Background: Differentiating between ulcerative colitis (UC), Crohn’s disease (CD) and intestinal tuberculosis (ITB) using endoscopy is challenging. We aimed to realize automatic differential diagnosis among these diseases through machine learning algorithms. Methods: A total of 6399 consecutive patients (5128 UC, 875 CD and 396 ITB) who had undergone colonoscopy examinations in the Peking Union Medical College Hospital from January 2008 to November 2018 were enrolled. The input was the description of the endoscopic image in the form of free text. Word segmentation and key word filtering were conducted as data preprocessing. Random forest (RF) and convolutional neural network (CNN) approaches were applied to different disease entities. Three two-class classifiers (UC and CD, UC and ITB, and CD and ITB) and a three-class classifier (UC, CD and ITB) were built. Results: The classifiers built in this research performed well, and the CNN had better performance in general. The RF sensitivities/specificities of UC-CD, UC-ITB, and CD-ITB were 0.89/0.84, 0.83/0.82, and 0.72/0.77, respectively, while the values for the CNN of CD-ITB were 0.90/0.77. The precisions/recalls of UC-CD-ITB when employing RF were 0.97/0.97, 0.65/0.53, and 0.68/0.76, respectively, and when employing the CNN were 0.99/0.97, 0.87/0.83, and 0.52/0.81, respectively. Conclusions: Classifiers built by RF and CNN approaches had excellent performance when classifying UC with CD or ITB. For the differentiation of CD and ITB, high specificity and sensitivity were achieved as well. Artificial intelligence through machine learning is very promising in helping unexperienced endoscopists differentiate inflammatory intestinal diseases. Conference: The abstract of this article has won the first prize of the Young Investigator Award during the Asian Pacific Digestive Week (APDW) 2019 held in Kolkata, India.

Innate lymphoid cells (ILCs) are a group of innate immune cells that have garnered considerable attention due to their critical roles in regulating immunity and tissue homeostasis. They are particularly abundant in the gastrointestinal tract, where they have been shown to interact with commensal bacteria, pathogens, and other components of the local microenvironment to influence host immune responses to infection and oncogenesis. Their tissue-residency properties enable gastric ILCs a localized and rapid response to alert and stress, which indicates their key potential in regulating immunosurveillance. In this review, we discuss the current understanding of the role of ILCs in the gastric mucosa, with a focus on their interactions with the gastric microbiota and Helicobacter pylori and their contributions to tissue homeostasis and inflammation. We also highlight recent findings on the involvement of ILCs in the pathogenesis of gastric cancer and the implications of targeting ILCs as a therapeutic approach. Overall, this review provides an overview of the diverse functions of ILCs in gastric mucosa and highlights their potential as targets for future therapies for gastric cancer.

Gastroblastoma is an extremely rare stomach tumor that primarily presents in adolescent and early adulthood, with a biphasic cell morphology of epithelioid and spindle cells. In light of its similarity to other childhood blastomas, it has been named gastroblastoma. Few patients showed a potential of metastasis and recurrence, however, most of the reported cases were alive, with no evidence of the disease after surgical treatment. Commonly, MALAT1-GLI1 fusion has been considered to be the most relevant mutation. Herein, we present a case of an asymptomatic 58-year-old man who happened to find a submucosal gastric mass during a gastroscope and received endoscopic submucosal excavation (ESE). He turned out to have a gastroblastoma with a novel PTCH1::GLI2 fusion confirmed by Sanger sequencing. The patient was discharged two days after ESE without any complication and was recurrence-free during his one-year follow-up. According to the previous literature and our own experience, in cases with characteristic histopathology and immunohistochemistry patterns, a diagnosis of gastroblastoma should be considered even without a MALAT1-GLI1 fusion. Gastroblastoma pursues a favorable clinical outcome and endoscopic therapy could be an effective alternative treatment choice.

The malignant potential of pancreatic cystic lesions (PCLs) varies dramatically, leading to difficulties when making clinical decisions. This study aimed to develop noninvasive clinical‐radiomic models using preoperative CT images to predict the malignant potential of PCLs. It also investigates the biological mechanisms underlying these models. Patients from two retrospective and one prospective cohort, all undergoing surgical resection for PCLs, are divided into four datasets: training, internal test, external test, and prospective application sets. Eleven machine learning classifiers are employed to construct radiomic models based on selected features. Cyst fluid from the prospective cohort is collected for proteomic and lipidomic analysis. The radiomic models demonstrated high accuracy, with area under the receiver operating characteristic curves (AUCs) > 0.93 across the training (n = 262), internal test (n = 50), and external test (n = 50) sets. AUCs ranged from 0.92 to 0.96 for the prospective cohort (n = 34). Meanwhile, differentially‐expressed proteins and lipid molecules, along with their associated signaling pathways, are identified between high and low groups of clinical‐radiomic scores. This models can effectively and accurately predict the malignant potential of PCLs, with multi‐omics evidence suggesting the biological mechanisms involving secretion function and lipid metabolism underlying clinical‐radiomic models.

Obstructive jaundice is a common digestive disorder with multiple etiologies. Non-tuberculous mycobacterial (NTM) infection is an opportunistic disease that may present with localized pulmonary involvement or disseminated multi-organ manifestations. However, biliary involvement in disseminated NTM infection is rare, and its characteristics and progression remain poorly understood. We report a patient with progressive jaundice who was eventually considered to have probable biliary NTM infection after comprehensive evaluation, exclusion of alternative etiologies, and a favorable therapeutic response despite negative microbiological testing. Endoscopic retrograde cholangiopancreatography (ERCP) was performed to relieve biliary obstruction, but the patient developed recurrent and refractory ampullary bleeding requiring repeated endoscopic interventions. Clinical improvement was achieved following combined antimycobacterial therapy and immunomodulatory treatment. Biliary NTM infection is a rare cause of obstructive jaundice, and ERCP remains necessary for biliary decompression, while post-ERCP bleeding risk should be carefully monitored.

Gastric cancer is a globally prevalent malignancy, with early detection being pivotal for improving patient survival. While endoscopy remains the diagnostic gold standard, it frequently faces challenges such as missed lesions and operator dependency. Artificial intelligence (AI) has emerged as a powerful tool to address these limitations. This narrative review synthesizes recent evidence from PubMed and Web of Science, focusing on four core functional domains of AI-assisted gastric endoscopy: lesion detection and characterization, margin delineation, invasion depth prediction, and blind-spot monitoring. Furthermore, we summarize current limitations, including single-center data biases and algorithmic “black-box” issues, and discuss future directions such as multimodal data integration and real-time video analysis systems. Ultimately, carefully validated AI represents a vital clinical adjunct that holds great potential to significantly enhance diagnostic accuracy and patient outcomes.