Mauna Kea Technologies confirms 96% diagnostic accuracy in two new studies
Mauna Kea Technologies reported two peer-reviewed studies showing its Cellvizio platform achieves 96% accuracy in diagnosing pancreatic cysts and 87% accuracy in detecting high-risk lesions using AI. The findings highlight reduced unnecessary surgeries and improved detection rates compared to conventional methods.

*this image is generated using AI for illustrative purposes only.
Mauna Kea Technologies (Euronext Growth: ALMKT) has announced two new peer-reviewed clinical studies that reinforce the diagnostic value of its Cellvizio® probe and needle-based confocal laser endomicroscopy (p/nCLE) platform for pancreatic cystic lesions (PCLs). The findings demonstrate significant improvements in diagnostic accuracy and physician agreement, addressing critical challenges in distinguishing mucinous from non-mucinous cysts and detecting high-grade dysplasia. These results support the company’s strategy to scale adoption across U.S. centers and integrate artificial intelligence into its diagnostic workflow.
The first study, published in Gastrointestinal Endoscopy, was an international interobserver assessment involving 15 independent physicians of varying experience levels. The research evaluated the reproducibility of pancreatic cyst diagnostics when Cellvizio imaging was added to standard evaluation protocols. The addition of real-time cellular imaging increased the accuracy for distinguishing mucinous versus non-mucinous cysts to 96%. Furthermore, subtype-specific accuracy rose to approximately 85%, and agreement among the independent observers improved significantly. This reproducibility across clinicians is cited as a key factor for broad clinical adoption.
The second study, published in Techniques and Innovations in Gastrointestinal Endoscopy, focused on an nCLE-guided artificial intelligence algorithm designed to detect high-grade dysplasia or early cancer within intraductal papillary mucinous neoplasms (IPMNs). In higher-risk, non-gastric IPMN subtypes, the AI algorithm achieved 87% accuracy, a substantial increase from the 53% accuracy rate of conventional methods. The study noted comparable sensitivity across subtypes, providing early evidence supporting the company’s roadmap to deploy AI-assisted decision support tools on the Cellvizio platform.
Clinical Impact and Expert Commentary
Dr. Somashekar (Som) Krishna, Professor of Medicine and Director of Advanced Endoscopy at The Ohio State University Wexner Medical Center, highlighted the clinical stakes of accurate diagnosis. He noted that in pancreatic cyst diagnosis, errors can lead to either unnecessary surgery for benign cysts or missed cancers. In the ongoing multicenter CLIMB study evaluating EUS-nCLE, adding real-time EUS-nCLE to standard evaluation raised diagnostic accuracy from 73% to 85%. This improvement reduced unnecessary surgery for benign cysts by more than 40% and cut missed cancers roughly six-fold. Dr. Krishna emphasized that these gains were produced by 15 international endosonographers new to the technique, demonstrating consistent performance across users.
Sacha Loiseau, Ph.D., Chairman and Chief Executive Officer of Mauna Kea Technologies, stated that the growing body of independent, peer-reviewed evidence confirms the unique value of Cellvizio in providing a window into pancreatic cysts that no other technology offers. He added that avoiding unnecessary surgeries and detecting cancers earlier strengthens the company’s conviction as it scales adoption and brings AI-assisted diagnosis to the platform.
What the Numbers Show
The data reveals a clear divergence between conventional diagnostic methods and those augmented by Cellvizio technology. While standard evaluations struggle with consistency, the integration of nCLE imaging raises overall accuracy to 96% for primary classification. More notably, the application of AI to this imaging data bridges a significant gap in detecting high-risk lesions, nearly doubling the accuracy rate from 53% to 87% in non-gastric IPMN subtypes. This suggests that the combination of hardware-based cellular imaging and software-driven analysis addresses both the reproducibility and sensitivity limitations inherent in traditional pancreatic cyst management.
| Metric | Conventional Methods | With Cellvizio / AI | Improvement |
|---|---|---|---|
| Mucinous vs. Non-mucinous Accuracy | N/A | 96% | Significant |
| High-Risk Lesion Detection (AI) | 53% | 87% | +34 percentage points |
| Unnecessary Surgery Reduction | Baseline | >40% reduction | In CLIMB study |
| Missed Cancers | Baseline | ~6-fold reduction | In CLIMB study |
Mauna Kea Technologies is listed on Euronext Growth under the ticker ALMKT. The company’s 2025 Annual Report was filed with the Autorité des marchés financiers (AMF) on April 30, 2026.
How might the demonstrated 96% diagnostic accuracy influence insurance reimbursement policies for Cellvizio procedures in the U.S. market?
What are the regulatory hurdles Mauna Kea Technologies faces in obtaining FDA clearance for its AI-assisted decision support tools?
Could the reduction in unnecessary surgeries by over 40% lead to significant cost savings for healthcare providers, thereby accelerating adoption rates?


























