Future Health - 9/14/18
Delivering care in the new virtual world…
Let’s Be Careful – You will not find a bigger fan on the use of artificial intelligence and machine learning as a tool for “augmenting clinical intelligence” among providers. At the same time, I took pause at a recent issue of JAMA Network Open (2018;1(3): e180926.doi: 10.1001 – 2018-0926) which report on the application of machine learning as an approach toward assessing short-term mortality among cancer patients who were starting chemotherapy. The study included a cohort of 26,946 patients with cancer who were starting a wide variety of discrete chemotherapy regimens. With the use of machine learning and AI, those patients who were at high risk of 30-day mortality were accurately identified across palliative and curative chemotherapy regimens and across many different types/stages of cancer. The researchers also compared the computer algorithm to more conventional tools such as predictions based on randomized clinical trials and population-based registry data. The algorithm did a far better job. Their conclusion is what drew my attention. It said:
“A machine learning algorithm accurately identified individuals at high risk of short-term mortality and may help to guide patient and physician decisions about chemotherapy initiation and advance care planning.” [emphasis added]
Lest we forget, it is an imperative – from my perspective – to always keep in mind their conclusion that the use of machine learning and AI is a “guide” – not the definitive approach. In our enthusiasm to embrace new technologies, we always need to keep at the forefront that computers and machines still do not think – and, until they do, we should always hold their outcomes as support for “clinically augmented intelligence”. Make sense?
Bioprinting Your Medications – The new HP D300e Digital Dispenser BioPrinter technology is being tested by the company and the CDC on the effectiveness of printing (i.e. making and dispensing) and testing antimicrobial resistant strains of “super bugs” which is an increasing problem. The idea is to provide a select number of hospitals access to the technology nationally to cut down on the drug resistance problem by identifying the bugs earlier and dispensing the appropriate medications quicker. In addition to identifying resistant strains, the HP BioPrinter can also dispense medications in many doses thereby alleviating the frequent “wait” in pharmacies who need to obtain the correct dose prescribed by the clinicians. Soon, the antimicrobial resistance testing will become a common place component of dispensing medications for patients with the end result being a reduction in the 23,000+ people that die in the USA every year from infections by super bugs that are not treated as rapidly as can be managed using the new HP Bioprinter devices. But, the use for drug dispensing is only the beginning of a revolution in the “printing” business. Bioprinters are also being explored for use in human tissue and organ production, drug research and other clinical areas. Amazing!!