As new cyber tech has a greater effect on medical diagnostic practices, giving an accurate prognosis is undergoing the same transformation. Now artificial intelligence (AI) is coming into play, and it’s making prognoses and diagnoses even more accurate and effective.
For those who don’t know, a diagnosis is the identification of a disease, and the prognosis is basically a doctor’s notion of what success any treatment may have.
Traditionally, the effectiveness of certain medical treatments (drugs, for example) have been studied through randomized trials wherein patients are divided into two groups. One group is given the treatment to be tested, and the other a placebo (a faux treatment with all the psychological effect of the drug with none of the therapeutic, curative effect, like a sugar pill).
But that was then, and this is now.
Finnish researchers at the University of Eastern Finland, Kuopio University Hospital and Aalto University have a better way. They used AI computer modeling to compare different treatments and to identify which patients may benefit most from treatment.
Professor Emeritus Olli-Pekka Ryynänen from the University of Eastern Finland, said, “We can now predict the treatment outcome in individual patients and to evaluate existing and new treatment methods. With this method, it is also possible to replace some randomized trials with modeling.”
Researchers used AI-based computer modeling to study treatment effectiveness in obstructive sleep apnea. But the method can also be applied to various treatments of virtually countless diseases and conditions.
In cardiac physiology and cardiology, models are being used to illustrate combined activity within millions of cardiac cells, activity both electrical (the combined activity of all ion channels) or mechanical (contraction-related processes). Cardiac computer modeling owes its many great strides to research advancements and the ever-increasing computer processing power.
But modeling is still based on pre-existing template circumstances for their diagnoses and prognoses. The real goal, biomedical experts agree, will be the creation of patient-specific models.
And the figures tell us all that there’s little time to waste and an incredible amount at risk.
Latest studies indicate that 8.4 % of all people who die in US hospitals (850,000) have at least one major misdiagnosis, wherein the misdiagnosis directly contributes to the patient’s death. These are patients who could have been saved if they’d been correctly diagnosed. That amounts to 71,400 Americans who died needlessly. Taking into account minor misdiagnoses, wherein the misdiagnosis did not relate directly to the patient’s death, increases the percentage to a terrifying 28%.
While misdiagnoses get a lot more focus than faulty prognoses, the two are inextricably connected. On website bmj.com, Dhruv Khullar and Anupam Jena urge readers that the connection is more crucial to their personal wellbeing than most people realize: “…We argue that efforts to improve clinical decision making and patient outcomes by minimizing misdiagnosis will be limited if these efforts do not also seek to reduce [mis-prognoses]. To truly address the human and economic costs of failures in clinical decision making we must recognize that prognostic errors may be as important as diagnostic errors.”