Predicting (and Reducing) Cancer-Treatment Side Effects

According to the American Cancer Society, about 1 in 9 men will be diagnosed with prostate cancer during their lifetime; and prostate cancer is the most common noncutaneous malignancy in men.

In a recent article published in the Journal of the American Medical Informatics Association, informaticians discussed using Natural Language Processes (NLP) techniques to parse millions of free text clinical notes on prostate cancer treatment-related side effects that alter a patient’s quality of life, such as sexual, urinary and bowel dysfunction.

What did they discover? NLP methods can convert prostate cancer treatment-related data into a structured representation, which offers an opportunity to train models to automatically predict these outcomes for future patients.

Why informatics? So, we can reduce the side effects for men when they are being treated for prostate cancer.