Endoscopic ultrasound · Pancreatic lesion evaluation
See the lesion. Clarify its nature.
AI-assisted identification and differentiation of solid pancreatic lesions during live and previously acquired endoscopic ultrasound images.
ENDOSCOPIC ULTRASOUND LIVE AI
LESION2 cm4 cm
AI DIFFERENTIATIONPANCREATIC LESION
P-DAC
P-NET
01 · Intended use
AI assistance for solid pancreatic lesions.
Smart Pancreas® is a deep-learning technology developed for endoscopic ultrasound examinations, supporting identification and differentiation of pancreatic solid lesions.
It works during live procedures and with previously acquired EUS images, helping clinicians select the most relevant target for sampling.
AI
The clinical challenge
Similar ultrasound patterns can conceal meaningful differences.
Pancreatic solid lesions can share overlapping sonographic features. Visual uncertainty may complicate characterization, target selection and timely clinical decisions.
“The critical step is turning subtle EUS patterns into a precise, clinically useful target.”
02
How it works
One continuous path from ultrasound to targeted sampling.
Smart Pancreas® supports the EUS workflow while keeping sampling strategy and final interpretation in expert hands.
EUS acquisition
Live or previously acquired endoscopic ultrasound images enter the analysis workflow.
Lesion localization
The AI detects a solid pancreatic lesion and follows it across ultrasound frames.
Pattern analysis
Deep-learning models assess the lesion’s sonographic texture and structural features.
Lesion differentiation
The finding is characterized to support differentiation between pancreatic lesion types.
Targeted sampling
A precise bounding box supports sampling while the clinician retains final interpretation.
AI capabilities
Designed for focused pancreatic‑lesion assessment.
>98%
Diagnostic accuracy presented by DigestAID
Solid-lesion detection
Automatically identifies pancreatic solid lesions in endoscopic ultrasound images.
Lesion differentiation
Analyzes sonographic patterns to assist differentiation between pancreatic lesion types.
Real-time and retrospective analysis
Works during live EUS and with previously acquired examinations.
Sampling-target guidance
Places bounding boxes around relevant tissue to support precise sampling.
Clinical decision support
Presents detection and differentiation results for expert validation.
03
Scientific validation
Transatlantic validation across pancreatic lesion types.
Deep Learning and Automatic Differentiation of Pancreatic Lesions in Endoscopic Ultrasound: A Transatlantic Study
Clinical and Translational Gastroenterology · 2024 · 378 EUS examinations
Read publication ↗Clinical benefits
Earlier clarity. Better‑targeted sampling.
01
Earlier and more accurate diagnosis of pancreatic solid lesions
↗02
Fewer unnecessary procedures through more focused decision-making
↗03
Target selection for sampling with bounding-box guidance
↗04
Real-time and post-procedural EUS analysis in one workflow
↗05
Reduced need for repeat procedures
↗Smart Pancreas®
See the lesion. Target with confidence.
Talk to our team ↗This technology is part of DigestAID's research and development pipeline. Availability and regulatory status may vary by market.