Norwegian researchers are facing a potentially small revolution.
At the University Hospital Hoekland in Bergen, they have developed a method to be able to diagnose tumors with much greater accuracy than is possible today.
Researchers believe that using artificial intelligence could lead to faster and more accurate treatment.
– It’s information that is difficult for the human eye to perceive, but is then retrieved with the help of a computer program, says mathematician Erlind Hodiland.
– Results are tabulated and evaluated against a database that tells how much risk the patient is exposed to, he continues.
Revolution on the stairs
Nina Hirvendahl recently had a long battle against cancer.
After diagnosis, treatment, and rehabilitation, she is now cancer-free, but is undergoing a final medical examination.
She is very tired and struggles a lot to imitate having a little embroidery during the day.
– DrIt’s best to be cohesive to keep the limbs moving, Hirfindahl says.
The image of the tumor in her womb formed the basis of the treatment she received.
– Here we see a uterus with a large tumor that lights up. Here is a small tumor that spreads to the lymph node at the bottom of the pelvis, says OThe words of Professor John Trovik and pointers to the luminous fields.
Their radiologist usually sits down and studies a number of different images of the patient to see where the cancer is and whether it has spread.
Trovik says it takes time and is labour-intensive.
But the hospital’s researchers are now facing what could be a small revolution: being able to diagnose tumors faster and more accurately than is possible today with the help of artificial intelligence.
– nQ Can we get information from the image if the tumor is particularly aggressive. We can’t just see this from this image, because it’s digital information that must be interpreted mechanically according to very advanced mathematical models, she says.
Then the radiologists can get help to see if there is anything unusual about the tumor. Then they can review it themselves and confirm it, Trovik says.
More detailed treatment
To make this happen, computer experts, doctors, and other specialists at the Mohn Center for Medical Imaging and Visualization at Hauckland University Hospital collaborated.
Mathematician Erlend Hodneland and molecular biologist Erling Høivik are among the authors. Research Report which recently appeared in the acclaimed medical journal Communications Biology.
It could lead to more tailored treatment, gentler treatment for those who don’t need such extensive treatment and more comprehensive treatment for those who don’t want to survive their disease, says research leader Ingfried Haldorsen.
The researchers think they’ve taken a huge step toward machines being able to interpret data that humans can’t or don’t have time for.
“In the long run, we can get to the point where the machine first maps the tumor, then quickly analyzes the characteristics of the tumor and tells you that patient is at risk of relapse,” Haldorsen says.
– It can refer to the treatment that the patient should ideally receive, without the need for a biopsy.
“Web specialist. Lifelong zombie maven. Coffee ninja. Hipster-friendly analyst.”