GISTs from other submucosal tumors (SMTs) were 0.92 (95% confidence interval [CI] 0.89-0.95), 0.82 (95% CI 0.75-0.87), 4.55 (95% CI 2.64-7.84), and 0.12 (95% CI 0.07-0.20), respectively. The summary diagnostic odds ratio (DOR) and the area under the curve were 64.70 (95% CI 23.83-175.69) and 0.950 (Q* = 0.891).

CONCLUSIONS

AI-assisted EUS showed high accuracy for the automatic endoscopic diagnosis of GISTs, which could be used as a valuable complementary method for the differentiation of SMTs in the future.

Endoscopic Ultrasound Advanced Techniques for Diagnosis of Gastrointestinal Stromal Tumours

Gastrointestinal Stromal Tumors (GISTs) are subepithelial lesions (SELs) that commonly develop in the gastrointestinal tract. GISTs, unlike other SELs, can exhibit malignant behavior, so differential diagnosis is critical to the decision-making process. Endoscopic ultrasound (EUS) is considered the most accurate imaging method for diagnosing and differentiating SELs in the gastrointestinal tract by assessing the lesions precisely and evaluating their malignant risk. Due to their overlapping imaging characteristics, endosonographers may have difficulty distinguishing GISTs from other SELs using conventional EUS alone, and the collection of tissue samples from these lesions may be technically challenging. Even though it appears to be less effective in the case of smaller lesions, histology is now the gold standard for achieving a final diagnosis and avoiding unnecessary and invasive treatment for benign SELs. The use of enhanced EUS modalities and elastography has improved the diagnostic ability of EUS. Furthermore, recent advancements in artificial intelligence systems that use EUS images have allowed them to distinguish GISTs from other SELs, thereby improving their diagnostic accuracy.

Deep learning and radiomics to predict the mitotic index of gastrointestinal stromal tumors based on multiparametric MRI

Introduction

Preoperative evaluation of the mitotic index (MI) of gastrointestinal stromal tumors (GISTs) represents the basis of individualized treatment of patients. However, the accuracy of conventional preoperative imaging methods is limited. The aim of this study was to develop a predictive model based on multiparametric MRI for preoperative MI prediction.

Methods

A total of 112 patients who were pathologically diagnosed with GIST were enrolled in this study. The dataset was subdivided into the development (n = 81) and test (n = 31) sets based on the time of diagnosis. With the use of T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) map, a convolutional neural network (CNN)-based classifier was developed for MI prediction, which used a hybrid approach based on 2D tumor images and radiomics features from 3D tumor shape. The trained model was tested on an internal test set. Then, the hybrid model was comprehensively tested and compared with the conventional ResNet, shape radiomics classifier, and age plus diameter classifier.

Results

The hybrid model showed good MI prediction ability at the image level; the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), and accuracy in the test set were 0.947 (95% confidence interval [CI]: 0.927-0.968), 0.964 (95% CI: 0.930-0.978), and 90.8 (95% CI: 88.0-93.0), respectively. With the average probabilities from multiple samples per patient, good performance was also achieved at the patient level, with AUROC, AUPRC, and accuracy of 0.930 (95% CI: 0.828-1.000), 0.941 (95% CI: 0.792-1.000), and 93.6% (95% CI: 79.3-98.2) in the test set, respectively.

Discussion

The deep learning-based hybrid model demonstrated the potential to be a good tool for the operative and non-invasive prediction of MI in GIST patients.

Efficacy of ultrasound endoscopy with artificial intelligence for the differential diagnosis of non-gastric gastrointestinal stromal tumors

Gastrointestinal stromal tumors (GISTs) are common subepithelial lesions (SELs) and require treatment considering their malignant potential. We recently developed an endoscopic ultrasound-based artificial intelligence (EUS-AI) system to differentiate GISTs from non-GISTs in gastric SELs, which were used to train the system. We assessed whether the EUS-AI system designed for diagnosing gastric GISTs could be applied to non-gastric GISTs. Between January 2015 and January 2021, 52 patients with non-gastric SELs (esophagus, n = 15; duodenum, n = 26; colon, n = 11) were enrolled. The ability of EUS-AI to differentiate GISTs from non-GISTs in non-gastric SELs was examined. The accuracy, sensitivity, and specificity of EUS-AI for discriminating GISTs from non-GISTs in non-gastric SELs were 94.4%, 100%, and 86.1%, respectively, with an area under the curve of 0.98 based on the cutoff value set using the Youden index. In the subanalysis, the accuracy, sensitivity, and specificity of EUS-AI were highest in the esophagus (100%, 100%, 100%; duodenum, 96.2%, 100%, 0%; colon, 90.9%, 100%, 0%); the cutoff values were determined using the Youden index or the value determined using stomach cases. The diagnostic accuracy of EUS-AI increased as lesion size increased, regardless of lesion location. EUS-AI based on gastric SELs had good diagnostic ability for non-gastric GISTs.

Application of artificial intelligence in the diagnosis of subepithelial lesions using endoscopic ultrasonography: a systematic review and meta-analysis

Endoscopic ultrasonography (EUS) is the most common method for diagnosing gastrointestinal subepithelial lesions (SELs); however, it usually requires histopathological confirmation using invasive methods. Artificial intelligence (AI) algorithms have made significant progress in medical imaging diagnosis. The purpose of our research was to explore the application of AI in the diagnosis of SELs using EUS and to evaluate the diagnostic performance of AI-assisted EUS. Three databases, PubMed, EMBASE, and the Cochrane Library, were comprehensively searched for relevant literature. RevMan 5.4.1 and Stata 17.0, were used to calculate and analyze the combined sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and summary receiver-operating characteristic curve (SROC). Eight studies were selected from 380 potentially relevant studies for the meta-analysis of AI-aided EUS diagnosis of SELs. The combined sensitivity, specificity, and DOR of AI-aided EUS were 0.92 (95% CI, 0.85-0.96), 0.80 (95% CI, 0.70-0.87), and 46.27 (95% CI, 19.36-110.59), respectively). The area under the curve (AUC) was 0.92 (95% CI, 0.90-0.94). The AI model in differentiating GIST from leiomyoma had a pooled AUC of 0.95, sensitivity of 0.93, specificity of 0.88, PLR of 8.04, and NLR of 0.08. The combined sensitivity, specificity, and AUC of the AI-aided EUS diagnosis in the convolutional neural network (CNN) model were 0.93, 0.81, and 0.94, respectively. AI-aided EUS diagnosis using conventional brightness mode (B-mode) EUS images had a combined sensitivity of 0.92, specificity of 0.79, and AUC of 0.92. AI-aided EUS diagnosis based on patients had a combined sensitivity, specificity, and AUC of 0.95, 0.83, and 0.96, respectively. Additionally, AI-aided EUS was superior to EUS by experts in terms of sensitivity (0.93 vs. 0.71), specificity (0.81 vs. 0.69), and AUC (0.94 vs. 0.75). In conclusion, AI-assisted EUS is a promising and reliable method for distinguishing SELs, with excellent diagnostic performance. More multicenter cohort and prospective studies are expected to be conducted to further develop AI-assisted real-time diagnostic systems and validate the superiority of AI systems. Systematic Review Registration: PROSPERO (https://www.crd.york.ac.uk/PROSPERO/), identifier CRD42022303990.

Diagnostic accuracy of endoscopic ultrasound with artificial intelligence for gastrointestinal stromal tumors: A meta-analysis

OBJECTIVES

Gastrointestinal stromal tumors (GISTs) are thought to have a malignant potential. However, utilizing endoscopic ultrasound (EUS) to differentiate GISTs from other types of subepithelial lesions (SELs) remains challenging. Artificial intelligence (AI)-based diagnostic systems for EUS have been reported to have a promising performance, although the results of the previous studies remain controversial. In this meta-analysis, we aimed to assess the diagnostic accuracy of AI-based EUS in distinguishing GISTs from other SELs.

METHODS

A literature search was conducted on MEDLINE and EMBASE databases to identify relevant articles. The sensitivity, specificity, and area under the summary receiver operating characteristic curve (AUROC) of eligible studies were analyzed.

RESULTS

Seven studies were eligible for the final analysis. The combined sensitivity and specificity of AI-based EUS were 0.93 (95% confidence interval [CI] 0.88-0.96) and 0.78 (95% CI 0.67-0.87), respectively. The overall diagnostic odds ratio of AI-based EUS for GISTs was 36.74 (95% CI 17.69-76.30) with an AUROC of 0.94.

CONCLUSIONS

AI-based EUS showed high diagnostic ability and might help better differentiate GISTs from other SELs. More prospective studies on the diagnosis of GISTs using AI-based EUS are warranted in clinical setting.