Magnet Controlled Capsule Endoscopy
Conditions
Keywords
Artificial intelligence, Deep learning
Brief summary
With the rapid development of artificial intelligence technology, more and more deep learning technology has been applied to medicine. Our research is to develop a set of quality control system for magnetic capsule gastroscope using deep learning technology, and conduct a randomized controlled trial to verify its practical efficiency.
Detailed description
The accuracy of magnetic controlled capsule endoscopy(MCCE) in the diagnosis of stomach lesions is highly consistent with the traditional electronic gastroscopy. It has become a new comfortable and safe form for screening and It is also the beneficial complementarity of the traditional electronic gastroscope. To ensure the medical quality of magnetic controlled capsule endoscopy system, Based on Artificial Intelligence Deep Learning Technology ,investigators developed the Magnetic-controlled Capsule Endoscopic Assisted Quality Control System(AI-box).Randomized controlled trials will be conducted on prospective subjects to verify the quality control efficiency.
Interventions
OTHERAI-box
Automatic quality-control system（AI-box）could real-time measure endoscopic inspection completeness, evaluate gastric cleanliness.
Sponsors
Ancon Technologies Ltd
Shandong Provincial Hospital
Binzhou Medical University
Shandong University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SCREENING
Masking
DOUBLE (Subject, Outcomes Assessor)
Masking description
Both participates and outcomes Assesors will be masked
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
No
Inclusion criteria
* Consecutive patients aged 18-75 years old who underwent MCCG examination in Qilu Hospital of Shandong University, Shandong Provincial Hospital affilated to Shandong First Medican University, and the Binzhou Medical University Hospital.
Exclusion criteria
* People who are allergic to the ingredients prepared in the stomach before examination; * Patients with astrointestinal bleeding or ulcers, or prior gastrointestinal bleeding or ulcers within the last 24 months; * Patients with esophageal and gastric tumor diseases that are undergoing active surgical treatment; * Contraindications to the MCE test, including suspected or known gastrointestinal obstruction, stenosis, fistula, diverticula, etc; presence of gastrointestinal obstruction symptoms such as pain or dysphagia; * Prior gastrointestinal tract or abdominal surgery other than simple procedures which would not change the gastrointestinal tract anatomy, such as polyp removal, cholecystectomy or appendectomy; * Inoperative conditions or refusal to undergo abdominal surgery if required (ie, if the capsule will not pass and cannot be removed by endoscopy); * A cardiac pacemaker is installed in the body, except when the pacemaker is a new MRE-compatible product; * Electronic devices such as cochlear implant, magnetic metal drug infusion pump, nerve stimulator and magnetic metal foreign body are implanted in vivo; * Planned MRI examination before capsule endoscopy discharge; * Pregnant women; * Patients with deglutition disorders or gastric emptizing disorders.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Mean inspection completeness | 4 months | After the inspection，investigators will review the whole process of inspection and record the sites observed as well as unobserved sites，then calculate the inspection completeness（number of observed sites in each patient/10）besides the the blind spot rate(number of unobserved sites in each patient/10 ) per procedure in control group and AI-assisted group. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cleanliness judgement consistency between expert and AI-box | 4 months | After the inspection，investigators will review the whole process of inspection and make a judgement of cleanliness of different sites of gastric then compared with the cleanliness judgement of AI-box. So at last investigators will calculate the accuracy of the AI result. |
Countries
China
Outcome results
None listed