A blood test which doesn’t require drawing any blood is being submitted for approval to the U.S. Food and Drug Administration (FDA).
Scientists at the Mayo Clinic, a non-profit medical practice and research group in Rochester, Minnesota, have developed an algorithm which can determine serum potassium levels in the blood by monitoring small changes in electrocardiogram (ECG) waves.
Fluctuations in potassium levels can lead to life-threatening irregular heartbeats or sudden death and can be caused by the medication used to treat patients with high risk: those with heart and kidney disease.
“Detecting abnormal blood potassium levels earlier would allow medicines to be more effectively titrated, problems to be addressed before they occurred, hospitalizations to be prevented, and costs to be reduced.”
They hypothesized that ECG signals could serve as a full range of blood potassium levels and developed their algorithm based on data containing concurrent ECG and potassium measurements from more than 2 million Mayo Clinic patients.
In order to assess and validate how the algorithm worked, Mayo Clinic researchers carried out a supervised learning process where human experts assessed changes in ECG features related to potassium during dialysis.
Mayo Clinic validated their approach in a series of clinical studies and found the potassium scores their algorithm generated closely correlated with serial blood tests taken from the same patients.
They then partnered with AliveKor, an AI driven meditech company specialising in smart phone and apple watch-based ECG readers.
The bloodless blood test technology could now leverage AliveCor’s devices to capture ECG data from patients in their homes and use these data to predict blood potassium levels.
Article extracted from News section of AIMed Magazine. Access Full Magazine Free Here.