Abstract: A machine learning algorithm can be trained to process signals received from a non-invasive sensor to determine an analyte. The machine learning algorithm can be a neural network trained using data obtained from the non-invasive sensor and associated analyte detection results from another sensor. The features of the data can be smoothed using a Savitzky-Golay filter, scaled, and the feature space can also be reduced, such as by Gaussian random projection. The training of the algorithm can be tested using other data obtained from the non-invasive sensor and associated analyte detection results from another sensor. Once trained, the machine learning model is used to determine levels of an analyte from signals received at a non-invasive sensor in response to transmission of a transmit signal into a subject.

Abstract: A highly accurate and safe analyte sensor that detects an analyte by transmitting and receiving sensing signals in a radio or microwave frequency range of the electromagnetic spectrum is provided. The analyte sensor has at least two antennas at least one of which operates as a transmit antenna to transmit one or more of the sensing signals and at least one of which operates as a receive antenna, and the analyte sensor meets or exceeds performance parameters for integrated continuous glucose monitoring systems as set forth in 21 C.F.R. 862.1355 (Feb. 18, 2022) and is configured to emit radiofrequency radiation less than radiofrequency radiation exposure limits as set forth in 47 C.F.R. 1.1310 (Apr. 1, 2020).

Abstract: A non-invasive analyte sensor includes one or more motion sensors and/or one or more temperature sensors. Data from the one or more motion sensors and/or one or more temperature sensors can be used to post-process the data obtained by the non-invasive analyte sensor.

Abstract: A watch comprising a deadfront window; a first OLED display disposed beneath the deadfront window such that the first OLED display can be observed through the deadfront window when the first OLED display is active; a movement for tracking time, the movement having a first portion that resides beneath the deadfront window and a second portion that protrudes through the deadfront window; and time-indicating members disposed above the deadfront window, the time indicating members being secured to the second portion of the movement.