Precision AI Surgical Planning System (PAI-SPS)

The Precision AI Surgical Planning System aims to enhance shoulder arthroplasty outcomes by providing surgeons with advanced pre-operative 3D visualization, simulation, and planning tools, leveraging patient-specific CT data and machine learning for semi-automated analysis. Its methodology involves creating customized, 3D-printed surgical guides from biocompatible Polyamide-12 to translate the virtual plan into precise intraoperative execution, thereby improving implant positioning accuracy for adult patients undergoing total or reverse shoulder replacement.

The Precision AI Surgical Planning System (PAI-SPS) is a patient-specific medical device designed to assist surgeons in planning and executing shoulder joint arthroplasty, including total anatomic and reverse shoulder replacement. The system comprises two main components: a software module and 3D-printed hardware. The software allows for pre-surgical simulation by creating a 3D bone construct from patient CT scans, enabling surgeons to visualize joint mechanics, identify issues, and virtually position implant components to determine the optimal placement. This planning phase results in a surgical report and a data file that can be used to design patient-specific surgical guides and biomodels compatible with specific Enovis implant systems. If requested, the system's hardware component—Precision AI Surgical Guides and Biomodels—are manufactured using the CAD files generated by the software. These single-use guides are 3D printed from a biocompatible Polyamide-12 material and are customized to the individual patient's anatomy. Intraoperatively, these guides assist in accurately positioning implant components, such as placing K-wires for the glenoid and humeral pins for humeral head resection, by referencing anatomic landmarks identified on preoperative CT scans, specifically for a delto-pectoral approach. The system's software functionality incorporates non-adaptive machine learning algorithms for semi-automatic image segmentation and landmark identification. Its performance and accuracy were evaluated through non-clinical tests including biocompatibility, dimensional stability, and cadaveric studies, as well as a post-market clinical case series of 35 subjects which assessed measurement accuracy using post-operative CT scans.

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 is the quantifiable improvement in surgical accuracy (e.g., implant positioning, resection accuracy) when using the PAI-SPS compared to traditional methods or other patient-specific instrumentation systems?; How does the non-adaptive machine learning algorithm account for anatomical variability or pathological deformities beyond its training dataset in the semi-automatic segmentation and landmark identification process?; What are the learning curve implications for surgeons adopting the PAI-SPS, and what specific training is recommended to maximize its benefits?; The system is indicated for a delto-pectoral approach; what are the technical limitations or contraindications for its use with other surgical approaches to the shoulder?; Could the pre-surgical plan data file generated by the software be integrated with robotic-assisted surgical systems, and are there plans for such compatibility?