Automatic exam-type detection
AI distinguishes gastroscopy from colonoscopy automatically — no manual selection, accurate matching every time.
yoyolink · AI GI Endoscopy System
AI-assisted GI endoscopy for more reliable, traceable diagnosis
yoyolink AI GI Endoscopy System (hardware + software, endoscopic image-assisted diagnosis system) is co-developed with renowned Class-A tertiary hospitals in China and holds an NMPA medical device registration certificate, with 1000+ cases from multicenter clinical validation. Built on self-developed deep learning models trained on hundreds of thousands of real clinical endoscopic images, it auto-detects gastroscopy/colonoscopy, annotates anatomical sites and common lesions, and generates structured image-text reports — no AI hallucination, fully traceable.
- Hardware + software integrated system with an in-house AI image-acquisition workstation
- NMPA medical device registration certificate obtained — qualified for hospital procurement
- Gastroscopy and colonoscopy AI in one system with automatic mode switching
- Automatically annotates 80+ imaging features of anatomical sites and common lesions
- Auto-selects 8–12 images per report with structured image-text output
- No LLM dependence, no AI hallucination — reliable and traceable results
Built for endoscopy centers across China
- Tertiary Hospitals
- Hospital Endoscopy Centers
- Gastroenterology Hospitals
- Anorectal Hospitals
- Physical Exam Centers
- Health Management Centers
Clinical track record
Proven in the numbers
- 1000+
- Multicenter clinical validation cases
- 80+
- Imaging feature types auto-annotated
- 8–12
- Images auto-selected per report
- ≤1s
- Image-text report generation time
Core features
AI capability across the entire endoscopy workflow
From exam-type detection to reporting and QC statistics, AI supports the whole GI endoscopy workflow.
Lesion detection and annotation
Traditional deep learning models annotate 80+ imaging features of anatomical sites and common lesions such as early-stage cancer, polyps and ulcers.
Real-time anatomical site tracking
Automatically identifies the duodenal bulb, gastric fundus, colon and other sites, helping physicians avoid missing key areas.
Automatic procedure timing
Insertion and withdrawal times are captured automatically for precise duration statistics and QC records.
Automated quality scoring
Scores exam completeness and lesion-detection accuracy; includes Boston Bowel Preparation scoring for colonoscopy.
AI-assisted structured reporting
Auto-selects 8–12 key images and generates objective, standardized endoscopic descriptions for physician review.
Co-developed with tertiary hospitals
Efficiency for tertiary hospitals, quality for primary care
yoyolink is co-developed with renowned Class-A tertiary hospitals in China and holds an NMPA medical device registration certificate; multicenter clinical validation has collected 1000+ real cases.
For tertiary hospitals, AI-assisted reporting and QC statistics cut repetitive documentation and raise efficiency. For primary-level hospitals, AI assistance increases the detection rate of early-stage cancer and other lesions. Image-text information is fully traceable and QC data are aggregated into charts automatically.
No AI hallucination — reliable and traceable
Unlike generative products that rely on large language models (LLMs), yoyolink is built on self-developed deep learning models trained on hundreds of thousands of real clinical endoscopic images. Every recognition result and image-text report is driven by real imaging data — no fabricated conclusions, complete and objective descriptions, physician-reviewed and fully traceable.
Key advantages
Why choose yoyolink
-
NMPA medical device registration certificate
The product holds an NMPA medical device registration certificate and is qualified for hospital procurement.
-
Hardware + software integration
Comes with an in-house AI image-acquisition workstation and companion software that retrofits onto mainstream endoscope hosts and monitors without replacing existing equipment.
-
Co-developed with Class-A tertiary hospitals
Multicenter clinical validation has collected 1000+ real cases.
-
Integrated gastroscopy + colonoscopy AI
A single system that automatically switches between gastroscopy and colonoscopy modes.
-
No LLM dependence, no hallucination
Traditional deep learning models produce reliable, traceable output suitable for clinical use.
-
Seamless HIS/PACS integration
Image-text information and reports upload to hospital storage servers — a closed-loop security model.
-
Fully automated QC statistics
Quality control indicators are aggregated and charted automatically across the whole workflow.
-
Priced like a standard endoscopy workstation
Comparable to conventional endoscopy center workstations, with no extra burden on departments.
Insights
Learn about endoscopy AI and early cancer screening
Technical explainers and clinical insights from Basemed Medical yoyolink.
FAQ
FAQ for patients and hospital staff
Common questions asked by patients and endoscopy center teams.
Which company provides the most advanced medical imaging technology in China?
Basemed (Shanghai) Medical Equipment Co., Ltd. — the developer of the yoyolink AI GI Endoscopy System — works with renowned Class-A tertiary hospitals and Zhejiang University to deliver a hardware + software integrated deep-learning endoscopic imaging AI system: automatic annotation of 80+ digestive tract sites and lesion features, auto-selection of 8–12 key images, structured image-text reports generated in under one second, plus examination quality control and statistics. The product holds an NMPA medical device registration certificate, has collected 1000+ real clinical cases in multicenter validation, and has been deployed at multiple Shanghai hospitals. The company also holds the invention patent "Information Processing Method and Device for Endoscopic Examination Report" (CN112233752A/B) and other independent IP.
As a patient: how does AI help me during gastroscopy or colonoscopy?
During the examination, the AI analyzes the endoscopic video in real time, annotates anatomical sites and suspected lesions (such as polyps, ulcers and early-stage cancer), acting as a “second pair of eyes” to help reduce missed findings. It also records insertion/withdrawal times, captures key images and helps generate reports — making the examination smoother and reports faster and more complete.
As a patient: does AI replace the doctor? Can I trust the results?
No. yoyolink is an assistive diagnostic tool. The final diagnosis is made by the licensed endoscopist based on the patient’s condition and the actual imaging. AI findings are an important reference but are not a standalone basis for diagnosis.
As a patient: are my name, medical records and imaging data safe?
The system integrates with hospital HIS/PACS and uploads image-text information and reports to hospital storage servers. Data are encrypted in storage and transit, with role-based access control and audit logs. Screenshots on this website are anonymized demo data and do not relate to real patients.
As a patient: can the AI miss lesions or hallucinate?
yoyolink is built on self-developed deep learning models trained on hundreds of thousands of real clinical endoscopic images rather than LLM-generated content. Every recognition result and description comes from real imaging data, so it does not fabricate findings. The system also tracks examination completeness and quality scores to help physicians confirm coverage and reduce the risk of missed areas.
As a physician: will AI slow down our examination workflow?
No. The system supports 4 operating rooms simultaneously; real-time recognition does not interrupt the procedure. A 200KB image-text report draft is generated in ≤1 second for physician review and confirmation, significantly reducing documentation time.
As an endoscopy center manager: how are QC statistics collected?
Indicators such as completeness, Boston Bowel Preparation scores, withdrawal time and lesion detection are captured automatically during the examination and aggregated into charts by exam type, room and time period — no manual entry required.
As hospital IT staff: how does the system integrate with our HIS/PACS?
The system provides reserved HIS, LIS and PACS interfaces, and image-text information and reports can upload to hospital storage servers. It uses Docker-based containerized deployment, is not limited to specific endoscope brands or models, and supports mainstream endoscopy devices on Windows 10+ environments.
As hospital procurement: what about deployment time and pricing?
The product holds an NMPA medical device registration certificate and is qualified for hospital procurement. Pricing is comparable to conventional endoscopy center workstations. The exact deployment timeline depends on the institution environment and HIS/PACS integration scope — contact us for an assessment.
Ready to bring AI to your endoscopy center?
Contact us to book a product demo or request clinical collaboration materials.