Machine Learning to Predict Future Symptoms of Oncology Patients

Machine Learning to Predict Future Symptoms of Oncology Patients

Researchers from University of Surrey developed an Artificial Intelligence that can predict symptoms of cancer and their severity

Effective management of symptom is a critical component of cancer treatment. Now a team of researchers from the Centre for Vision, Speech and Signal Processing (CVSSP) at the University of Surrey developed two machine learning models that can precisely foresee the severity of three common symptoms — depression, anxiety, and sleep disturbance — faced by cancer patients. These symptoms are associated with severe reduction in cancer patients’ quality of life.

The team analyzed existing data of the symptoms experienced by cancer patients during the course of computed tomography X-ray treatment. To test whether the machine learning algorithms are able to accurately predict when and if symptoms surfaced, the team used different time periods during this data analysis. The results found that the actual reported symptoms were closely similar to those predicted by the machine learning methods. According to Payam Barnaghi, Professor of Machine Intelligence at the University of Surrey, the exciting results demonstrate that there is an opportunity for machine Artificial Intelligence (AI) in enhancing the lives of people suffering from cancer. AI can aid clinicians to identify high-risk patients, assist and support their symptom experience and pre-emptively plan an approach to manage the symptoms in order to improve quality of life.

Nikos Papachristou, who worked on designing AI algorithms for this project stated that the prospect of use of machine learning and AI to create solutions is revolutionary and can have a positive influence on the quality of life of patients living with various types of cancer. The research was conducted in collaboration between the University of Surrey and the University of California in San Francisco (UCSF). The UCSF research in the joint collaboration was led by Professor Christine Miaskowski. The findings were published in the journal PLOS ONE on December 31, 2018.

You May Also Like