Back

Novel data preprocessing techniques in an expanded dataset improve machine learning model accuracy for a non-invasive blood glucose monitor

Klyve, D.; Pandya, K.; Ward, C.; Shelton, B.

2023-07-25 endocrinology
10.1101/2023.07.24.23293113 medRxiv
Show abstract

To determine the accuracy of a novel sensor designed to measure blood glucose (BG) non-invasively using Radio Frequency (RF) waves, we present results from a study that validates the stability of a machine learning model on an expanded dataset. In this study, we trained a Light Gradient-Boosting Machine (lightGBM) model to predict BG values using 3,311 observations from over 330 hours of data collected from 13 healthy participants, where an observation is defined as data collected from 13 sweeps from the novel Bio-RFID sensor paired with a single Dexcom G6(R) value as reference.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.