Sensing system and method for monitoring time-dependent processes

7:

• The method of claim 1, wherein training the machine learning algorithm includes:
• receiving labeled sensor data, the labeled sensor data including first labels identifying usable and unusable samples in the labeled sensor data;
• assigning second labels to samples in the labeled sensor data, wherein assigning second labels includes applying the machine learning algorithm to the labeled sensor data to detect usable and unusable samples in the labeled sensor data;
• and
• modifying the expected outlier ratio as a function of discrepancies between the first and second labels.

11:

• A system comprising:
• a sensor device configured to monitor at least one biological function;
• one or more processors connected to the sensor device;
• and
• memory connected to the one or more processors, wherein the memory includes instructions that, when executed by the one or more processors, cause the one or more processors to:
• receive sensor data from the sensor device, the sensor data representing time-series measurement samples of one or more parameters associated with the biological functions being monitored by the sensor device, wherein the time-series measurement samples include usable and unusable samples;
• extract data representing two or more features of samples of the time-series measurement samples;
• filter the extracted data, wherein filtering includes eliminating outliers in the extracted data based on an expected outlier ratio;
• and
• train a machine learning algorithm to identify events based on the filtered extracted data.

13:

• The system of claim 11, wherein the instructions that cause the one or more processors to train the machine learning algorithm comprise instructions that, when executed, cause the one or more processors to:
• receive labeled sensor data, the labeled sensor data including first labels identifying usable and unusable samples in the labeled sensor data;
• assign second labels to samples in the labeled sensor data, wherein assigning second labels includes applying the machine learning algorithm to the labeled sensor data to detect usable and unusable samples in the labeled sensor data;
• and
• modify the expected outlier ratio as a function of discrepancies between the first and second labels.

14:

• The system of claim 13, wherein the time-series measurement samples include measurement samples associated with breathing, and
• wherein the instructions that cause the one or more processors to train the machine learning algorithm comprise instructions that, when executed, cause the one or more processors to:
• receive labeled sensor data includes concurrently capturing measurement samples associated with breathing via the sensor device and via a breath measurement apparatus;
• and
• label the breath measurements captured by the sensor device based on the measurement samples associated with breathing captured by the breath measurement apparatus.