Customize

Customize is a medical image management and processing system designed for the pre-operative planning of total ankle arthroplasty in adult patients. It utilizes CT and X-ray imaging data, employing AI-driven segmentation and 3D modeling, to assist surgeons in visualizing, modifying, and validating surgical plans with an integrated implant library for optimal implant selection and placement.

Customize is a software interface designed to assist clinicians in the pre-operative planning of total ankle arthroplasty for adult patients. Its methodology involves processing medical imaging data, specifically multi-slice DICOM CT scans and, notably, X-ray images, to generate 3D CAD models of the patient's anatomy. The software facilitates the visualization, modification, and validation of surgical plans, allowing physicians to review these plans in 2D (CT and X-ray) and 3D, and adjust the position and orientation of 3D objects, including selecting appropriate implant models from an integrated library for optimal positioning and sizing. Scientifically, it employs nonadaptive machine learning algorithms, trained on a dataset of 126 CT scans (including arthroplasty-eligible and cadaveric cases), for initial image segmentation of bone structures, which is then reviewed and potentially refined under human supervision by a 3D-Side engineer. This AI-driven segmentation, combined with the ability to incorporate landmark information from X-ray images, aims to enhance planning accuracy for ankle arthroplasty. Functionally, Customize serves as a comprehensive platform for managing ankle arthroplasty cases, encompassing CT data segmentation, 3D model creation, communication of treatment options, and the saving of finalized plans for surgical reference; these plans can also be exported for external design and manufacturing of patient-specific devices. Non-clinical performance data demonstrated the software's validation for its intended use, including segmentation validation, repeatability and reproducibility studies for ankle anatomy segmentation, and accuracy assessments for 3D model generation of the tibia and talus. Customize is intended for prescription use only and is not designed for spine surgeries, direct implant or instrument design within the software, or for diagnostic purposes. Its proper use necessitates clinical experience and assessment from the user.

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 quantitative accuracy of the 3D model generation for the tibia and talus, and how does it compare to manual methods or other software?; How does the incorporation of X-ray image data specifically improve the pre-operative planning parameters for ankle arthroplasty compared to using CT data alone?; What is the typical extent and nature of manual editing required by the 3D-Side engineer after the AI-based initial segmentation, and what are the most common reasons for these edits?; Given the training dataset composition (70% cadaveric, 30% arthroplasty-eligible, average age 62), how does the AI segmentation performance generalize to younger patients or those with significantly different bone pathologies not well-represented in the training data?; How does the 'Customize' software integrate with the workflow for designing and manufacturing patient-specific devices, given that these functions are performed using external software?