Detecting material type using low-energy sensing

A non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to execute a process for detecting material type using low-energy sensing, the process comprising:

• receiving frequency words from a tag attached to a container containing material, wherein the tag provides a low-energy sensing configured to transmit the frequency words, wherein each frequency word is based on a frequency calibration of the tag with respect to a currently specified transmission frequency for transmission of data by the tag, wherein a frequency word changes when the tag is out of a calibration with respect to the currently specified transmission frequency for transmission of data by the tag;
• extracting at least a first data feature from the received frequency words, wherein the extracted at least a first data feature changes in response to a type of material in the container;
• classifying the extracted at least a first data feature to label a type of the material in the container, wherein the classifying is performed by a trained machine classifier that employs a trained machine learning model;
• and
• sending a notification indicating the type of material in the container as classified by the trained machine classifier.

A system for developing a treatment plan using multi-stage machine learning, comprising:

• a processing circuitry;
• and
• a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:
• receive frequency words from a tag attached to a container containing material, wherein the tag provides a low-energy sensing configured to transmit the frequency words, wherein each frequency word is based on a frequency calibration of the tag with respect to a currently specified transmission frequency for transmission of data by the tag;
• extract at least a first data feature from the received frequency words, wherein the extracted at least a first data feature changes in response to a type of material in the container;
• classify the extracted at least a first data feature to label a type of the material in the container, wherein the classifying is performed by a trained machine classifier that employs a trained machine learning model;
• and
• send a notification indicating the type of material in the container as classified by the trained machine classifier.