Researchers at the US-based Harvard Medical School have designed a versatile, ChatGPT-like artificial intelligence (AI) model capable of performing various diagnostic tasks across multiple forms of cancers.
According to the researchers, the new AI system, described in the journal Nature, goes a step beyond many current AI approaches to cancer diagnosis.
Also read: Researchers to use AI to spot signs of dementia from 1.6 mn brain scans
The current AI systems are usually trained for specific tasks, like detecting the presence of cancer or predicting the genetic profile of a tumour. They often only work for a few types of cancer. In contrast, the new model can perform a wide range of tasks and was tested on 19 different types of cancer, making it flexible similar to large language models such as ChatGPT, the researchers mentioned.
The new AI model called CHIEF (Clinical Histopathology Imaging Evaluation Foundation), was trained on 15 million unlabeled images chunked into sections of interest. The tool was then trained further on 60,000 whole-slide images of tissues including lung, breast, prostate, colorectal, stomach, esophageal, kidney, brain, liver, thyroid, pancreatic, cervical, uterine, ovarian, testicular, skin, soft tissue, adrenal gland, and bladder.
Also read: AI-powered robot performs world’s 1st fully automated dental procedure on human
The new AI model outperformed other state-of-the-art AI methods by up to 36 per cent on the following tasks — cancer cell detection, tumour origin identification, predicting patient outcomes, and identifying the presence of genes and DNA patterns related to treatment response, the researchers noted.
“Our ambition was to create a nimble, versatile ChatGPT-like AI platform that can perform a broad range of cancer evaluation tasks,” said study senior author Kun-Hsing Yu, assistant professor at Harvard Medical School.
“Our model turned out to be very useful across multiple tasks related to cancer detection, prognosis, and treatment response across multiple cancers,” he added.
Moreover, the researchers mentioned that CHIEF achieved nearly 96 per cent accuracy in detecting cancers and significantly outperformed present AI approaches across 15 datasets containing 11 cancer types.
The researchers hope that the AI model will assist clinicians in more accurately evaluating a patient’s tumour.