'Instant' blood test for heart attacks

BBC News

A blood test that could rule out a heart attack in under 20 minutes should be used routinely, say UK researchers. A team from King's College London have tested it on patients and say the cMyC test could be rolled out on the NHS within five years. They claim it would save the health service millions of pounds each year by freeing up beds and sending well patients home. About two-thirds of patients with chest pain will not have had a heart attack. A heart trace, called an ECG, can quickly show up major heart attacks, but it is not very good at excluding more common, smaller ones that can still be life-threatening.


Measuring the Stability of EHR- and EKG-based Predictive Models

arXiv.org Machine Learning

Databases of electronic health records (EHRs) are increasingly used to inform clinical decisions. Machine learning methods can find patterns in EHRs that are predictive of future adverse outcomes. However, statistical models may be built upon patterns of health-seeking behavior that vary across patient subpopulations, leading to poor predictive performance when training on one patient population and predicting on another. This note proposes two tests to better measure and understand model generalization. We use these tests to compare models derived from two data sources: (i) historical medical records, and (ii) electrocardiogram (EKG) waveforms. In a predictive task, we show that EKG-based models can be more stable than EHR-based models across different patient populations.


The Challenge of Imputation in Explainable Artificial Intelligence Models

arXiv.org Artificial Intelligence

Even though the field of Artificial Intelligence is more than sixty years old, it is only in the last decade or so that AI systems are being increasingly interwoven into the fabric of the socio-technical apparatus of the society and are thus having a massive impact on society. This increasing incorporation of AI has led to increased calls for accountability and regulation of AI systems [8]. Model explanations are considered to be one of the most important ways to provide accountability of AI systems. The model explanations, however, can only be as good as the data on which the algorithms are based. This is where the issue of missing and imputed data becomes pivotal for model explanations. In some domains like healthcare, almost all datasets have missing values [6]. As many applications of AI in healthcare are patient-oriented, decisions that are informed by AI and ML models can potentially have significant clinical consequences.


DIY or die: How an Australian outback nurse diagnosed his own heart attack and saved his life

Los Angeles Times

He notes that patients living within a two-hour drive to a hospital will more likely have their electrocardiogram read while they're en route in an ambulance, and have their plaques removed by a clot-retrieving device rather than dissolved with tenecteplase. Even in remote parts of the United States, emergency medical technicians, emergency physicians and cardiologists are devising ways to speed the clearance of coronary arteries blocked by clots and plaques.


Nurse Suffers Heart Attack In Hospital, Diagnoses And Treats Self

International Business Times

An Australian nurse who sensed he was having a heart attack took necessary measures to save his own life. The 44-year-old man, whose name has been withheld for security reasons, was the only medical professional on duty at a nursing post in Coral Bay, Western Australia, when he started experiencing chest pain and dizziness, South China Morning Post reported. The first thing the man did was perform an electrocardiogram (EKG), which revealed that the male nurse had a complete heart blockage – the result of a developing heart condition. A second EKG confirmed the fact that the man was indeed suffering from a heart attack. Using the Emergency Telehealth Service (ETS), he emailed the results to a doctor and connected to an emergency physician who was willing to help him out.