All articles tagged with #predictive technology
Researchers at the University of Surrey have developed an AI system that can predict future knee X-ray images and assess disease risk, potentially transforming osteoarthritis treatment by enabling earlier and personalized care. The system, trained on a large dataset, outperforms existing tools in speed and efficiency and could lead to similar innovations for other chronic diseases.
Researchers at Los Alamos have identified electron accumulation as a key factor in satellite failures, with a 75% correlation between electron peaks and electrical discharges, suggesting that continuous electron monitoring could help predict and prevent satellite malfunctions in the future.
Researchers have conducted a study using artificial intelligence (AI) to analyze speech patterns and predict intoxication levels with 98% accuracy. Participants were given alcohol and asked to recite tongue twisters while their breath alcohol levels were recorded. The AI program was able to determine if someone was within legal sobriety limits, suggesting potential applications such as ignition locks on cars and managing alcohol consumption in bars and restaurants. However, the study had a small sample size and lacked diversity, highlighting the need for further testing.
A study conducted by the University of Oxford has found that artificial intelligence (AI) can predict a person's risk of having a heart attack up to 10 years in advance. By improving the accuracy of cardiac CT scans, AI technology could potentially save lives and enhance treatment for nearly half of all patients. The study, funded by the British Heart Foundation, analyzed data from over 40,000 patients and found that AI-generated risk scores accurately predicted the likelihood of a heart attack. The implementation of this AI tool across the NHS could help prevent thousands of avoidable deaths from heart attacks each year in the UK.
A new smartphone app called HearO uses AI to analyze changes in users' voices and predict the risk of hospitalization due to heart failure. By detecting subtle changes that indicate increasing fluid in the lungs, the app was able to predict deterioration and the need for hospitalization three weeks in advance with 76% accuracy. This advanced warning allows doctors to start preventative treatment and potentially save lives. The app has received Breakthrough Device Designation and is on track for FDA approval by 2024.
The Centers for Medicare and Medicaid Services (CMS) will implement new rules next year to restrict the use of predictive technology tools by Medicare Advantage plans in making coverage decisions. These tools, such as UnitedHealthcare's NaviHealth, use algorithms to predict patients' care needs and discharge dates. However, patients, providers, and advocates have raised concerns that the predictions often align with the date coverage is cut off, even if further treatment is necessary. The new rules aim to ensure that coverage decisions are based on individual circumstances and reviewed by healthcare professionals, rather than relying solely on algorithms. Critics question how the rules will be enforced, while CMS states that audits and enforcement actions may be taken for noncompliance.