CANCER
Representational picture REUTERS/Stefan Wermuth

A new study suggested that two machine learning (ML) models Support Vector Regression (SVR) and Non-linear Canonical Correlation Analysis by Neural Networks (n-CCA) are able to predict the severity of three common symptoms faced by cancer patients such as depression, anxiety and sleep disturbance.

The researchers believe that these type of models can be used to identify high-risk patients, educate them about the symptom and the experience. It can also improve the timing of preemptive and personalised symptom management interventions.

Researchers from the University of Surrey in UK said that these three symptoms are associated with a severe reduction in cancer patients' quality of life.

In addition, Professor from the varsity, Payam Barnaghi said, "These exciting results show that there is an opportunity for machine learning techniques to make a real difference in the lives of people living with cancer."

"They can help clinicians identify high-risk patients, help and support their symptom experience and preemptively plan a way to manage those symptoms and improve quality of life," he further added.

The new study, which is published in the journal PLOS One, stated the team of the researchers analysed the data of the symptoms experienced by over 2000 cancer patients during the computed tomography x-ray treatment. It also stated that they used different periods during this data to test whether the ML method can predict when and if symptoms surfaced with accuracy or not. But, the team discovered that the actual reported symptoms were very close to those predicted by ML method.

Experts said that in among all the cases, between 35 and 53 percent of patients face depression, while 30 to 50 percent of the patients suffer from sleep disturbance. The data also showed that between 35 percent and 53 percent of the patients told about the anxiety they faced during cancer treatment.