| CPC G16H 20/13 (2018.01) [G06N 3/08 (2013.01); G16H 10/60 (2018.01); G16H 80/00 (2018.01)] | 18 Claims |
|
1. A method comprising:
receiving, by one or more processors, data associated with a patient, wherein the received data includes medication regiment information for a plurality of medications and patient specific information;
applying a model to the received data to generate a score for at least one medication of the plurality of medications, wherein:
the score is indicative of a likelihood that the patient will successfully convert from a current dispensing process to an alternate dispensing process,
the model includes a machine learning technique having a neural network, and
the machine learning technique is trained to establish a relationship between a plurality of training patient-drug combination features and medication dispensing process conversion likelihoods;
training the machine learning technique by:
obtaining a batch of training data including a first set of the plurality of training patient-drug combination features and medication dispensing process conversion likelihoods;
processing the first set of the plurality of training patient-drug combination features with the machine learning technique to generate estimated medication dispensing process conversion likelihoods;
computing a loss function based on a deviation between the estimated medication dispensing process conversion likelihoods and corresponding medication dispensing process conversion likelihoods of the first set of the plurality of training patient-drug combination features; and
updating parameters of the machine learning technique based on the computed loss function; and
in response to the generated score exceeding a threshold value, triggering a notification associated with converting the patient from the current dispensing process to the alternate dispensing process.
|