By the time you finish reading this sentence, 3 people will have an Osteoporotic fracture somewhere on the globe. That’s 25,000 people per day, and 9M every year. A 50 year old woman has a 2.8% risk of death related to hip fracture – equivalent to her risk of death from breast cancer.
Osteoporosis is heavily under-diagnosed in most countries, and it is extremely prevalent. In the US alone there are 2 million fractures every year. Nearly 25% of all people who suffer from a hip fracture in the US enter a nursing home within 12 months of their fracture, due to reduced mobility, poor rehabilitation, loss of self confidence and the inability to reclaim their quality of life.
So what can we do about it? How can we know who is at risk? Can we prevent some of these fractures?
One of the limitations for early diagnosis of Osteoporosis is the lack of external symptoms. People are always reluctant to undergo screening for conditions they don’t believe they have. Some forward thinking organizations, such as the FLS (Fracture Liaison Service) and Kaiser SCAL, have put Bone Health and Fracture Prevention programs in place, to try and manage patients who have already suffered a fracture – and prevent them from having subsequent ones. But these programs only identify people after a first fracture has occurred!
Enter the Zebra Community crystal ball
Early detection of Osteoporosis can save lives. It can improve care, and it can reduce the overall cost of patient care. This has been proven time and time again by numerous studies.
So how can we detect Osteoporosis at an earlier phase? There’s an algorithm for that!
At Zebra, we developed an algorithm that examines a standard CT scan, performed for any reason, and extracts the bone mineral density of the individual. Using the algorithm, we can provide a risk profile for Osteoporosis, by scanning historic imaging studies, as well as new incoming studies. This risk profile, both at an organizational and individual level, allows healthcare providers to accurately identify people at risk, tailor specific preventative care programs for them, and manage their overall clinical risk in a much more efficient manner. At a personal level, it has tremendous impact on your overall well-being. Our algorithm was developed using the Zebra massive imaging database – correlating thousands of Bone Mineral Density scans to their corresponding CT scans to build the correct model.
And this is just the tip of the iceberg.
Early detection of cancer, liver disease, heart disease, vascular disease and many more will be developed over the next few years by researchers joining the Zebra research community. We continue to invite researchers interested in pushing the boundaries of machine learning and computer vision – and who want to make a difference in the world, to come check us out.
Also, if you happen to be in Las Vegas next week for HIMSS – come say hello at booth 4416, and be on the lookout for some exciting news as well!