The goal of this clinical trial is to evaluate the diagnostic accuracy of the AI-integrated Capsule Gastroscopy (ACG) system in simulated home-use conditions for detecting upper gastrointestinal (UGI) abnormalities. It will also compare the diagnostic accuracy and time efficiency of AI-assisted interpretation versus manual reading of ACG data. The main questions it aims to answer are: What is the diagnostic accuracy of the ACG system, using conventional esophagogastroduodenoscopy (EGD) as a standard of reference? Does AI-assisted ACG reading improve diagnostic accuracy or reduce reading time compared to manual ACG video reading? Researchers will compare ACG results to conventional EGD findings (standard of reference) to determine if ACG can serve as a reliable method for UGI disease detection in home scenarios. Participants will: Undergo an ACG examination in a simulated home environment. Complete an EGD procedure within 24 hours post-ACG ingestion.
Gastroscopy stands as the gold standard for diagnosis of upper gastrointestinal (UGI) diseases (1, 2). However, China faces a significant challenge due to its large population base and an insufficient number of gastrointestinal endoscopists, which limits the widespread adoption of conventional electronic gastroduodenoscopy (EGD). Additionally the discomfort associated with EGD and the low compliance rate reduce its effectiveness in meeting public health needs (3). Capsule endoscopy (CE), primarily comprising magnetic capsule endoscopy (MCE) and powerless capsule endoscopy (PCE), is an innovative medical technology that enables comfortable and comprehensive examinations of the digestive tract (4-6). MCE relies on costly magnetic guidance equipment for capsule manipulation (7), a process that requires specialized technicians. These inherent limitations pose barriers to its clinical use and acceptance in various settings, such as primary hospitals and community health centers. To overcome these challenges, we have developed an automated, wireless, artificial intelligent (AI) integrated Capsule Gastroscopy (ACG) System (GICE-1000, AI Mobile Gastroscopy, Guangzhou Side Medical Technology Co., Ltd) for detecting gastric lesions. After the stomach is distended with ingested, the capsule examines the entire stomach through standardised body position changes by the participant, eliminating the need for any magnetic guidance equipment. The video sequence can be viewed in real-time using a cellphone and transmitted via WIFI to a cloud server for remote reading. Characterized by its comfort, operational simplicity, and remote controllability, GICE-1000 is poised to enhance the early detection and treatment rate of UGI abnormalities in different settings, including both community hospitals and homes, thereby alleviating healthcare system burdens. However, there is no prospective study to assess the diagnostic accuracy of GICE-1000 in home scenarios. The study is structured into three distinct phases: the screening period, the examination period, and the follow-up period. 1. Screening Period: The investigators or research assistants will identify eligible participants, including patients and volunteers in the hospital, and provide a detailed explanation of the study, including relevant information and potential risks. Once a participant signs the informed consent form, the researcher will assign a screening number, record baseline demographic information, and evaluate eligibility based on the study's inclusion and exclusion criteria. Participants who meet the inclusion criteria and do not meet any exclusion criteria will be assigned an enrollment number. 2. Intervention Period: 2.1 Preparation One day prior to the examination, investigators or research assistants will instruct enrolled participants on preparation for the ACG examination. Participants will be advised to fast for 8 hours before the procedure but may drink clear, non-carbonated beverages. 2.2 Procedure of ACG examination 1. Thirty minutes before the examination, participants will report to the hospital's examination room, designed to simulate a home environment. Participants will watch an instructional video on a mobile device and ingest a streptozotocin granule-dimethylsiloxane suspension to clear gastric mucus and foam, improving visualization for the ACG examination. 2. Participants will undergo the ACG examination with video guidance from the mobile app for the ACG device. Following the video instructions, participants will connect the capsule to a mobile device, activate the capsule independently, and ingest it with water to begin the examination. Participants will adjust their body positions as instructed in the video until the examination is complete. Participants are not allowed to view the real-time videos of the ACG examination. During the procedure, researchers will only observe and record whether the participants complete the examination as instructed and note any adverse events. 3. After the ACG examination, the examination data will be uploaded to a cloud server or exported using the device manufacturer's reading system. 2.3 EGD examination 1. Participants should fast for at least 8 hours before the examination with clear water allowed, and take the streptozotocin granule-dimethylsiloxane suspension 30 minutes prior to the procedure. Anesthesia (either local or general) will be administered based on the patient's preference and clinical condition. 2. White-light EGD will be performed by endoscopists (with \> 500 EGD procedures performed) who are blinded to the results of the ACG examination results, within 24 hours after the ingestion of the capsule. The procedure for this study will adhere to the Systematic Screening Protocol for the Stomach (SSS), as outlined by Yao et al(8). This approach includes taking 22 standardized endoscopic photographs to ensure comprehensive visualization of the stomach. Additionally, at least 5 images will be captured from the esophagus, and 2 images will be obtained from the duodenum-one from the first portion (D1) and one from the second portion (D2). This ensures systematic and thorough documentation for diagnostic purposes. At least two images will be captured for each abnormal lesion: one from the top view and one from the side view. This dual imaging approach ensures comprehensive documentation of the lesion's characteristics, aiding in accurate diagnosis and assessment.Any abnormal lesions observed will be documented, including lesion location, morphology, number, type, estimated size etc. Biopsies or treatments will be performed, if necessary, after completion of the white-light EGD examination, following standard clinical practices. 3. Anonymous EGD reports, images, and videos will be exported from the Endoscopy Unit's database. 2.4 ACG reading The ACG video will first be reviewed by a capsule reader (with experience reading \>100 capsules) who is blinded to the EGD results at the center where the patient was enrolled (non-randomized). The initial reading will be conducted in standard mode, according - at 10 frames per s in single-view mode in the small bowel, and 20 frames per s in the oesophagus or stomach. Landmarks, including the first image of the gastrointestinal tract, the first stomach image, the first duodenal image, and representative images of anatomical structures (including esophageal, EGJ, gastric fundus, gastric angle, cardia, body, antrum, pylorus, 1st and 2nd part of the duodenum), were manually selected by the reader. Observed findings were recorded through mouse clicks, ensuring comprehensive documentation of the anatomical and pathological features identified during the video review. This systematic approach allowed for detailed tracking and analysis of abnormalities and structure coverage. The reader should record any abnormalities noted during the ACG video reading, and take at least one representative image of each lesion. The recorded data will include lesion location, morphology, number of lesions, type of lesion, a visual estimation of lesion size, and coverage of gastric anatomical structures (fundus, gastric angle, cardia, body, antrum, pylorus). Additional data will include timestamps captured at the start and end of the reading, image quality, gastric cleanliness, and the degree of gastric filling, among other factors. The ACG video will also be anonymized and randomly allocated to another center for blinded AI-assisted reading, with a reader (with experience reading \>100 capsules) who is unaware of the results from the initial manual reading and the EGD results. The reader will analyze the images and video data provided by the AI platform (Endonet) to make an AI-assisted diagnosis. Following the same documentation protocol as the manual reading, landmarks such as the first images of the gastrointestinal tract, stomach, and duodenum, as well as key anatomical regions (e.g., esophagus, EGJ, fundus, gastric angle, cardia, body, antrum, pylorus, D1 and D2) will be manually identified and captured through mouse clicks if available. Observations, abnormalities, and representative lesion images (at least one image for each lesion) will be recorded, along with lesion characteristics (location, morphology, number of lesions, type of lesion, estimated size), coverage of gastric anatomical structures (fundus, gastric angle, cardia, body, antrum, pylorus), overall image quality, cleanliness, and gastric filling, ensuring thorough documentation and analysis. The study is aim to compare the diagnostic accuracy and reading time of AI-assisted ACG reading with standard manual reading. The documented data will include lesion location, morphology, number of lesions, type of lesion, size, coverage of gastric anatomical structures (fundus, gastric angle, cardia, body, antrum, pylorus), as well as image quality, gastric cleanliness, and degree of gastric filling. Additionally, the number of AI-selected images and AI-assisted ACG reading time will be recorded. The diagnostic outcomes of EGD will serve as the reference standard for assessing the diagnostic accuracy of ACG (including AI-assisted ACG reading). 3\. Study Completion After the completion of all clinical trial cases, the researcher will retain the informed consent forms, review and finalize the case report forms, verify the accuracy and clarity of the data, complete data entry and verification, conduct statistical analysis, and prepare the clinical trial report.
Study Type
INTERVENTIONAL
Allocation
NON_RANDOMIZED
Purpose
DIAGNOSTIC
Masking
NONE
Enrollment
482
Product name: Gastric Capsule Endoscope System Specification model: GICE-1000 Manufacturer: Guangzhou Side Medical Technology Co., Ltd.
EGD was the reference standard against which ACG was compared, and it was performed within 24 hours post-ACG ingestion time. Product name: Electronic gastrointestinal endoscope (EGD) Specifications and models: GIF-290 series Manufacturer: Olympus, Tokyo, Japan
Guangdong Provincial Hospital of Traditional Chinese Medicine
Guangzhou, Guangdong, China
RECRUITINGSecond Affiliated Hospital of Guangzhou Medical University
Guangzhou, Guangdong, China
RECRUITINGNanfang Hospital, Southern Medical University
Guangzhou, Guangdong, China
RECRUITINGRenhua People's Hospital
Shaoguan, Guangdong, China
RECRUITINGLonggang District People's Hospital of Shenzhen
Shenzhen, Guangdong, China
RECRUITINGPingshan People's Hospital
Shenzhen, Guangdong, China
RECRUITINGZhuhai People's Hospital
Zhuhai, Guangdong, China
RECRUITINGWuxi People's Hospital.
Wuxi, Jiangsu, China
RECRUITINGGanzhou People's Hospital
Ganzhou, Jiangxi, China
RECRUITINGMianyang Central Hospital
Mianyang, Sichuan, China
RECRUITING...and 2 more locations
Diagnostic accuracy of ACG in screening for common UGI abnormalities
EGD will serve as the reference standard. Metrics such as sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, diagnostic agreement rate, positive agreement rate (PAR), negative agreement rate (NAR), and diagnostic agreement rate of each individual lesion will be calculated. A participant will be classified as positive if any lesions are detected, including but not limited to: chronic atrophic gastritis, gastric ulcer, gastric polyps, gastric dysplasia, gastric masses, gastric cancer, gastric hemorrhage and other gastric lesions. Esophageal abnormalities (reflux esophagitis, Barrett's esophagus, esophageal cancer, esophageal masses and other abnormalities), duodenal abnormalities (duodenal ulcer, duodenal polyps, hemorrhage and other abnormalities). If no lesions are detected, or only chronic superficial gastritis is observed, the participant will be classified as negative.
Time frame: Through study completion, an average of 1 year.
Diagnostic accuracy and average diagnostic time between AI-assisted reading and manual reading of ACG
First, we will calculate metrics including accuracy, sensitivity, specificity, diagnostic agreement rate, PAR, NAR, PPV, NPV, and per-lesion agreement rate between EGD and AI-assisted ACG reading. Collected data will encompass lesion location, morphology, count, type, and estimated size. The diagnostic performance of AI-assisted reading will be compared with manual reading. Additionally, the average diagnostic time, number of images required for reading for both methods will be analyzed and compared. The capsule reading interface displays the current time. To calculate the diagnostic time, the reader will capture and save an image at both the start and end of each reading session.The average diagnostic time is calculated as follows: Average Diagnostic Time= Total diagnostic time/ Total number of diagnoses
Time frame: Through study completion, an average of 1 year.
Coverage rate of gastric anatomical sites of ACG
The coverage rates of gastric anatomical regions (including the fundus, antrum, cardia, body, sinus, and pylorus) detected by ACG and EGD will be analyzed. Coverage for each anatomical site and overall coverage will be calculated as follows: Coverage rate=Number of Covered Anatomical Sites / Total Number of umber of Anatomical Sites
Time frame: Through study completion, an average of 1 year.
Inspection completion rate of ACG
To assess the proportion of cases that complete ACG examination as required. The inspection completion rate will be calculated as follows: Inspection Completion Rate= Number of Completed Inspections / Total Number of Inspections X 100%
Time frame: Through study completion, an average of 1 year.
Average duration of ACG
The time from the capsule ingestion to expulsion of the capsule during the ACG examination will be recorded for each case, and the average duration will be calculated upon completion of all cases.
Time frame: Through study completion, an average of 1 year.
Average inspection time of ACG
The time from the capsule ingestion to the completion of the gastric examination will be recorded, and the average time will be calculated after all cases are completed.
Time frame: Through study completion, an average of 1 year.
72-hour discharge rate of ACG
This metric refers to the proportion of cases in which the capsule is successfully expelled within 72 hours without any human intervention, relative to the total number of completed examinations. Normal discharge is defined as occurring when the time from swallowing the capsule to natural expulsion is ≤72 hours. If expulsion cannot be confirmed within this timeframe, an abdominal X-ray will be performed on the 7th day, and on the 14th day if necessary, to verify expulsion from the participant. For participants for whom expulsion of the capsule endoscope cannot be confirmed after 72 hours, if the capsule is verified to have been expelled through abdominal X-ray examination on the 4th day, the last defecation time will be recorded as the final expulsion time. The 72-hour discharge rate is calculated as follows: 72-hour Discharge Rate = Number of Discharges within 72 Hours/Total Number of Cases
Time frame: Through study completion, an average of 1 year.
Lesion detection rate of ACG
The proportion of cases with pathological changes among all cases will be assessed. The lesion detection rate of ACG is calculated according to the following formula: Lesion Detection Rate = Number of cases with Lesions Detected/Total Number of Cases
Time frame: Through study completion, an average of 1 year.
Evaluation of image quality of ACG
A three-level grading scale (Excellent/ Good/ Poor) will be used to objectively assess the visibility of the gastric mucosa in six anatomical landmark areas of the stomach. The assessment indicators include: Clarity, Field of view, Imaging, Stability, Structure and Detail Resolution, Ability, Stereoscopic Sense, Overall image impression.
Time frame: Through study completion, an average of 1 year.
Satisfaction evaluation of ACG
Comfort rating Excellent: No obvious discomfort after swallowing, does not affect activities Good: Some discomfort after swallowing and inconvenience in normal activities. Poor: Obvious discomfort after swallowing, affecting normal activities Convenience reviews Excellent: The examination is highly convenient and markedly superior to EGD in terms of ease. Good: The examination is fairly convenient, roughly comparable to EGD. Poor: The examination is inconvenient and noticeably less favorable than EGD in terms of convenience. Operation convenience evaluation Excellent: Simple operation, highly convenient, excellent performance. Good: Relatively easy to operate, convenient, better performance. Poor: Complex operation, very inconvenient, average performance. Comprehensive experience evaluation Excellent: ACG examination provides a better overall experience compared to EGD. Good: ACG examination is comparable to EGD. Poor: ACG examination is noticeably inferior to EGD.
Time frame: Through study completion, an average of 1 year.
Evaluation of image quality of ACG
A three-level grading scale (Excellent/ Good/ Poor) will be used to objectively assess the visibility of the gastric mucosa in six anatomical landmark areas of the stomach. Assessment indicators include Clarity, Field of view, Imaging, Stability, Structure and Detail Resolution, Ability, Stereoscopic Sense, Overall image impression.
Time frame: Through study completion, an average of 1 year.
Evaluation of gastric cleanliness of ACG
A four-level grading scale will be used to objectively assess gastric cleanliness, The grades are classified as Excellent, Good, Moderate, and Poor. Excellent: Satisfactory filling with no obvious gastric fold formation, allowing for complete and clear observation. Good: Gastric fold formation is present, but there is no significant folding; the folds are low and the adjacent spacing is large, which does not affect complete observation. Moderate: A small amount of gastric fold folding is noted, with higher folds and reduced adjacent spacing, affecting the ability to make a complete observation. Poor: A large number of folded gastric folds are present, with high and twisted folds that cause adjacent spacing to disappear, preventing effective observation.
Time frame: Through study completion, an average of 1 year.
Evaluation of gastric filling of ACG
A four-level grading scale will be used to objectively assess gastric cleanliness, filling. The grades are classified as Excellent, Good, Moderate, and Poor. Excellent: A clear field of view, with no obvious mucus, foam, gastric contents, or liquid turbidity present in the stomach, allowing for complete observation. Good: Clear vision with a small amount of mucus, foam, contents, or liquid turbidity present in the stomach, but this does not impede complete observation. Moderate: A blurred field of vision, with increased mucus, foam, or cloudy gastric contents, affecting the ability to make a complete observation. Poor: A severely blurred vision due to a large amount of mucus, foam, contents, or liquid turbidity in the stomach, making observation impossible.
Time frame: Through study completion, an average of 1 year.
Evaluation of degree of gastric mucosal visualization of ACG
A four-level grading scale will be used to objectively assess the visualization of the gastric mucosa. The grades are classified as Excellent, Good, Moderate, and Poor. Excellent: Gastric mucosa is visualized in ≥90% of the target anatomical site. Good: Gastric mucosa is visualized in ≥75% of the target anatomical site. Moderate: Gastric mucosa is visualized in ≥50% of the target anatomical site. Poor: Gastric mucosa is visualized in \<50% of the target anatomical site.
Time frame: Through study completion, an average of 1 year.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.