Research

Smartphone Application For Effective Prostate Cancer Screening With Machine Learning Enhanced PSA-density Measurement

Principal investigator: Zion TSE
Prostate cancer (PCa) is the most type of cancer amongst men. Prostate-specific antigen (PSA) testing is the first-line investigation used for referral to secondary care. Less than half of the 120,000 patients each year referred in the UK are ultimately diagnosed with PCa, highlighting the inefficiencies in the system. These inefficiencies include the use of MRI as an expensive resource and biopsy as an invasive procedure. A common reason for raised PSA levels is benign gland overgrowth, and therefore PSA-density corrects for overgrowing gland volume, and therefore has utility for and indicating the presence of clinically significant cancer. Ultrasound (US) can measure gland volume provide such information. However, this US is currently performed in secondary care by specialised practitioners, which increases costs and may delay cancer treatment pathways. Making US volume calculations automated, cheap, and potentially available in primary care would address such limitations. The aims of this project are to develop a prototype device for automated US measurement of prostate volume and validate performance in a patient cohort.