News summary produced by Claude AI
The NHS is introducing a new artificial intelligence-based blood test called PinPoint that could reduce the need for invasive diagnostic procedures in women suspected of having womb cancer. Around 90,000 postmenopausal women are referred annually in England for investigation of possible uterine cancer due to heavy bleeding, with 10,000 diagnosed and 2,700 dying from the disease each year.
The test demonstrated a 99% accuracy rate in detecting and ruling out gynaecological cancers in a trial involving 3,313 women referred for suspected womb cancer. Based on these results, researchers estimate the test could prevent approximately 18,000 women annually from undergoing transvaginal ultrasound scans and related procedures that many find uncomfortable or painful. The technology uses machine learning to analyze 30 blood markers and classify patients as having low, elevated, or high cancer risk.
The trial included 16,481 patients referred from 170 general practices across Yorkshire for nine different cancer types. Mid Yorkshire NHS Teaching Trust and Leeds Teaching Hospitals NHS Trust have announced plans to implement the test for detecting gynaecological and upper gastrointestinal cancers. Healthcare professionals note that currently, women referred with suspected cancer undergo multiple procedures, with most ultimately found to be cancer-free.
Developed by Leeds-based firm PinPoint Data Science, the test aims to streamline the diagnostic process and reduce unnecessary invasive procedures. Medical professionals highlight that the test could expedite diagnosis for women who do have cancer while sparing low-risk patients from invasive examinations. Cancer Research UK stated the results appear promising but noted that additional research is needed to fully understand the benefits for patients and the healthcare system.