AI, NLG, and Machine Learning

How AI Can Help Reduce Burnout among Radiologists

Burnout is prevalent in the medical field. Artificial intelligence (AI) could help relieve radiologists from having to perform tedious tasks that can hamper productivity.

Burnout has become a serious problem for medical professionals. Radiologists often experience extreme emotional, physical, and mental stress. Now artificial intelligence (AI) could help relieve radiologists from having to perform tedious tasks that can hamper productivity.

A 2020 report shows that 46 percent of radiologists have felt burnout, which is among the highest rate for any physician specialty.

“Twenty years ago, radiologists read a quarter of the scans that we do today,” Steven S. Raman, director of Prostate MR Imaging Research Group, told the Radiology Society of North America. “We are working faster than ever before, but we are also facing greater demands.”

That’s why more and more radiologists are turning to AI to reduce the administrative load they carry while working.

In fact, over 70 percent of the US-based healthcare institutions are planning to use or have already begun using artificial intelligence to help alleviate the problem.

“AI delivers fully automated, rapid, on-demand, and standardized imaging results that enable prioritization of work lists and second reader evaluation, quantification, and classification of measures in support of radiologists’ clinical decision-making,” Mo Abdolell, CEO of Densitas, told discover.bot.

Abdolell said his company developed an AI product that helps radiologists to perform digital mammography with greater precision.

AI picks up the slack for radiologists by relieving the burden of performing monotonous work that can take time away from more important clinical engagement.

AI can automate the clinical image quality assessments of any exam taken at a medical facility once the image is captured. That gives radiologists a chance to review and retake the image while the patient is still in the room. AI can help a radiologist become more accurate, lowering the number of faulty images taken, which can ultimately decrease the need for patients to take another mammogram.

“AI is a supportive tool that can help boost performance when used to augment a radiologist’s work. But because AI algorithm performance depends entirely on the data upon which the algorithm is trained, and on the data to which the algorithm is applied, human expertise is still needed,” Abdolell said.

AI can give instant feedback for a mammogram, which research has shown can enhance radiologists’ performance. Decreasing the number of exams also reduces rescheduled visits. And fewer repeat exams for patients means less administrative work for radiologists and technologists. That also means less burnout for healthcare professionals.

Radiologists are often plagued by heaping administrative tasks that wear down energy and drain their desire to stay in the field. They’re often called on to perform highly repetitive reporting practices and read countless patient examinations. Those are two examples of the duties that can lead to burnout among radiologists and radiological technologists.

The limitations of traditional mammogram technology can produce flawed images, which bring on extra work for radiologists.

Standard mammogram images can be obscured by skin folds, cut off at certain portions of the image, or captured by poor quality imaging and microscopy facility equipment.

The Densitas AI solution integrates into whatever automated image assessment software radiologists are using. The product standardizes the image quality for all mammograms it evaluates. Two radiologists may view the same image drastically differently. But AI standardization eliminates the problem of variability between multiple radiologists by consistently reading an image the same way every time.

It also delivers on-demand analytics that help radiologists spot trends in mammogram images.

The Densitas solution analyzes image quality after the mammogram in real time. This allows radiologists to pinpoint image quality errors to decide, on the spot, whether to reexamine a patient—without having them return to the facility.

“AI automation combined with digitalization can deliver better reporting and workflow efficiencies, help radiological technologists to improve clinical image quality acquisition, and can improve patient and process management. As a result, radiologists can be liberated from tedious and repetitive reporting, and administrative burdens, so they can dedicate more time to more rewarding interpretive tasks that mitigate burnout,” Abdolell said.

Two AI solutions from Israeli-based Aidoc claim to save time for radiologists by speeding up the diagnosis and treatment of ischemic and hemorrhagic strokes. This is important since research shows earlier treatment of stroke patients leads to lower a likelihood of disabilities associated with the condition.

The first Aidoc solution uses head CT scans to target blockages in the eye’s blood vessel. Then the second solution examines those scans with an AI algorithm and alerts the radiologist of serious intracranial hemorrhaging. Having AI that can quickly interpret these results has the potential to free up radiologists to do more important clinical work.

According to Aidoc, the combination of solutions helped to lower by 36.6 percent the turnaround time for patients suffering intracranial hemorrhage.

That speed may help to reduce the negative outcomes from strokes, which cause about 5 percent of all deaths in the United States.

Each solution is a part of the company’s stroke package and has received FDA approval.

"The faster we can identify, Aidoc's comprehensive stroke package flags both large vessel occlusion and hemorrhages inside our existing workflows, ensuring we can diagnose stroke faster and decide on the best course of treatment. We're already seeing how this has a positive impact on department efficiency and patient length of stay." Dr. Marcel Maya, Co-chair Department of Imaging, Cedars-Sinai Medical Center said in a press release.

The AI medtech industry continues to grow. And the technology has the potential to cut down on burnout among radiologists, which in the long run, may also save lives.