Minimally Invasive Scoliosis for AF

Minimally Invasive Spinal Technology (MIST) offers an all-in-one solution for the diagnosis and clinical management of several spinal deformities and injuries. We have developed a computer vision and machine learning (ML) based software – the Vision Augment Predictive Screening (VAPS) system – which uses anatomical data from X-ray images of patients to determine accurate and multi-level measures of spinal curvature (e.g., scoliosis, kyphosis, lordosis). We use ML to predict disease progression by combining extracted anatomical data with demographic parameters of the patient (e.g., age, gender, height, weight). Current diagnosis methods are time consuming, error prone (with an average inaccuracy of 5-7°), and not representative of the 3D deformity, resulting in patients being under- or over-diagnosed. MIST provides an automated and accurate process (allowing for a maximum error of 2°), remote diagnosis (results are stored in a cloud and immediately available), faster analysis (2 min as opposed to 30 min of manual analysis by a physician), and the potential for predictive capabilities (not offered by any other system). This will result is optimized medical workflows and reduced overall cost of care (by 20%). With this Phase I Project, we want to approach department of defense (DoD) Program Executive Offices (PEOs) Healthcare Management and Digital Systems.