Background And Aims: This study aimed to investigate whether AI via a deep learning algorithm using endoscopic ultrasonography (EUS) images could predict the malignant potential of gastric gastrointestinal stromal tumors (GISTs).
Methods: A series of patients who underwent EUS before surgical resection for gastric GISTs were included. A total of 685 images of GISTs from 55 retrospectively included patients were used as the training data set for the AI system. Convolutional neural networks were constructed to build a deep learning model. After applying the synthetic minority oversampling technique, 70% of the generated images were used for AI training and 30% were used to test AI diagnoses. Next, validation was performed using 153 EUS images of 15 patients with GISTs. In addition, conventional EUS features of 55 patients in the training cohort were evaluated to predict the malignant potential of GISTs and mitotic index.
Results: The overall sensitivity, specificity, and accuracy of the AI system for predicting malignancy risk were 83%, 94%, and 82% in the training dataset, and 75%, 73%, and 66% in the validation cohort, respectively. When patients were divided into low-risk and high-risk groups, sensitivity, specificity, and accuracy increased to 99% in the training dataset and 99.7%, 99.7%, and 99.6%, respectively, in the validation cohort. No conventional EUS features were found to be associated with either malignant potential or mitotic index (P > 0.05).
Conclusions: AI via a deep learning algorithm using EUS images could predict the malignant potential of gastric GISTs with high accuracy.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1007/s10620-021-06830-9 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Comparison of different puncture needles used for endoscopic ultrasound-guided fine-needle biopsy of Gastrointestinal subepithelial lesions (≤ 2 cm) with respect to the adequacy of specimen collection: study protocol for a multicenter randomized prospective trial.
Trials
December 2024
Second Department of Internal Medicine, Wakayama Medical University, 811-1, Kimiidera, Wakayama City, 641-0012, Japan.
Background: Gastrointestinal subepithelial lesions (SELs) range from benign to malignant. Endoscopic ultrasound (EUS)-guided fine-needle biopsy (EUS-FNB) is used widely for pathological diagnosis of SELs. Early diagnosis and treatment are important because all Gastrointestinal stromal tumors (GISTs) have some degree of malignant potential.
View Article and Find Full Text PDFNovel insights into the interaction between IGF2BPs and ncRNAs in cancers.
Cancer Cell Int
December 2024
Department of General Surgery, Affiliated Zhongda Hospital of Southeast University, Nanjing, 210009, China.
Insulin-like growth factor II mRNA-binding proteins (IGF2BPs), a family of RNA-binding proteins, are pivotal in regulating RNA dynamics, encompassing processes such as localization, metabolism, stability, and translation through the formation of ribonucleoprotein complexes. First identified in 1999 for their affinity to insulin-like growth factor II mRNA, IGF2BPs have been implicated in promoting tumor malignancy behaviors, including proliferation, metastasis, and the maintenance of stemness, which are associated with unfavorable outcomes in various cancers. Additionally, non-coding RNAs (ncRNAs), particularly long non-coding RNAs, circular RNAs, and microRNAs, play critical roles in cancer progression through intricate protein-RNA interactions.
View Article and Find Full Text PDFEvaluating the adherent perinephric fat risk score in East Asian populations and its correlation with non-alcoholic fatty liver disease.
Nutr Metab Cardiovasc Dis
November 2024
Department of Urology, Longyan First Hospital Affiliated to Fujian Medical University, Longyan, 364000, China.
Background And Aims: This study evaluated the predictive value of the APF risk score in East Asian patients undergoing open nephrectomy and its correlation with hypertension and NAFLD.
Methods And Results: A retrospective study used the clinical data of 82 patients who underwent ON between January 2010 and December 2022. Per their APF score, patients were categorized into groups A (0-2 points) and B (3-4 points).
Transcriptomic data integration and analysis revealing potential mechanisms of doxorubicin resistance in chondrosarcoma cells.
Biochem Pharmacol
December 2024
Department of Medical Laboratory Science and Biotechnology, Asia University, Taichung 41354, Taiwan; Graduate Institute of Biomedical Science, China Medical University, Taichung 40402, Taiwan; Department of Pharmacology, School of Medicine, China Medical University, Taichung 40402, Taiwan; Chinese Medicine Research Center, China Medical University, Taichung 40402, Taiwan; Department of Medical Research, China Medical University Hsinchu Hospital, Hsinchu 30205, Taiwan. Electronic address:
Chondrosarcoma is a type of bone cancer that originates from cartilage cells. In clinical practice, surgical resection is the primary treatment for chondrosarcoma, but chemotherapy becomes essential for patients with metastasis or tumors in surgically inaccessible sites. However, drug resistance often leads to treatment failure.
View Article and Find Full Text PDFPrognostic impact of tumor cell nuclear size assessed by artificial intelligence in esophageal squamous cell carcinoma.
Lab Invest
December 2024
Department of Surgery, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama 359-8513 Japan.
Tumor cell nuclear size (NS) indicates malignant potential in breast cancer; however, its clinical significance in esophageal squamous cell carcinoma (ESCC) is unknown. Artificial intelligence (AI) can quantitatively evaluate histopathological findings. The aim was to measure NS in ESCC using AI and elucidate its clinical significance.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!