
-
Brazil binman finds newborn baby on garbage route
-
US senator smashes record with marathon anti-Trump speech
-
Trump advisor Waltz faces new pressure over Gmail usage
-
Niger junta frees ministers of overthrown government
-
Trump set to unleash 'Liberation Day' tariffs
-
Boeing chief to acknowledge 'serious missteps' at US Senate hearing
-
Real Madrid hold Real Sociedad in eight-goal thriller to reach Copa del Rey final
-
Nuno salutes 'special' Elanga after stunning strike fires Forest
-
PSG survive scare against Dunkerque to reach French Cup final
-
Sundowns edge Esperance as crowd violence mars quarter-final
-
Nottingham Forest beat Man Utd, Saka scores on Arsenal return
-
Elanga wonder-goal sinks Man Utd as Forest eye Champions League berth
-
Stock markets mostly advance ahead of Trump tariffs deadline
-
US movie theaters urge 45-day 'baseline' before films hit streaming
-
Saka scores on return as Arsenal beat Fulham
-
Third-division Bielefeld shock holders Leverkusen in German Cup
-
Ball-blasting 'Torpedo bats' making waves across MLB opening weekend
-
Newsmax shares surge more than 2,000% in days after IPO
-
Thousands of Hungarians protest against Pride ban law
-
GM leads first quarter US auto sales as tariffs loom
-
Tesla sales tumble in Europe in the first quarter
-
No 'eye for an eye' approach to US tariffs: Mexico
-
NFL club owners back dynamic kickoffs, delay tush push vote
-
Trump 'perfecting' new tariffs as nervous world braces
-
Trump nominee says to press UK on Israel arms
-
French court says Le Pen appeal ruling could come before presidential vote
-
The battle to control assets behind Bosnia crisis
-
Prabhsimran powers Punjab to IPL win over Lucknow
-
Mass layoffs targeting 10,000 jobs hit US health agencies
-
Tiger's April Foolishness: plan to play Masters just a joke
-
Myanmar quake toll passes 2,700, nation halts to honour victims
-
Turkish fans, artists urge Muse to cancel Istanbul gig
-
US seeks death penalty for accused killer of insurance CEO
-
UK govt moves to block sentencing guidelines for minority defendants
-
Trump puts world on edge as 'Liberation Day' tariffs loom
-
Swedish journalist jailed in Turkey kept 'isolated': employer
-
Stock markets advance ahead of Trump tariffs deadline
-
Gulf between Everton and Liverpool has never been bigger, says Moyes
-
Finland to withdraw from anti-personnel mine ban treaty
-
UK vows £20 million to boost drone and 'flying taxi' services
-
Ford's US auto sales dip in first quarter as tariffs loom
-
Digging for box office gold, 'A Minecraft Movie' hits cinemas
-
Southampton boss Juric desperate to avoid Premier League 'worst team' tag
-
Thailand rescue dogs double as emotional support
-
Five takeaways from Marine Le Pen verdict
-
Stock markets split ahead of Trump tariffs deadline
-
Turkish fans, artists urge Muse to cancel Istanbul gig over protest dispute
-
Former captain Edwards named new England women's cricket coach
-
Haaland ruled out for up to seven weeks: Man City boss Guardiola
-
UK Supreme Court opens car loans hearing as banks risk huge bill

'AI doctor' better at predicting patient outcomes, including death
Artificial intelligence has proven itself useful in reading medical imaging and even shown it can pass doctors' licensing exams.
Now, a new AI tool has demonstrated the ability to read physicians' notes and accurately anticipate patients' risk of death, readmission to hospital, and other outcomes important to their care.
Designed by a team at NYU Grossman School of Medicine, the software is currently in use at the university's affiliated hospitals throughout New York, with the hope that it will become a standard part of health care.
A study on its predictive value was published Wednesday in the journal Nature.
Lead author Eric Oermann, an NYU neurosurgeon and computer scientist, told AFP that while non-AI predictive models have been around in medicine for a long time, they were hardly used in practice because the data they needed requires cumbersome reorganization and formatting.
But "one thing that's common in medicine everywhere, is physicians write notes about what they've seen in clinic, what they've discussed with patients," he said.
"So our basic insight was, can we start with medical notes as our source of data, and then build predictive models on top of it?"
The large language model, called NYUTron, was trained on millions of clinical notes from the health records of 387,000 people who received care within NYU Langone hospitals between January 2011 and May 2020.
These included any records written by doctors, such as patient progress notes, radiology reports and discharge instructions, resulting in a 4.1-billion-word corpus.
One of the key challenges for the software was interpreting the natural language that physicians write in, which varies greatly among individuals, including in the abbreviations they choose.
By looking back at records of what happened, researchers were able to calculate how often the software's predictions turned out to be accurate.
They also tested the tool in live environments, training it on the records from, for example, a hospital in Manhattan then seeing how it fared in a Brooklyn hospital, with different patient demographics.
- Not a substitute for humans -
Overall, NYUTron identified an unnerving 95 percent of people who died in hospital before they were discharged, and 80 percent of patients who would be readmitted within 30 days.
It outperformed most doctors on its predictions, as well as the non-AI computer models used today.
But, to the team's surprise, "the most senior physician who's actually a very famous physician, he had superhuman performance, better than the model," said Oermann.
"The sweet spot for technology and medicine isn't that it's going to always deliver necessarily superhuman results, but it's going to really bring up that baseline."
NYUTron also correctly estimated 79 percent of patients' actual length of stay, 87 percent of cases where patients were denied coverage by insurance, and 89 percent of cases where a patient's primary disease was accompanied by additional conditions.
AI will never be a substitute for the physician-patient relationship, said Oermann. Rather, they will help "provide more information for physicians seamlessly at the point-of-care so they can make more informed decisions."
C.Garcia--AMWN