• AI
  • Molecular Imaging
  • CT
  • X-Ray
  • Ultrasound
  • MRI
  • Facility Management
  • Mammography

Decision support: so long as we know who is right...


The holiday season is over, and it is the timeto make New Year's resolutions. What toresolve? My perception that the finishedarticle could hardly be improved upon isfrequently challenged by my family.

The holiday season is over, and it is the time to make New Year's resolutions. What to resolve? My perception that the finished article could hardly be improved upon is frequently challenged by my family. "You could lose weight, you should drink less, you could get home earlier, you could be less grumpy," they tell me.

With so many problems, it was difficult deciding which to address first. I needed help. Perhaps my first resolution should be to consult the computer more often. Google indicated that I needed a decision support system. I really had no idea of the extent to which I could delegate decisions to an inanimate object, thereby blaming not someone else but something else for all my errors and deficiencies.

These computerized information systems store data that can help individuals or organizations make evidence-based decisions. They have increasing applications throughout industry and commerce, as well as healthcare. Apparently, these approaches can be passive (predominantly an information resource), active (the system will provide solutions based on the data), or collaborative (the individual accessing the system can modify the suggestions). Computer-aided diagnosis and clinical support systems fall, at present, into the last category.

Some systems, such as www.mapofmedicine.com, represent an aide memoire for the clinician seeking to determine the best clinical pathway for a patient presenting with a set of symptoms. Others, such as CAD within radiology, allow differentiation of images into specific diagnostic groups that can subsequently be reviewed by the radiologist. This may have specific advantages (e.g., when using imaging for screening).

Like me, perhaps you thought that artificial intelligence was a recent concept in clinical radiology. Not so. I am not sure that Thomas Bayes foresaw the imaging applications of his theories as early as 1763, but CAD of bone tumors was first suggested in 1963, and general radiological applications were proposed in 1966. I must have been watching the soccer World Cup Final because the latter completely passed me by.

CAD has the potential to address a number of issues, the first of which could be our ever-burgeoning workload. Imagine it: an entire department of virtual colleagues. I could pick and choose the examinations I reported. I could schedule my duties so that I never worked during public holidays but was always on duty during the January retail sales. I wonder if we could get these colleagues to do the defecating proctograms?

I would have to learn a completely new language, but then I have always been adept at spouting meaningless drivel (ref. A Dose of Dubbins). I am sure I could intersperse a few sage comments, such as "using knowledge to create knowledge is the major concept of the emerging knowledge society." I could refer to my own webliography, supporting my own opinion, and wax lyrical about compunetics, perhaps even creating a wikidubbins without really having any idea what I was talking about.

Decision support could go much further. When asked, "Do you prefer the little black dress or the trouser suit?" or "Does my bum look big in this?", I could check with an active decision support system, thereby ensuring an evidence-based answer and avoiding marital strife.

In medicine, CAD can be applied not only to image interpretation but also to clinical diagnosis, to determine management pathways and support remote prescribing. It can streamline the patient's journey, ensuring greater consistency of care and saving costs, particularly where the cost of different management options with similar clinical outcome is included in the data. In remote communities, clinical care could be supported by access to an expert knowledge base structured to guide management decisions.

Within radiology, the potential to address workload and workforce issues is immeasurable. "There is an urgent need to provide assistance in handling the several million radiographs read by radiologists each year. The need for computer-aided diagnosis is becoming increasingly urgent." Amen to that. Except that this quote was made nearly 40 years ago, and I have still not seen a reduction in my PACS work list.

The biggest disappointment to come out of my newfound enthusiasm for all things IT is that my vision of a stress-free workplace with compliant virtual colleagues has been challenged. Apparently we cannot get on. Three different types of workplace tension are likely when using decision support or electronic knowledge resource: between doctor and inanimate object, between doctor and other healthcare workers, and between doctor and healthcare organization.

Of course, it wouldn't happen in my department, so long as everyone and everything realized that I am always right. And I could always kick the computer like I always have done. They don't kick back . . . yet.

Related Videos
Does Initial CCTA Provide the Best Assessment of Stable Chest Pain?
Making the Case for Intravascular Ultrasound Use in Peripheral Vascular Interventions
Can Diffusion Microstructural Imaging Provide Insights into Long Covid Beyond Conventional MRI?
Assessing the Impact of Radiology Workforce Shortages in Rural Communities
Emerging MRI and PET Research Reveals Link Between Visceral Abdominal Fat and Early Signs of Alzheimer’s Disease
Reimbursement Challenges in Radiology: An Interview with Richard Heller, MD
Nina Kottler, MD, MS
The Executive Order on AI: Promising Development for Radiology or ‘HIPAA for AI’?
Related Content
© 2024 MJH Life Sciences

All rights reserved.