Can Decision Management Help to Diagnose Rare Diseases?
cheryl wilson 270003VHSH firstname.lastname@example.org | | Tags:  business-rules decision-management predictive-analytics medical-diagnosis rare-disease-diagnosis
1 Comments | 3,250 Visits
Imagine ... Ed Snow comes home from jogging to find his wife Elyse sick in bed. It's believed to be a tumor, but scans turn up nothing, and treatments do nothing to stop Elyse from having seizures and hallucinations. Physicians narrow it down to two diseases – rabbit fever and African sleeping sickness – both of which have dangerous treatments and are very unlikely. She is treated for rabbit fever, after which she slips into a coma. When she's finally treated for sleeping sickness her condition improves. She lives.
That’s from an episode of the popular television show House. Dr. Gregory House is known for two things: his medical expertise and the acerbic wit with which he denigrates the doctors on his staff for making incorrect diagnoses.
Cut to the real world. Medical diagnosis is probably the most complex form of decision making, which is why it makes for good TV drama. The variables are numerous and the stakes are potentially life and death. There are thousands of diseases and any given disease can cause a variety of symptoms. Symptoms can overlap diseases and more than one disease can cause a given symptom.
To make matters more complicated, many diseases are rare. If it's been ten years since a disease has been seen at a hospital, how can you expect a doctor who has been practicing for five years to make the correct diagnosis? Misdiagnosis remains a significant obstacle to helping rare disease patients. Even in the case of correct diagnosis, it often takes too long, with potentially dire consequences.
Can available software help diagnose rare diseases? You sometimes hear that a doctor’s best asset is his or her memory, so it remains a challenge for any software package to out-diagnose the human brain’s ability to store and synthesize information and experience. But if you remember back to Watson’s victory over two renowned Jeopardy champions, it’s clear we’ve reached the point where information technology can have a seat at the table of experts (figuratively, of course). In the case of medical diagnosis, software can be invaluable in helping physicians identify rare diseases.
Decision management software is already in use at one of IBM’s clients in Belgium, the University Hospital of Antwerp. The hospital uses business rules management to consider all the known information about a patient -- along with predictive analytics to deduce things that are not known -- to help to determine the presence of rare diseases. Just like Watson on Jeopardy, the system accesses a database of information, and then makes sense of that information by applying knowledge-based logic (business rules) and predictive logic (analytics). The net result is that diagnoses are made earlier and more accurately, saving lives and improving the speed and quality of health care delivery.
Making a decision means taking information and turning it into action. This is never more true and relevant than in the medical profession, where the success or failure of decision making can determine the quality of someone’s life. Information technology can aid in this decision-making process and is increasingly doing so. Don’t be surprised if a future medical show includes a software-based system as a key member of the diagnostic team. As an added bonus, the digital “Dr. House” will probably communicate with much less sarcasm and cynicism than its human counterpart.
If you’d like to learn more about applying state-of-the-art decision management in virtually any industry, download this collection of real life case studies: Case Studies in Operational Decision Management.
This post was co-authored by Brian Safron and Cheryl Wilson, Business Process and Decision Management, IBM.