NEWS

The detection skills of colon cancer could weaken with AI, discovers the study

NEWYou can now listen to Fox News articles!

The advantages of artificial intelligence (AI) in the medical space are constantly increasing, but the evidence suggests that it can also include risks.

A new study by European researchers examined how AI could change the behavior of endoscopists when driving a colonoscopy, and how their performance decreases when they do not use AI.

Research followed clinicians from four endoscopy centers in Poland participating in the ACCTST (artificial intelligence in colonoscopy for cancer prevention) trial, where Polyps detection was introduced at the end of 2021.

The FDA approves the first AI tool to predict the risk of breast cancer

The colonoscopies of these centers have been selected at random to be administered with or without AI assistance in AI.

The researchers evaluated the quality of the colonoscopies by comparing the identification of tumors (also known as the adenoma detection rate, or ADR) three months before and three months after the implementation of the AI.

Doctor examining View colonoscopy

A study discovered that doctors who used AI in colonoscopies have worsened to locate the cancer themselves. (istock)

From September 2021 to March 2022, 1,443 patients underwent a colonoscopy not assisted by AI before and after the introduction of the AI.

The study revealed that the tumor detection rate decreased “significantly”, ranging from 28.4% before the AI ​​exposure to 22.4% after exposure to AI.

The results were published in the journal Lancet gastroenterology and hepatology.

Colorectal cancer diagnostics soar in young adults for a key reason

According to the analysis of researchers, exposure to AI, the sex and age of the patient were “independent factors” associated with the tumor identification rate.

The researchers concluded that “continuous exposure to AI could reduce the ADR of standard colonoscopy not assisted by AI, suggesting a negative effect on endoscopic behavior”.

Medical illustration of colorectal cancer

Research has identified a 6% drop in detection rates during colonoscopies not assisted by AI. (istock)

Harvey Castro, MD, emergency doctor and AI specialist and Texas -based health care, said that he considered this study an “important work”.

“This study highlights a paradox in medicine: artificial intelligence can help us detect cancer, but it can also weaken the doctor’s own capacity to see what matters when the tool is not available,” he said in an interview with Fox News Digital.

Click here to obtain the Fox News app

Castro stressed that even a change of 1% can affect the survival of colon cancer for “thousands of patients”, which makes the 6% drop in significant detection rates.

“Even small changes in adenoma detection can change cancer results,” he said. “A reduction in a few percentage points is significant in terms of the population.”

Click here to register for our Health Newsletter

The expert recommends integrating the “wisely” AI instead of avoiding all of this together.

“As a futuristic doctor, I believe that artificial intelligence is one of the most powerful tools we have to improve detection and save lives,” said Castro.

Doctor of colonoscopy reading

“Even small changes in adenoma detection can change cancer results,” said a doctor and an AI expert. “A reduction in a few percentage points is significant in terms of the population.” (istock)

“At the same time, this study reminds us of a simple truth: medicine is always a human profession. The eye, judgment and recognition of doctor’s models remain irreplaceable.”

Castro said the right way to follow with medicine is balanced, conceiving it to strengthen clinicians instead of weakening.

For more health items, visit www.foxnews.com/health

“This requires training, monitoring and intentional design,” he said.

“The best future is the one where technology and humanity work side by side, Doctor Plus Machine, guaranteeing patients the safest and most effective care as possible.”

Related Articles

Back to top button