PeekMed web

PeekMed web is a cloud-based, prescription-use decision support tool designed for healthcare professionals to conduct comprehensive 2D/3D pre-operative planning for adult patients with injuries or disabilities, utilizing patient imaging and surgical material databases. The system employs validated Machine Learning models for tasks like segmentation and landmarking, and its performance has been rigorously tested through non-clinical and internal validation, confirming its substantial equivalence to existing market devices.

Peek Health, S.A. has developed "PeekMed web," a cloud-based system designed to assist healthcare professionals in pre-operative planning for various surgical procedures. This software tool utilizes imported patient imaging studies and integrates a database of digital representations of surgical materials from manufacturers. It enables medical professionals to perform 2D or 3D planning, conduct relevant measurements, and add templates, providing a comprehensive overview of the intended surgery. PeekMed web is classified as a prescription-use device, serving as a decision support tool for qualified healthcare professionals to efficiently plan procedures for adult patients with diagnosed injuries or disabilities, typically used in a computer-equipped room.The PeekMed web system has been determined to be substantially equivalent to an existing device on the market, sharing the same intended use, indications for use, user base, patient population, and overall technical and functional capabilities. Non-clinical performance testing was conducted, including comprehensive verification of its features and rigorous validation of its embedded Machine Learning (ML) models for segmentation, landmarking, classification, and detection. These ML models were developed, trained, and externally validated using extensive, independent datasets, consistently meeting predefined acceptance criteria. Internal validation tests, simulating real-world environments, confirmed the software's acceptable performance across all anatomical areas and primary functions, reinforcing its equivalence.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What specific types of surgical procedures and anatomical areas does PeekMed web support, and are there any limitations?; How does the "substantially equivalent" determination translate into specific performance metrics or clinical benefits compared to the predicate device?; What were the specific datasets used for training and validating the ML models, and how was their independence ensured to prevent bias?; Can the system integrate with existing hospital PACS or EMR systems, and what are the workflow implications for clinicians?; What post-market surveillance or ongoing validation processes are in place to ensure continued accuracy and performance of the ML models over time and with diverse patient populations?