CHARTSaaS

”My own induction into diagnosis began in the fall of 1997, in Boston, as I started my clinical rotations. To prepare, I read a textbook, a classic in medical education, that divided the act of diagnosis into four tidy phases. First, the doctor uses a patient’s history and a physical exam to collect facts about her complaint or condition. Next, this information is collated to generate a comprehensive list of potential causes. Then questions and preliminary tests help eliminate one hypothesis and strengthen another—so-called “differential diagnosis.” Weight is given to how common a disease might be, and to a patient’s prior history, risks, exposures. (“When you hear hoofbeats,” the saying goes, “think horses, not zebras.”) The list narrows; the doctor refines her assessment. In the final phase, definitive lab tests, X-rays, or CT scans are deployed to confirm the hypothesis and seal the diagnosis. Variations of this stepwise process were faithfully reproduced in medical textbooks for decades, and the image of the diagnostician who plods methodically from symptom to cause had been imprinted on generations of medical students.

But the real art of diagnosis, I soon learned, wasn’t so straightforward …

The ‘black box’ problem is endemic in deep learning. The system isn’t guided by an explicit store of medical knowledge and a list of diagnostic rules; it has effectively taught itself to differentiate moles from melanomas by making vast numbers of internal adjustments—something analogous to strengthening and weakening synaptic connections in the brain. Exactly how did it determine that a lesion was a melanoma? We can’t know, and it can’t tell us. All the internal adjustments and processing that allow the network to learn happen away from our scrutiny. As is true of our own brains. When you make a slow turn on a bicycle, you lean in the opposite direction. My daughter knows to do this, but she doesn’t know that she does it. The melanoma machine must be extracting certain features from the images; does it matter that it can’t tell us which? It’s like the smiling god of knowledge. Encountering such a machine, one gets a glimpse of how an animal might perceive a human mind: all-knowing but perfectly impenetrable.”

The foregoing quote from The New Yorker April 3, 2017, edition article by Siddhartha Mukherjee, MD, PhD, entitled “A.I. Versus M.D. — What happens when diagnosis is automated?” (http://bit.ly/2okzZfM) discusses the methods and mysteries of human learning in the context of medical diagnosis. Taking the discourse and included examples into account, it can be inferred that part if not all of the “black box” nature of “deep learning” is the ability to recognize complex patterns and similar occurrences thereof. This is the essence of “similarity analytics,” which is an multi-variate analytic technique that can identify degrees of likeness among records or “cases” based upon a comparison of the data points or elements of information that each comprises. Thus, given the approximately four-hundred fifty data points that accrue from a complete medical “work-up” consisting of a patient history and physical exam, real-time access to a similarity analytics program would enable real-time comparison of the work-up data points of the current case with records in one or more databases of closed cases described by equivalent data points.

Similar case records could be presented in an order of probability of equivalence with the case at hand in a video display including the labeling of each case dataset with the recorded diagnosis for each, thus producing a differential diagnosis report for consideration by the attending physician a.k.a. the primary care provider.   Once the physician had determined the final diagnosis from his/her professional experience confirmed by tests or consultation, (s)he could submit the selected case(s) for further inspection to reveal the treatment plan followed for each and display of these in order of case outcome success a.k.a. condition on discharge. A Cloud Healthcare Appliance Real-Time Solution as a Service (CHARTSaaS) solution compliant with the CHARTSaaS IT reference architecture (RA) includes easy-to-use tools for healthcare provider subject matter experts (SMEs) to design, develop, deploy and use an IT applications or “app” that could automate a process for extraction of work-up data from electronic health record (EHR) system currently in use, execution of the  similarity analytics as described, and presentation of differential diagnosis and optimal treatment plan with a much higher and reliable degree of certainty that a similar process reliant exclusively on the cognitive capabilities of the physician; but which diagnosis and treatment would rely on the physician for confirmation and implementation.

Please validate this proposition to your own satisfaction by reading the white paper at http://bit.ly/2nhwqpd and then by reviewing the details of CHARTSaaS and the CHARTSaaS RA in these presentations:

Healthcare providers will benefit significantly from appreciating and then applying a CHARTSaaS RA-compliant IT solution. To do so will mitigate medical mistakes (currently the third leading cause of patient deaths. per Makaray and Daniel http://www.bmj.com/content/353/bmj.i2139), thereby minimizing patient adverse events and optimizing clinical case outcomes while maximizing the cost-effectiveness of care and treatment while accelerating the accrual and application of medical knowledge.