Feature article

Beyond Workflows: 4 Ways AI in Radiology Helps Patients

How AI enhances the patient experience before, during, and after every clinical interaction

For the past several years, there's been ample publicity about the tremendous impact of artificial intelligence (AI) on radiology, from timesaving efficiencies to resource allocations.1 But in all the talk of operational benefits, what about the benefits to the patients themselves?

Of course, patients may not see the backend gears turn as machines pore through behemoth volumes of scans and data for their ultimate benefit, but they do see positive changes on the front-end. And they can see—as much as clinicians or administrators do—that the technology could make a significant and meaningful difference in their lives.

Consumers understand the potentially beneficial impact of AI in healthcare; according to a 2018 survey from the analytics company SAS, 6 in 10 consumers felt comfortable with the prospect of their physicians using AI for treatment plans.2 This could be because AI investments made on the back-end yield major care improvements on the patient-facing side of the business.3 In radiology, those investments can make a marked difference on a patient’s experience at the clinic, relationship with their doctor, and overall treatment journey. Here’s how:

1. Patients get appointments and results faster than ever.

People have great expectations for imaging turnaround. According to recent research in the Journal of the American College of Radiology, patients wanted test results within 72 hours—and if they had a chest X-ray, that window shrunk to just two days.4

Without AI, radiologists (and therefore, patients) rely on the manual review of imaging—scans viewed one-at-a-time by human eyes. With machine aid, that timeline can be accelerated to make a three-day turnaround (or faster) more reasonable.

Patients see that time savings through faster turnaround, and faster scheduling, too. By unlocking efficiencies in staff and resources, it’s feasibly easier to pack more appointments into the day without undue stress to all.5

But most importantly, hypothetically consider the young mother whose abnormal screening results require follow-up imaging. The faster she can get clarity, and the faster she can get answers, the less time she and her family spend agonizing over the unknown.

2. Patients get a faster, and more comfortable, scanning experience.

AI also accelerates the scanning experience itself. For example, some magnetic resonance (MR) software can run a total cardiovascular scan, including anatomy, function, and flow, within 10 minutes. That’s an immense step up from the current standard of one- or two-hour exams, a change that patients will obviously enjoy.

But it’s not just the time it takes to complete scans—it’s also how the patient feels from start to finish. With new AI technologies that improve scan precision and clarity—such as 3D-volume configurations—patients don’t have to hold their breath or lie impossibly still throughout the experience.5

Such enhancements can make an already daunting ordeal, like an MRI, less intimidating through the patient's eyes. Take that same hypothetical example of the young mother. She's already scared of what may be growing inside of her. She's already trying her hardest to be still when the fear sends shakes through her body. A more comfortable scan is a pleasant surprise, especially when that scan is over within minutes.

3. Patients get diagnosed (and treated) more quickly.

With expedited review afforded by machine learning, clinicians not only have more information at their fingertips, but they have more time to spend diagnosing—and treating—for better care delivery all-around. This benefits everyone, especially patients.

Together, those resources help physicians determine the level of urgency of interventions, like heart transplantation.5 From a patient’s perspective, they see more knowledgeable practitioners, more timely treatments, and a quicker return to their previous quality of life.

Returning to the X-ray example, radiologists today are overburdened by STAT chest X-rays marked for urgent reading. Turnaround to report may take over eight hours.6 Embedding AI in X-ray systemscan help identify critical conditions such as pneumothorax in the ER and ICU and enable prioritization.

4. Patients get more face-time with their doctors.

And finally, patients want practitioners with the bandwidth to answer their questions in a thoughtful manner, but physicians buried in heaps of data have limited time to do so. With AI, machines can take over the more machine-oriented jobs like calculation and review, while the humans can get back to where they’re really needed: the human component of medicine, write authors in European Radiology Experimental.7

That face-to-face interaction can make a significant difference from the patient's perspective: after all, at the end of the day, that patient doesn't want to talk to a machine. She's not interested in learning about the technology, however advanced it may be. She wants to talk to her doctor. Her family does too. She wants to hear a human's voice that there is a plan. She wants to look at her oncologist's face as he explains it to her, patiently and calmly. However invisible, AI helps make that happen.

A win-win-win for all

Indeed, AI has the potential to serve a very great purpose in radiology—that of enhancing the patient experience before, during, and after every clinical interaction. From faster scheduling and scans to improved turnaround and doctor-patient rapport, technology continues to hum in the background while patients, physicians, and administrators see tangible benefits on the front-end.

And all told, it’s a win-win-win solution for everyone involved.

References:

  1. How Artificial Intelligence Will Change Medical Imaging. Imaging Technology News. https://www.itnonline.com/article/how-artificial-intelligence-will-change-medical-imaging. Accessed Feb. 14, 2019.
  2. Consumers Support AI in Healthcare More than Other Industries. Health IT Analytics. https://healthitanalytics.com/news/consumers-support-ai-in-healthcare-more-than-other-industries. Accessed Feb. 14, 2019.
  3. How Artificial Intelligence is Changing Radiology, Pathology. Health IT Analytics. https://healthitanalytics.com/news/how-artificial-intelligence-is-changing-radiology-pathology. Accessed Feb. 14, 2019.
  4. Radiologists Want Patients to Get Results Faster. https://www.reuters.com/article/us-radiology-results-timeliness/radiologists-want-patients-to-get-test-results-faster-idUSKBN1DH2R6. Accessed Feb. 14, 2019.
  5. This Cardiac Software Originating from a Stanford Basement is Now One of the Top of Artificial Intelligence Solutions Available. GE Healthcare. http://newsroom.gehealthcare.com/cardiac-software-stanford-basement-artificial-intelligence/. Accessed Feb. 14, 2019.
  6. Reducing STAT Portable Chest Radiograph Turnaround Times: A Pilot Study. Current Problems in Diagnostic Radiology. doi: 1067/j.cpradiol.2017.05.012. Last accessed July 8, 2019.
  7. Artificial Intelligence in Medical Imaging: Threat or Opportunity? Radiologists Again at the Forefront of Innovation in Medicine. European Radiology Experimental. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6199205/. Accessed Feb. 14, 2019.