Utilising artificial intelligence (AI), Watson has been able to accurately detect, diagnose, and even predict illness, at a rate faster than scientists can ever achieve.
How did they do it?IBM started off by gathered scientific and research papers, clinical trials, genome sequences and other forms of healthcare data that they could attain. This was certainly a mammoth feat – with an estimated 1.8 million scientific papers published every year.
In addition to publications, Watson is stored with patient data, with an average individual generating over one million gigabytes of healthcare information in their lifetime – a load equivalent to 300 million books.
“In the healthcare system, there is three million times more data than all books ever written,” said president and CEO of IBM, Ginni Rometty. “So what’s a doctor going to do?”
Watson, through its computing power, is able to ‘speed-read’ the unstructured data – up to 200 million pages of text in 3 seconds – cross-reference the information, understand and even learn from it, in order to make better, more accurate diagnoses.
But that was not all.
In 2011, Watson was sent to medical school, where doctors from top schools worldwide spent two years teaching it about different diseases, testing the AI and correcting its mistakes. Watson graduated with the ability to diagnose patients with high accuracy and speed, and was also capable of looking for alternative treatments that were overlooked by doctors by 30% of cases.
Making a live-saving diagnosis – in ten minutesWatson was put to the test, when a Japanese woman diagnosed with leukaemia underwent a standard treatment of chemotherapy, however, without any clinical improvements. Baffled, doctors enlisted Watson to assist in their treatment approach for a solution.
Watson quickly sifted through the same information the doctors had, compared it with 20 million other research papers on cancer, and was able to provide a more accurate diagnosis – in a mere ten minutes.
“The team found she had another type of leukemia [that needed] a different therapy,” said Satoru Miyano, a professor at the Institute of Medical Science at the University of Tokyo.
“She got it and she recovered completely.”
“We would have arrived at the same conclusion by manually going through the data, but Watson’s speed is crucial in the treatment of leukemia, which progresses rapidly and can cause complications,” added Professor Arinobu Tojo, who led the research team.
“It might be an exaggeration to say AI saved her life, but it surely gave us the data we needed in an extremely speedy fashion.”
This particular incident has once again raised concerns over the dangers of medical errors caused by doctors, which is estimated to be the third leading cause of death in the United States.
With the use of AI like Watson, it is hoped that such medical errors can be prevented.
Globalisation of healthcare information for better patient careHaving graduated from medical school, Watson Health is currently focusing efforts to assist in cancer research, utilising the bulk of data that it has for researchers to start looking for potential cures.
IBM is also collaborating with a biopharmaceutical company called Celgene, which has extensive data on clinical drug trials, with aims to reduce unnecessary testing, and scientists at the Undiagnosed and Rare Diseases Centre in Germany are using Watson to gather data on rare diseases.
Watson has already showed promising possibilities of reshaping healthcare and transcend borders to enable doctors worldwide to utilise the wealth of information to assist in aspects of patient’s diagnosis or treatment.
As said by the research manager for IBM Health, Spyros Kotoulas, “We are moving from treating a set of problems to treating a person.” MIMS
AI Watson and other nanorobots to help ‘detective doctors’ solve medical mysteries
Is AI the solution to decipher 2.5 million scientific papers annually?
The artificially-intelligent doctor is still a dream – for now
The battle of doctors vs. machines in clinical diagnoses