Why Diagnose with AI/ML

Most of medicine is taught using what we call “clinical pearls”, the small bits of free standing, clinically relevant information often shared by a wise mentor, distilled from a life time of experience and observations. Each of these pearls are true, even if most of us are unable to fully explain why. Other clinical pearls don’t have a good physiologic underpinning. But, as soon as you begin to interrogate the physiology, it becomes quite clear there is a lot more going on. The explanation has interesting and important implications for treatment.

Many clinicians reflexively order monitoring labs and position their diagnostic approach according to the referance values of the labs’ given range of what is designated as “normal”. This in turn is to ascertain if the patient is within the specified range or not. Which may be valid for some cases.

This provides,

• Pertaining to the scheduled time for the consultation
• Less burn-out
• Presumably, clearance from potential law suits

All of which are required, to conclude a feasable practice. Yet, every patient depending on their specific condition may require subtle adjustments to the common referance values and ranges. Whatever the units of measurement may be, and however minute the difference between the custom designed reference markers and standard markers may be, it may make a great difference for the diagnosis, perhaps even turn it on its head.

In most cases it is highly difficult and time consuming, for the medical doctor or specialist to calculate and adjust to this “personal optimal range”. 

Each medical specialty has expertise in a set of symptoms for a specific pathology. There are some symptoms whose causes span many diseases and systems, that they can’t be owned just by the subspecialists but instead should also be “owned” by general internists.

Nevertheless, the exponentially growing complexity of systems biology, molecular biology, genetics, and the immense volumes of mega-data attached to these fields of science, makes it even more complicated for the MD to juggle all this knowledge without breaking a glass or two. This in turn, makes the ever-difficult medical profession even more compelling.