Humans have, so far, invested humongous amount of effort and resources in an attempt to better understand themselves, their origins, and everything they are capable of doing. Not every piece of this pursuit has been successful, but the kind of progress we have made showcases an interesting pattern. As per this pattern, our efforts have gotten more and more laser-focused over the course of time, thus signaling a significant shift in human mindset. Now, there is no scarcity of examples to back this up, however, nothing epitomizes it better than the man’s greatest creation i.e. technology. At the beginning, technology’s potential was far from apparent. In fact, the initial consensus pretty much adjudged it as another ‘funky’ attempt by humans to achieve something that might just never get into their reach. However, as we know, things panned out a lot differently. We would slowly come to notice the true value of what we had on our hands, and once that moment of realization happened, there was no looking back. Technology was suddenly the centerpiece of everything we did. It inspired us to set off on journeys that would literally transform the way we looked at our lives. One such journey we took under the guidance of technology was in medical sector. The idea of healthcare had remained grim for years, so it was only right to restructure it and open the doors to a better world. With cutting-edge medical solutions, we were able to get there and beyond. Today, people across the world are much more in the knowing about their bodies than ever before. Medical institutions, on the other hand, have also exponentially expanded their arsenal of treatments. Nevertheless, they still lack that proficiency to resolve some complicated medical problems. To change that up, Kaiser Permanente has just proposed an idea for the ages.
Folks at Kaiser Permanente, California, have suggested the use of Natural Language Processing tool for dealing with highly-specific echocardiography conditions like valvular heart disease, aortic stenosis, mild valve disease, and many more that still don’t have a reliable treatment. The institution has argued that the reason why these areas have remained somewhat underdeveloped is because the traditional reports haven’t done enough to supplement the diagnostic process. Hence, the system needs a segregation of meaningful data from the remaining pack, and there is no solution more efficient than NLP for this job.
NLP is quickly growing to become one of the most trusted branches of AI. In a world where data is almost exceeding the requirements, it has become really important to make a distinction between what is useful and what is not. Amidst these growing troves of data, it looks to be medical sector’s turn to rope in NLP skill-set for better patient management. As a way of establishing that this technology will be an effective addition across the board, Kaiser Permanente’s team ran it on their own database which constituted of nearly a million echocardiography reports. In a few minutes, the software diagnosed over 54,000 patients with different conditions, a process that would have taken years to complete, if done manually.
“Not only did we identify the patients, but we were able to also extract all the key detailed elements from each echocardiography report. Now we are using this data to examine our practice patterns and outcomes for these patients so that we can improve our care and understanding of these patients for ourselves and the broader medical community,” said Dr. Mathew Solomon, cardiologist and physician researcher at Kaiser Permanente.