ABSTRACT
AI virtual assistants have significant potential to alleviate the pressure on overly burdened healthcare systems by enabling patients to self-assess their symptoms and to seek further care when appropriate. For these systems to make a meaningful contribution to healthcare globally, they must be trusted by patients and healthcare professionals alike, and service the needs of patients in diverse regions and segments of the population. We developed an AI virtual assistant which provides patients with triage and diagnostic information. Crucially, the system is based on a generative model, which allows for relatively straightforward re-parameterization to reflect local disease and risk factor burden in diverse regions and population segments. This is an appealing property, particularly when considering the potential of AI systems to improve the provision of healthcare on a global scale in many regions and for both developing and developed countries. We performed a prospective validation study of the accuracy and safety of the AI system and human doctors. Importantly, we assessed the accuracy and safety of both the AI and human doctors independently against identical clinical cases and, unlike previous studies, also accounted for the information gathering process of both agents. Overall, we found that the AI system is able to provide patients with triage and diagnostic information with a level of clinical accuracy and safety comparable to that of human doctors. Through this approach and study, we hope to start building trust in AI-powered systems by directly comparing their performance to human doctors, who do not always agree with each other on the cause of patients' symptoms or the most appropriate triage recommendation.
ABSTRACT
INTRODUCTION: Kyphoplasty is gaining in popularity as a treatment for painful osteoporotic vertebral body fracture. It has the potential to restore vertebral shape and reduce spinal deformity, but the actual clinical and mechanical benefits of kyphoplasty remain unclear. In a cadaveric study, we compare the ability of vertebroplasty and kyphoplasty to restore spine mechanical function, and vertebral body shape, following vertebral fracture. METHODS: Fifteen pairs of thoracolumbar "motion segments" (two vertebrae with the intervening disc and ligaments) were obtained from cadavers aged 42-96 years. All specimens were compressed to induce vertebral body fracture. Then one of each pair underwent vertebroplasty and the other kyphoplasty, using 7 ml of polymethylmethacrylate cement. Augmented specimens were compressed for 2 hours to allow consolidation. At each stage of the experiment, motion segment stiffness was measured in bending and compression, and the distribution of loading on the vertebrae was determined by pulling a miniature pressure transducer through the intervertebral disc. Disc pressure measurements were performed in flexed and extended postures with a compressive load of 1.0-1.5 kN. They revealed the intradiscal pressure (IDP) which acts on the central vertebral body, and they enabled compressive load-bearing by the neural arch (F(N)) to be calculated. Changes in vertebral height and wedge angle were assessed from radiographs. The volume of leaked cement was determined by water displacement. Volumetric bone mineral density (BMD) of each vertebral body was calculated using DXA and water displacement. RESULTS: Vertebral fracture reduced motion segment compressive stiffness by 55%, and bending stiffness by 39%. IDP fell by 61-88%, depending on posture. F(N) increased from 15% to 36% in flexion and from 30% to 58% in extension (P<0.001). Fracture reduced vertebral height by an average 0.94 mm and increased vertebral wedging by 0.95 degrees (P<0.001). Vertebroplasty and kyphoplasty were equally effective in partially restoring all aspects of mechanical function (including stiffness, IDP, and F(N)), but vertebral wedging was reduced only by kyphoplasty (P<0.05). Changes in mechanical function and vertebral wedging were largely maintained after consolidation, but height restoration was not. Cement leakage was similar for both treatments. CONCLUSIONS: Vertebroplasty and kyphoplasty were equally effective at restoring mechanical function to an injured spine. Only kyphoplasty was able to reverse minor vertebral wedging.
