Luis Louro - Fotolia
Published: 18 Jan 2018
The healthcare industry may not be quite there yet, but someday soon, an archived clinical picture may be worth a thousand data points. And I'm intrigued by what results medical image data analytics will yield.
We've come a long way since the printed family photo albums my grandmother kept in her kitchen cabinet. Photos, nowadays, can hold important and revealing digital information.
Someone's mugshot, for example, is not just another pretty face. That particular image can be searched online, scrutinized by facial recognition software and possibly help a potential employer determine whether a candidate is the best person for a job.
Precision medicine and population health management benefit from the data contained in patient charts, electronic health records and clinical images. So it's no surprise that the potential spreads far and wide if medical image data analysis can peel away the layers of an MRI image of a tumor.
When aggregated by software, the location of lung tumors in 50-year-old men may help oncologists determine which treatments statistically work better for that particular group of patients. Tracking the data drawn from these images over time may show improved patient outcomes or suggest mortality-rate predictions.
We all, of course, give up a little bit of control to achieve these types of modern medical findings. Images from a knee injury may end up in an online tome of similar images, with an application churning the data points in a dozen different ways. That reality is discomforting for some, but it's not all that different from online retailers tracking your every click.
With this vast amount of medical image data comes the need for efficient storage of information. Anyone with a smartphone knows how quickly selfies eat up space, never mind the storage space required for sophisticated images involving MRIs or CT scans.
The cloud is the ultimate destination for medical image storage, but healthcare clinicians, CIOs and IT managers have long been wary of fully trusting cloud computing with patient information. For now, hybrid cloud is the popular approach, which we explore further in this issue of Pulse.