Meet the Bonebot; our AI solution to automatically detect fractured vertebrae
As part of our continuing effort to change the bone health landscape for the better, we have collaborated with University Hospital Brussels and KU Leuven to reinvigorate the current methodology of detecting vertebral (spinal) fractures, because of osteoporosis, which often go undetected.
Considered a silent epidemic in bone health; osteoporosis is estimated to affect 200 million people worldwide and is the most common bone disease, resulting in more than 8.9 million fragility fractures each year around the globe. Of these, vertebral fractures are the most common, with one occurring every 22 seconds worldwide in men and women over age 50.
The impact of these vertebral fractures can be distressing as they often lead to back pain, loss of height, deformity, immobility, increased number of bed days, and even reduced pulmonary function. The impact on a patient’s mental health is also significant with reports of loss of self-esteem, distorted body image and depression. Perhaps more importantly these fractures lead to a marked increase risk in additional fractures, a phenomenon referred to as the fracture cascade.
Despite the worrying impact of these fractures, it is estimated that only one-third of vertebral fractures come to clinical attention and more than half of vertebral fractures are under-reported by radiologists when reading spine-containing CT scans. As the complex nature of current radiological techniques are clearly not equipped to appropriately recognize these fractures, more must be done to address this deficit.
To help improve rates of diagnoses, we applied 3D convolutional neural networks to develop an artificial intelligence (AI)-based method that detects the presence of vertebral fractures in any spine-containing CT scans. This new technique does not require prior segmentation of the spine; instead, the model automatically learns compact 3D features of the scan and recognizes the fracture in a manner that is consistent to best clinical practice. This innovative prediction tool has the potential to significantly reduce the number of fractures that currently go undetected. We named it Bonebot.
With over 1.4 million patients each year suffering from vertebral fractures, the Bonebot can help to identify patients susceptible to future fractures and lead to more effective clinical intervention which could potentially help reduce the co-morbidities associated with osteoporosis.
The first results of the Bonebot were presented at this years RSNA annual congress in Chicago, USA and the online version of our paper can be accessed at https://arxiv.org/abs/1911.01816