Emerging Applications for Intelligent Diabetes Management

AAAI Conferences

Diabetes management is a difficult task for patients, who must monitor and control their blood glucose levels in order to avoid serious diabetic complications. It is a difficult task for physicians, who must manually interpret large volumes of blood glucose data to tailor therapy to the needs of each patient. This paper describes three emerging applications that employ AI to ease this task and shares difficulties encountered in transitioning AI technology from university researchers to patients and physicians.


Emerging Applications for Intelligent Diabetes Management

AI Magazine

It is a difficult task for physicians, who must manually interpret large volumes of blood glucose data to tailor therapy to the needs of each patient. This paper describes three emerging applications that employ AI to ease this task: (1) case-based decision support for diabetes management; (2) machine-learning classification of blood glucose plots; and (3) support vector regression for blood glucose prediction. The first application provides decision support by detecting blood glucose control problems and recommending therapeutic adjustments to correct them. The second provides an automated screen for excessive glycemic variability. The third aims to build a hypoglycemia predictor that could alert patients to dangerously low blood glucose levels in time to take preventive action.


[Report] β-cell–mimetic designer cells provide closed-loop glycemic control

Science

Chronically deregulated blood-glucose concentrations in diabetes mellitus result from a loss of pancreatic insulin-producing β cells (type 1 diabetes, T1D) or from impaired insulin sensitivity of body cells and glucose-stimulated insulin release (type 2 diabetes, T2D). Here, we show that therapeutically applicable β-cell–mimetic designer cells can be established by minimal engineering of human cells. We achieved glucose responsiveness by a synthetic circuit that couples glycolysis-mediated calcium entry to an excitation-transcription system controlling therapeutic transgene expression. Implanted circuit-carrying cells corrected insulin deficiency and self-sufficiently abolished persistent hyperglycemia in T1D mice. Similarly, glucose-inducible glucagon-like peptide 1 transcription improved endogenous glucose-stimulated insulin release and glucose tolerance in T2D mice. These systems may enable a combination of diagnosis and treatment for diabetes mellitus therapy.


Worried pasta will make you fat? Spaghettaboutit.

Popular Science

The cream-soaked, free-bread-basket excesses of fancy "Italian" dining of the past few decades have given pasta a bad name--one that Big Pasta has invested serious money into trying to expunge. Pasta is a processed carbohydrate, a category that also encompasses such dietary villains as white bread, canned fruit, and anything containing added sugar. But pasta differs from these others in that the way it's prepared actually slows down the time it takes to digest it, meaning that unlike Wonderbread, it has relatively low glycemic index (GI). A study published today in the journal BMJ Open suggests that pasta may have a place in a low-GI diet. Researchers at Toronto's St. Michael's Hospital looked at data gathered from 32 studies that analyzed the dietary effects of pasta in low-GI diets compared with the health effects of higher-GI diets.


Improving Heart Rate Variability Measurements from Consumer Smartwatches with Machine Learning

arXiv.org Machine Learning

The reactions of the human body to physical exercise, psychophysiological stress and heart diseases are reflected in heart rate variability (HRV). Thus, continuous monitoring of HRV can contribute to determining and predicting issues in well-being and mental health. HRV can be measured in everyday life by consumer wearable devices such as smartwatches which are easily accessible and affordable. However, they are arguably accurate due to the stability of the sensor. We hypothesize a systematic error which is related to the wearer movement. Our evidence builds upon explanatory and predictive modeling: we find a statistically significant correlation between error in HRV measurements and the wearer movement. We show that this error can be minimized by bringing into context additional available sensor information, such as accelerometer data. This work demonstrates our research-in-progress on how neural learning can minimize the error of such smartwatch HRV measurements.