Free Republic
Browse · Search
General/Chat
Topics · Post Article

Skip to comments.

Study Cautions Radiologists Not to Over-Rely on AI Tools for Diagnosis
Euronews ^ | 25/11/2024 | Oceane Duboust

Posted on 11/29/2024 2:59:15 PM PST by nickcarraway

Researchers also found that physicians were more likely to trust an AI explanation if it pinpointed a specific area on an X-ray.

While artificial intelligence (AI) is a revolutionising tool in medicine, radiologists may over-rely on its advice when it highlights a specific part of an X-ray, according to a new study.

A team of US researchers recruited 220 physicians across multiple sites in the country tasked with reviewing chest X-rays alongside AI-generated advice.

Participants included radiologists and internal or emergency medicine physicians who were tasked with reading X-rays with the help of an AI assistant. The physicians could accept, modify, or reject the AI suggestions.

The study, published in the journal Radiology, explored how the type of AI advice, either local or global, and its accuracy affected a diagnosis.

A local explanation is when the AI highlights specific areas of interest in an X-ray, while a global explanation is when the AI provides images from similar past cases to show how it made the suggestion.

"We found that local explanations improved diagnostic accuracy and reduced interpretation time when the AI’s advice was correct," Dr Paul H Yi, one of the study’s co-authors and director of intelligent imaging informatics at St Jude Children’s Research Hospital, told Euronews Health.

Will AI replace radiologists or make them more efficient?

When the AI provided accurate advice, local explanations led to reviewers having a diagnostic accuracy rate of 92.8 per cent and global explanations to an accuracy rate of 85.3 per cent.

However, when the AI’s diagnosis was incorrect, diagnostic accuracy dropped to 23.6 per cent for local explanations and to 26.1 per cent for physicians with global explanations.

"These findings emphasise the importance of carefully designing AI tools. Thoughtful explanation design is not just an add-on; it’s a pivotal factor in ensuring AI enhances clinical practice rather than introducing unintended risks," Yi said.

'Type of AI explanation' can impact trust An unexpected finding was how quickly physicians, both radiologists, and non-radiologists, trusted local explanations, even when the AI was incorrect.

"This reveals a subtle but critical insight: the type of AI explanation can shape trust and decision-making in ways users may not even realise," he added.

He has several suggestions for mitigating this risk of "automation bias" - the human tendency to over-rely on automation.

He said physicians learn through years of training and repetition to follow a pattern or checklist.

"The idea is that it creates a routine. It minimises variation which can cause unexpected mistakes to happen," he said.

However, introducing AI tools adds a new factor and can derail this routine.

"We have to stick to our checklists and make sure we adhere to them. But I envision a future where our checklists are actually going to change to incorporate AI," Yi said, adding that human-computer interaction should also be studied with factors such as stress or tiredness.


TOPICS: Computers/Internet; Health/Medicine; Science
KEYWORDS: ai; radiology; xray

1 posted on 11/29/2024 2:59:15 PM PST by nickcarraway
[ Post Reply | Private Reply | View Replies]

To: nickcarraway

I suspect their AI will suggest euthanasia too often.


2 posted on 11/29/2024 3:00:24 PM PST by aynrandfreak (Being a Democrat means never having to say you're sorry)
[ Post Reply | Private Reply | To 1 | View Replies]

To: ConservativeMind

Ping


3 posted on 11/29/2024 3:03:56 PM PST by nickcarraway
[ Post Reply | Private Reply | To 1 | View Replies]

To: aynrandfreak

Yes.


4 posted on 11/29/2024 3:04:05 PM PST by nickcarraway
[ Post Reply | Private Reply | To 2 | View Replies]

To: nickcarraway
The same issue will come up with Dermatology.

There is a central curated photographic database of skin lesions.

More than 100,000 high-def photos, as I recall.

Computer software predictive accuracy - cancer/not cancer - was more than 90% a couple years ago.

Skin cancer - in my opinion - is the most over-diagnosed disease in the history of modern medical science.

5 posted on 11/29/2024 3:47:24 PM PST by zeestephen (Trump "Lost" By 43,000 Votes - Spread Across Three States - GA, WI, AZ)
[ Post Reply | Private Reply | To 1 | View Replies]

To: nickcarraway

Lol I just had a MRI done today and happened to ask how the Radiologist went about reading them.


6 posted on 11/29/2024 4:37:19 PM PST by TalBlack (Time to use the Law and the Power. Good luck Mr. President.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: TalBlack

What did they say?


7 posted on 11/29/2024 4:38:21 PM PST by nickcarraway
[ Post Reply | Private Reply | To 6 | View Replies]

To: nickcarraway

Ha ha you think you ask a simple question.

The radiologist reads it at his leisure then reports to the Dr. I’ll find out in a week or two. What prompted me to get one was sudden almost paralyzing bolts of back pain. After a lifetime of abusing my back yet never getting a look in there I decided I need to know what’s going on in there. Got the MRI, X-rays and front side and back ultra sound to boot. I’m not screwing around. The happy part so far is that the Doc gave me a prescription for 100 pills 15mg Maloxicam. He said you could take one of these, the swelling go down “and everything pop back into place”. Damned if that didn’t happen! Took one and gone!


8 posted on 11/30/2024 7:38:53 AM PST by TalBlack (Time to use the Law and the Power. Good luck Mr. President.)
[ Post Reply | Private Reply | To 7 | View Replies]

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Free Republic
Browse · Search
General/Chat
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson