International Edition
Latest News
Health

AI Detects Bladder Cancer Risk Up to Five Years Before Diagnosis, Study Finds

Researchers at the University of Plymouth have developed an artificial intelligence system that identifies bladder cancer risk years before clinical diagnosis, outperforming current NHS referral guidelines in a study of nearly 70,000 patients. The model, named PRECISE-AGZ, analyzed…

AI Detects Bladder Cancer Risk Up to Five Years Before Diagnosis, Study Finds

Researchers at the University of Plymouth have developed an artificial intelligence system that identifies bladder cancer risk years before clinical diagnosis, outperforming current NHS referral guidelines in a study of nearly 70,000 patients.

The model, named PRECISE-AGZ, analyzed electronic health records spanning 25 years from the SAIL database in Wales. It correctly identified 85% of bladder cancer cases and accurately classified 91% of patients without the disease, according to findings published in IEEE Transactions on Biomedical Engineering.

Bladder cancer ranks as the ninth most common cancer worldwide, with approximately 220,000 deaths annually. No routine screening program exists for the general population, and detection relies heavily on visible blood in the urine (hematuria) — a symptom that overlaps with benign conditions such as kidney stones and prostate enlargement.

“Bladder cancer has a major unmet need in early detection and the prospect of a screening tool is hugely exciting for patients and their families,” said Dr. Helen Winter, Clinical Director for the Somerset, Wiltshire, Avon and Gloucestershire (SWAG) Cancer Alliance. “It could identify those most at risk more accurately, reduce harm from invasive procedures, and catch cancer earlier, when survival outcomes and quality of life can be dramatically improved.”

How the AI Model Works

The research team, led by Professor Shang-Ming Zhou at the Centre for Health Technology, examined 48,261 possible health indicators from patient records covering 1995 through 2020. The system narrowed these to 38 features most predictive of bladder cancer risk.

Ai model detects bladder cancer warning signs five years before diagnosis
Photo: technewsreel.com

Beyond established risk factors like smoking and hematuria, the model uncovered unexpected associations. Patients diagnosed with Parkinson’s disease or dementia appeared to have lower bladder cancer risk, while long-term tamoxifen use — a breast cancer medication — was associated with higher risk. The researchers emphasized these findings do not establish causation but warrant further investigation into shared biological mechanisms.

The significance of hematuria also varied by patient context.

Three-Tier Risk Stratification

PRECISE-AGZ sorts patients into three risk categories: low-risk (below 7% probability), uncertain or “grey zone” (7–55%), and high-risk (above 55%). This approach could help clinicians prioritize patients for cystoscopy — the invasive gold-standard diagnostic procedure — while sparing low-risk individuals unnecessary examinations.

“This work represents a paradigm shift towards precision screening for bladder cancer,” Professor Zhou said. “By harnessing the power of interpretable machine learning and comprehensive health records, we’re moving closer to detecting this disease at its earliest, most treatable stages.”

Validation Needed Before Clinical Use

The researchers caution that the model is not ready for widespread deployment. All training data came from the SAIL database in Wales, and performance must be validated across different healthcare systems and patient populations before clinical adoption.

“It’s important to emphasize that further validation across different healthcare systems is essential before anything is rolled out more widely,” Professor Zhou said. “But it’s a very exciting early study.”

Additional studies are also needed to confirm whether the associations identified by the model reflect causal relationships. The research involved collaborators from North Bristol NHS Trust and University Hospitals Bristol NHS Foundation Trust.

What This Means for Patients

If validated in broader populations, AI-driven risk stratification could transform bladder cancer detection from reactive symptom-chasing to proactive screening. Earlier diagnosis typically allows less aggressive treatment, better preservation of bladder function, and improved survival rates.

Founded in 1862 as the Plymouth School of Navigation, the University of Plymouth traces its origins to maritime education before evolving through polytechnic status to become a university in 1992. Today it serves 18,410 students from its main campus in Plymouth, England, and affiliated sites across South West England.

PhD student Xu Wang, who led the data analysis, said: “Our system goes far beyond traditional symptoms, and it’s great to have uncovered the results that we have. The research shows that the combination of multiple factors — from medication histories to seemingly unrelated conditions — can reveal patterns invisible to conventional screening methods.”

About the author: Health Desk

Health Desk at Archynewsy.