Medical researchers have created a new AI model that could potentially help some cancer patients get treated.
Nearly 10 million people died from cancer in 2020, according to the World Health Organisation. An estimated 20 million new cancer cases were recorded in 2022. Even worse, the WHO predicts that the burden will increase by 77% from the 2022 record to over 35 million new cases in 2050.
Immunotherapy with AI
Even as deadly as it appears, there is unfortunately no “single” cure for cancer. However, the disease is managed through some effective treatments, especially chemotherapy, which is a drug treatment that uses powerful chemicals to kill fast-growing cells in the body.
Immunotherapy is another alternative that has shown promise in treating cancer. It harnesses the body’s own immune system to recognize and combat cancer cells. But the treatment is not all that perfect.
It reportedly does not work for every patient, and some 10% to 15% of the people who undergo the treatment develop “significant toxicities.”
However, it’s deemed “life-saving” for the people who successfully undergo the treatment. So, the researchers intend to use AI to maximize the results by predicting who immunotherapy will work for and those who could develop significant toxicities from the treatment.
Researchers Say the AI Model Has a “Very Good Result”
“The model predicts which patients are likely to derive the benefit from immunotherapy versus those patients who may not,” said Jan Wolber, global digital product leader at GE HealthCare’s pharmaceutical diagnostics segment.
The AI model has been under development by GE HealthCare and Vanderbilt University Medical Center (VUMC) for the past five years. It is trained on thousands of patients’ electronic health records (EHRs), Fox News reported Tuesday.
The model looks at patients’ data, such as demographic information, preexisting diagnoses, medication history, lifestyle habits, and so on, to recognize patterns in how they will respond to immunotherapy treatment.
Early studies found that the AI model has 70% to 80% accuracy in predicting patients’ responses to immunotherapies.
“While the models are not perfect, this is actually a very good result,” Wolber said. “We can implement those models with very little additional effort because there are no additional measurements required in the clinic.”
