Machine learning detects, treats UTIs earlier
Scientists at the University of Surrey in Guildford, England, have developed a tool that uses machine learning to identify and treat urinary tract infections at early stages in dementia patients, according to a study published in PLOS One.
"I am delighted to see that the algorithms we have designed have an impact on improving the healthcare of people with dementia and providing a tool for clinicians to offer better support to their patients,” Shirin Enshaeifar, PhD, study author and senior research fellow at University of Surrey’s Center for Vision, Speech and Signal Processing (CVSSP), said in a statement.
For the study, scientists remotely monitored the health dementia patients at home using a network of internet-enabled and vital body signal monitoring devices. They designed a UTI detection machine-learning algorithm using a a Non-negative Matrix Factorisation (NMF) technique to analyze the data streamed from the monitoring devices. The NMF model was then compared to a another UTI-detection solution that uses a binary support vector machine (SVM) classifier.
“If a UTI is detected at early stages, it can be resolved by taking antibiotics; however, remaining undiagnosed, a UTI can cause major health issues resulting in hospital admissions,” Enshaeifar et al. wrote.
Data was collected from a total of 53 participants. The results showed the NMF model outperformed the SVM model and reduced the number of false positive alerts, according to the study. Researchers were also to design an algorithm, using the Isolation Forest technique, that detected changes in activity patterns and identified early signs of cognitive or health decline, which were later followed up by a clinical monitoring team.
“We are confident our algorithm will be a valuable tool for healthcare professionals, allowing them to produce more effective and [personalized] plans for patients,” Payam Barnaghi, professor of machine intelligence at CVSSP, said in a prepared statement.
About 50 million people worldwide currently have dementia, and that number is expected by reach 82 million by 2030 and 152 million by 2050, according to the study. One in four hospital beds in the United Kingdom are occupied by dementia patients, with UTIs being one of the top causes of hospitalizations.
"Machine learning could provide improved care for people living with dementia to remain at home, reducing hospitalization and helping the NHS to free up bed space,” Adrian Hilton, professor and director of the CVSSP, said in a statement.
Danielle covers Clinical Innovation & Technology as a senior news writer for TriMed Media. Previously, she worked as a news reporter in northeast Missouri and earned a journalism degree from the University of Illinois at Urbana-Champaign. She's also a huge fan of the Chicago Cubs, Bears and Bulls.