Now, oncologists (medical practitioners specialized in treating cancer) could get a head start using AI that is able to predict symptoms and their severity throughout the course of a patient’s treatment. Researchers at the University of Surrey in collaboration with the University of California San Francisco (UCSF) have developed Artificial Intelligence, able to predict symptoms experienced by cancer patients and how severe they could be through the course of treatment.
The researchers have developed two machine learning models that have the ability to accurately predict the severity of three common symptoms in cancer patients: depression, sleep disturbance and anxiety. Effective symptom management is crucial in cancer treatment because bringing them under control could improve the quality of a patient’s life. Many patients who undergo computed tomography x-ray treatment experience anxiety, depression and sleep interruption during the course of treatment. Researchers analyzed this data (in different time periods) to test whether the machine learning algorithms are able to accurately predict whether the symptoms have surfaced and when. The results derived by the analysts from the tests were very close to those predicted by the machine learning methods.
The machine learning technique enables oncology clinicians to identify high risk patients, extend support to patients during the in-symptom experience and formulate tactical treatment plans. Believed to be the first study of its kind, this research was published in the PLOS One journal. The team of researchers was led by Payam Barnaghi, professor of machine learning intelligence, from the Center for Vision, Speech and Signal Processing (CVSSP). Adrian Hilton, Director of CVSSP says, “These exciting developments by Professor Barnaghi and his team show the incredible potential of machine learning in transforming the way healthcare professionals treat people suffering from cancer.” The University of California San Francisco’s team was led by Professor Christine Miaskowski, Department of Physiological Nursing, UCSF.
These type of predictive learning models can be employed to educate patients about their symptom experience and improve the timing of pre-emptive and personalized symptom management interventions.