All articles tagged with #mortality prediction
A recent study suggests that body fat percentage (BF%) measured via bioelectrical impedance analysis (BIA) is a more accurate predictor of 15-year all-cause and cardiovascular mortality than BMI, especially in younger adults, and advocates for wider adoption of BF% measurement in clinical practice.
Researchers have developed the metabolic dysbiosis score (MDS), a fecal biomarker that may predict 30-day mortality in critically ill patients by analyzing gut microbiome imbalances, though further validation is needed before clinical use.
A new AI system called FaceAge can predict biological age and mortality risk from a simple selfie, revealing that how old you look may be linked to your lifespan, raising ethical concerns about privacy and misuse of such technology.
At-home tests claiming to predict longevity based on physical abilities have gained popularity, but experts caution against relying on them. Tests like the sitting-rising test, one-legged stance test, and grip strength assessment have been linked to increased risk of mortality in some studies. However, experts warn that these tests may not provide a complete picture of overall health and aging, and caution against older adults attempting them without supervision due to the risk of falls and injury.
Researchers in Denmark have developed an AI "death calculator" called life2vec, which analyzes personal and medical data to predict a person's mortality timeline with 79% accuracy. Using a Danish public health registry dataset, the algorithm outperformed other tools at predicting early mortality, suggesting it could be a valuable tool for identifying key factors that affect life outcomes and enabling earlier interventions.
A pilot study at the University of Illinois Chicago suggests that dads can also develop postpartum depression and should be screened for it. The study found that a third of the 24 dads screened tested positive for postpartum depression or suicidal ideation. Additionally, a study in JAMA suggests that an individual's waist-to-hip ratio is a better predictor of mortality than body mass index (BMI). Researchers found that the waist-to-hip ratio had the strongest association with mortality, regardless of BMI. Furthermore, researchers from the UK found that common conditions such as high blood pressure, respiratory tract infection, heart conditions, and ear, nose, and throat infections precede the diagnosis of Type 2 diabetes, suggesting that an uptick in these conditions could prompt concern about diabetes and lead to earlier diagnosis. Lastly, a study in Neurology suggests that both high and low levels of high-density lipoprotein (HDL) cholesterol, the "good" kind, are associated with a small increased risk of dementia in older adults.
Researchers have developed a machine learning algorithm that can predict the risk of death within one month, one year, and five years of a patient being admitted to the hospital with an 85% accuracy rate, using ECG data and demographic information. The algorithm sorts patients into five categories from lowest to highest risk. The study is a proof-of-concept for using routinely collected data to improve individual care and allow the health-care system to "learn" as it goes.