Scientists are tasking a language-processing Artificial Intelligence (AI) with detecting and diagnosing the early signs of language-associated cognitive impairments in people with failing livers.
In their findings, the researchers report finding evidence that this cognitive function is likely to be restored following a liver transplant.
Liver failure and cognitive function
In their paper, published in the journal npj Digital Medicine (formerly Nature Digital Medicine), the researchers explained how they used natural language processing (NPL) to evaluate electronic message samples from patients with chronic liver failure.
This disease is associated with transient cognitive abnormalities. These include diminished attention spans, loss of memory, and a reduced ability for an individual to detect and respond to their surroundings.
This is due to a failing liver that can no longer properly remove toxins from the blood, allowing them to cross the brain-blood barrier.
As many as 20% of adults with chronic liver failure develop the worst form of cognitive impairment, overt hepatic encephalopathy. This has a mortality rate of 43% after one year.
"We currently do not have a reliable method for identifying cognitive abnormalities in patients who need a liver transplant," senior study author Douglas Mogul, said in a press release.
Natural language processing
"Our findings suggest that NLP may provide that early diagnosis of cognitive issues, guide us in managing the problem, and help improve a patient's quality of life until a donor organ is available."
Incredibly, the researchers were able to adapt their AI to detect distinct, yet subtle, pre-transplant, and post-transplant differences in language use by liver transplant patients. These include changes in sentence length, word length, and other language characteristics.
For patients with the worst form of liver failure, the scientists "found that their messages had fewer letters per word, fewer words of six letters or more, and more words per sentence before their transplants," said study co-author and computational linguist Masoud Rouhizadeh, M.Sc., Ph.D.
While more research is necessary, the scientists hope that their technology could be developed into a valuable tool for diagnosing patients.