Abstract: One or more electronic device may use motion and/or activity sensors to estimate a user's 6 minute walking distance. In particular, because users typically walk at less than their maximum output and in imperfect conditions, control circuitry within the device(s) may rely on walks of shorter distances to estimate the 6 minute walking distance. For example, the control circuitry may gather activity information for the user, such as heart rate, calories burned, and step count, and analyze a distance component and a speed component for periods in which the user has walked. Individual 6 minute walk distance estimates may be generated based on each of the activity information, distance component, and speed component. The distance and speed estimates may be corrected for walking behaviors that deviate from an ideal testing environment, and may then be fused with the activity estimate to generate a final 6 minute walk distance estimate.

Abstract: Systems and methods that analyze blood-based cancer diagnostic tests using multiple classes of molecules are described. The system uses machine learning (ML) to analyze multiple analytes, for example cell-free DNA, cell-free microRNA, and circulating proteins, from a biological sample. The system can use multiple assays, e.g., whole-genome sequencing, whole-genome bisulfite sequencing or EM-seq, small-RNA sequencing, and quantitative immunoassay. This can increase the sensitivity and specificity of diagnostics by exploiting independent information between signals. During operation, the system receives a biological sample, and separates a plurality of molecule classes from the sample. For a plurality of assays, the system identifies feature sets to input to a machine learning model. The system performs an assay on each molecule class and forms a feature vector from the measured values.

Abstract: Embodiments are disclosed for estimating heart rate recovery (HRR) after maximum or high-exertion activity based on sensor observations. In some embodiments, a method comprises: obtaining, with at least one processor, sensor data from a wearable device worn on a wrist of a user; obtaining, with the at least one processor, a heart rate (HR) of the user; identifying, with the at least one processor, an observation window of the sensor data and HR; estimating, with the at least one processor during the observation window, input features for estimating maximum or near maximum exertion HRR of the user based on the sensor data and HR; and estimating, with the at least one processor during the observation window, the maximum or near maximum exertion HRR of the user based on a machine learning model and the input features.

Abstract: Detecting a physiological parameter of a user at a first level during a first activity and at a second level during a second activity and displaying, based on the first level and the second level, a predictive change in the physiological parameter had the second activity been a third activity that is different from the second activity.

Abstract: Systems and methods that analyze blood-based cancer diagnostic tests using multiple classes of molecules are described. The system uses machine learning (ML) to analyze multiple analytes, for example cell-free DNA, cell-free microRNA, and circulating proteins, from a biological sample. The system can use multiple assays, e.g., whole-genome sequencing, whole-genome bisulfite sequencing or EM-seq, small-RNA sequencing, and quantitative immunoassay. This can increase the sensitivity and specificity of diagnostics by exploiting independent information between signals. During operation, the system receives a biological sample, and separates a plurality of molecule classes from the sample. For a plurality of assays, the system identifies feature sets to input to a machine learning model. The system performs an assay on each molecule class and forms a feature vector from the measured values.

Abstract: Embodiments are disclosed for identifying poor cardio metabolic health using sensors of wearable devices. In an embodiment, a method comprises: obtaining estimates of maximal oxygen consumption of a user during exercise; determining at least one confidence weight based on context data; adjusting the maximal oxygen consumption estimates using the at least one confidence weight; aggregating the adjusted maximal oxygen consumption estimates to generate a summary maximal oxygen consumption estimate and corresponding confidence interval for the user; and classifying cardiorespiratory fitness of the user based on at least one of the summary maximum consumption estimate, the corresponding confidence interval, a population error model or a low cardiorespiratory fitness threshold.

Abstract: One or more electronic device may use motion and/or activity sensors to estimate a user's maximum volumetric flow of oxygen, or VO2 max. In particular, although a correlation between heart rate and VO2 max may be linear at high heart rate levels, there is not a linear correlation at lower heart rate levels. Therefore, for users without extensive workout data, the motion sensors and activity sensors may be used to determine maximum calories burned by the user, workout data, including heart rate data, and body metric data. Based on these parameters, a personalized relationship between the user's heart rate and oxygen pulse (which is a function of VO2) may be determined, even with a lack of high intensity workout data. In this way, a maximum heart rate and therefore a VO2 max value may be approximated for the user.

Abstract: Systems and methods that analyze blood-based cancer diagnostic tests using multiple classes of molecules are described. The system uses machine learning (ML) to analyze multiple analytes, for example cell-free DNA, cell-free microRNA, and circulating proteins, from a biological sample. The system can use multiple assays, e.g., whole-genome sequencing, whole-genome bisulfite sequencing or EM-seq, small-RNA sequencing, and quantitative immunoassay. This can increase the sensitivity and specificity of diagnostics by exploiting independent information between signals. During operation, the system receives a biological sample, and separates a plurality of molecule classes from the sample. For a plurality of assays, the system identifies feature sets to input to a machine learning model. The system performs an assay on each molecule class and forms a feature vector from the measured values.

Abstract: Systems and methods that analyze blood-based cancer diagnostic tests using multiple classes of molecules are described. The system uses machine learning (ML) to analyze multiple analytes, for example cell-free DNA, cell-free microRNA, and circulating proteins, from a biological sample. The system can use multiple assays, e.g., whole-genome sequencing, whole-genome bisulfite sequencing or EM-seq, small-RNA sequencing, and quantitative immunoassay. This can increase the sensitivity and specificity of diagnostics by exploiting independent information between signals. During operation, the system receives a biological sample, and separates a plurality of molecule classes from the sample. For a plurality of assays, the system identifies feature sets to input to a machine learning model. The system performs an assay on each molecule class and forms a feature vector from the measured values.