VEA Align; spineEOS

VEA Align and spineEOS are cloud-based software systems designed to enhance orthopedic assessment and preoperative spine surgery planning. VEA Align employs a machine learning algorithm on 2D X-ray images for automated anatomical landmarking and global alignment analysis, providing data that can be utilized by spineEOS for detailed, patient-specific surgical strategy development.

VEA Align is a cloud-based software designed for orthopedic applications in pediatric (aged 7 years and older) and adult populations. It utilizes biplanar 2D X-ray images, exclusively from EOS imaging's systems, to display interactive anatomical landmarks. These landmarks are initially placed by a machine learning-based algorithm and can then be adjusted and validated by the user. The system aids healthcare professionals in assessing patient-specific global alignment and analyzing spinal deformities, degenerative diseases, and lower limb alignment disorders through precise angle and length measurements, with options to compare against published normative values. VEA Align can also export 3D datasets, and its AI algorithm for landmark placement was trained on over 10,000 X-ray images and validated for accuracy (median error ≤ 3 mm, 3rd quartile ≤ 5 mm) against ground truth data across diverse patient subgroups. spineEOS, also a cloud-based software, complements VEA Align by assisting healthcare professionals in preoperative planning for spine surgeries. It utilizes EOS images and their associated 3D datasets (which can be generated by VEA Align) along with clinical measurements to enable users to define patient-specific surgical strategies for adolescent and adult patients. spineEOS offers surgical planning tools for segmental alignment, interbody implants, osteotomy, spondylolisthesis, and rod curvature management. Both VEA Align and spineEOS require clinical judgment for appropriate use. Comprehensive nonclinical performance testing was conducted for both devices, including verification of design inputs, software and system integration, AI algorithm performance, as well as validation of multifunctional requirements, usability, and the accuracy of clinical parameter outputs.

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 quantified impact on surgical planning time or accuracy when using spineEOS with 3D datasets generated from VEA Align's AI-assisted landmarking versus manually generated datasets?; How does the performance of VEA Align's machine learning algorithm for landmark placement vary across different types or severities of spinal deformities not explicitly detailed in the subgroup analysis (e.g., very rare congenital conditions or severe post-traumatic deformities)?; What are the specific "published normative values" VEA Align uses for comparison, and how frequently are these values updated or validated for diverse demographic populations and evolving orthopedic standards?; Are there ongoing or planned clinical studies to evaluate the long-term clinical outcomes and patient-reported outcomes for patients whose surgical plans were developed using spineEOS, particularly concerning complex spinal reconstructions?; Given that VEA Align's AI algorithm is trained on EOS imaging system data, what are the specific technical challenges and retraining requirements anticipated for adapting it to utilize X-ray images from non-EOS imaging systems while maintaining comparable accuracy?