Dynamic recommendations of sensors for patient monitoring

A computer program product for a recommendation, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform:

• monitoring a patient using a plurality of sensors including missing inputs as sensor failures or a sensor disconnection;
• receiving patient information including a comfort level corresponding to a sensor of the plurality of sensors;
• determining a relevance of each sensor of the plurality of sensors to at least one health condition of the patient;
• and
• determining at least one sensor of the plurality of sensors to disconnect based on the comfort level and the relevance of each sensor, wherein a machine learning system determines the at least one sensor being an optimal sensor,
• wherein a discriminative classifier and a generative classifier are used in combination to operate concurrently with the machine learning system to consider the missing inputs to make a prediction of the relevance of each sensor regardless of the missing inputs,
• further comprising disconnecting the at least one sensor of the plurality of the sensors based on a result of the determining,
• wherein an output of the generative classifier is analyzed to turn off a sensor, and
• wherein the machine learning system is trained by deliberately turning off inputs in a systematic way to train the generative classifier and then a performance after turning off inputs is compared to the discriminative classifier.

A recommendation system, the system comprising:

• a processor;
• and
• a memory, the memory storing instructions to cause the processor to perform:
• monitoring a patient using a plurality of sensors including missing inputs as sensor failures or a sensor disconnection;
• receiving patient information including a comfort level corresponding to a sensor of the plurality of sensors;
• determining a relevance of each sensor of the plurality of sensors to at least one health condition of the patient;
• and
• determining at least one sensor of the plurality of sensors to disconnect based on the comfort level and the relevance of each sensor, wherein a machine learning system determines the at least one sensor being an optimal sensor,
• wherein a discriminative classifier and a generative classifier are used in combination to operate concurrently with the machine learning system to consider the missing inputs to make a prediction of the relevance of each sensor regardless of the missing inputs,
• further comprising disconnecting the at least one sensor of the plurality of the sensors based on a result of the determining,
• wherein an output of the generative classifier is analyzed to turn off a sensor, and
• wherein the machine learning system is trained by deliberately turning off inputs in a systematic way to train the generative classifier and then a performance after turning off inputs is compared to the discriminative classifier.