Stomach Diseases, Stomach Neoplasms
Conditions
Keywords
Magnifying Endoscopy, Gastroscopy, One-Click Autofocus, Segmented Focus, Image Quality
Brief summary
Magnifying endoscopy provides high-resolution images that enhance the detection of early gastrointestinal lesions. However, conventional manual zoom techniques require frequent focal adjustments, which can be technically demanding and may compromise image stability, especially in complex anatomical settings. This study evaluates a novel one-click autofocus system based on image recognition and computer vision algorithms. The system automatically adjusts focal distance within seconds, eliminating the need for manual operation. This trial aims to compare the efficiency and image quality of one-click autofocus versus segmented manual focus in patients undergoing magnifying endoscopy for gastric diseases. A tandem randomized controlled trial will be conducted to assess procedure time, image clarity score, and operator satisfaction between the two techniques.
Interventions
PROCEDUREOne-Click Autofocus
Magnifying endoscopy examination performed using the one-click autofocus mode. The system automatically adjusts focus to the optimal plane with a single button press.
PROCEDURESegmented Focus
Magnifying endoscopy examination performed using the segmented focus mode. The endoscopist manually adjusts the focus level stepwise through multiple pre-set focal planes.
Sponsors
Shandong University
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
DIAGNOSTIC
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Age ≥ 18 years * Patients with gastric diseases (e.g., suspicious lesions, known gastric disorders) requiring further magnifying endoscopy examination * Ability to understand and provide written informed consent voluntarily
Exclusion criteria
* Absolute contraindications to upper gastrointestinal endoscopy (e.g., severe cardiopulmonary insufficiency, coagulopathy) * Pregnant or lactating women * History of gastric surgery (e.g., gastrectomy) that may affect endoscopic manipulation or anatomy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Image Quality Quantification by Edge Detection Analysis | Day 1 (During the endoscopic examination) | Image quality is objectively quantified using edge detection analysis applied to standardized images captured by each technique. The standard deviation of pixel intensity after edge detection processing is used as the quantitative metric. A higher standard deviation indicates higher image contrast and edge definition, reflecting superior image quality. All image analyses are performed offline by an independent researcher blinded to the technique used. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Subjective Image Quality Assessment by Blinded Expert Review | Day 1 (During the endoscopic examination) | Image quality is subjectively assessed by two to three independent expert reviewers blinded to the technique used. Recorded videos of magnifying endoscopy examinations are reviewed offline and scored using a 5-point Likert scale, where 1 = poor and 5 = excellent. Higher scores indicate superior image quality. The final score for each image is the average of the reviewers' scores. |
| Image Acquisition Time per Capture | Day 1 (During the endoscopic examination) | Time (in seconds) measured for each image acquisition, defined as the duration from the moment the endoscope makes contact with the target gastric mucosa to the capture of the final magnified image for each technique. |
Countries
China
Contacts
CONTACTRui Ji, MD PHD
Outcome results
None listed