Computer-based system for applying machine learning models to healthcare related data

Abstract

Techniques for responding to a healthcare inquiry from a user are disclosed. In one particular embodiment, the techniques may be realized as a method for responding to a healthcare inquiry from a user, according to a set of instructions stored on a memory of a computing device and executed by a processor of the computing device, the method comprising the steps of: classifying an intent of the user based on the healthcare inquiry; instantiating a conversational engine based on the intent; eliciting, by the conversational engine, information from the user; and presenting one or more medical recommendations to the user based at least in part on the information.

What is claimed is: 
     
         1 . A computer system that applies machine learning modeling to a healthcare-related input of a patient, comprising:
 (a) a conversational module configured to carry out an automated conversation with a patient that elicits the healthcare-related input from the patient; and   (b) a diagnostic reasoning module that applies a machine learning model to at least a portion of the healthcare-related input to determine at least one diagnosis or differential diagnosis of the patient or course of action for the patient.   
     
     
         2 . The computer system of  claim 1 , further comprising a knowledge base that is arranged based on one or more ontological relationships of medical terms and wherein the machine learning model receives an output from the knowledge base that comprises a relationship between at least a portion of the healthcare-related input to a potential differential diagnosis or a potential course of action. 
     
     
         3 . The computer system of  claim 2 , wherein the knowledge base is dynamically updated when no relationship exists in the knowledge base between the at least the portion of the healthcare-related input to the potential differential diagnosis or the potential course of action. 
     
     
         4 . The computer system of  claim 1 , wherein the computer system is configured to receive multimodal input as part of the conversation. 
     
     
         5 . The computer system of  claim 4 , wherein the multimodal input comprises one or more of written input, spoken input, non-speech audio input, static image input, and video input. 
     
     
         6 . The computer system of  claim 1 , wherein natural language processing is applied to at least a portion of the healthcare-related input that is unstructured in order to structure the at least the portion of the healthcare-related input. 
     
     
         7 . A computer system that applies machine learning modeling to a healthcare-related input of a patient, the computer system comprising:
 (a) a patient-interface module configured to receive the healthcare-related input from the patient via a user-interface; and   (b) a conversational module configured to:
 (i) apply a machine learning model, that models text, to the healthcare-related input; and 
 (ii) determine, using the machine learning model, one or more words or sentences to provide to the patient to create at least a portion of a computer-generated conversation with the patient that relates to or includes the healthcare-related input or a portion thereof. 
   
     
     
         8 . The computer system of  claim 7 , further comprising (c) a diagnostic module configured to determine a diagnosis or a differential diagnosis of the patient based on the at least the portion of the computer-generated conversation with the patient. 
     
     
         9 . The computer system of  claim 8 , wherein the diagnostic module determines the diagnosis by applying a machine learning model to the healthcare-related input. 
     
     
         10 . The computer system of  claim 8 , comprising a knowledge base that is arranged based on one or more ontological relationships of medical terms, and wherein the healthcare-related input is applied by the computer system to the knowledge base to determine at least one relationship between data in the knowledge base and at least a portion of the healthcare-related input. 
     
     
         11 . The computer system of  claim 10 , wherein the diagnostic module receives an output from the knowledge base, based on the healthcare-related input, that comprises the at least one relationship between the data in the knowledge base and the at least the portion of the healthcare-related input. 
     
     
         12 . The computer system of  claim 10 , wherein the knowledge base is dynamically updated when no relationship exists between the data in the knowledge base and the at least the portion of the healthcare-related input. 
     
     
         13 . The computer system of  claim 7 , wherein the patient interface module is configured to receive multimodal input as the healthcare-related input. 
     
     
         14 . The computer system of  claim 13 , wherein the multimodal input comprises one or more of written input, spoken input, non-spoken audio input, static image input, and video input. 
     
     
         15 . The computer system of  claim 7 , wherein natural language processing is applied to at least a portion of the healthcare-related input that is unstructured in order to structure the at least the portion of the healthcare-related input. 
     
     
         16 . The computer system of  claim 7 , wherein the one or more words or sentences is determined by the conversational module to elicit input types that will lead to a final disposition with respect to the patient using a fewest amount words or sentences. 
     
     
         17 . The computer system of  claim 16 , wherein the final disposition with respect to the patient comprises a diagnosis of a medical condition of the patient or a course of action. 
     
     
         18 . The computer system of  claim 16 , wherein the final disposition comprises connecting the patient with a healthcare provider. 
     
     
         19 . The computer system of  claim 16 , wherein the final disposition comprises an automated task carried out by the computer system comprising issuing or renewing a prescription or providing a referral. 
     
     
         20 . The computer system of  claim 7 , wherein the computer system creates a profile for the patient based on one or more of an electronic medical record and a previous use of the computer system by the patient. 
     
     
         21 . A non-transitory computer readable medium including software configured to cause at least one processor to:
 (a) receive an initial input from a patient comprising medical information of the patient in a natural language format; and   (b) determine, using a conversational module of the software, one or more responses to the initial input to provide to the patient that will elicit subsequent input from the patient, related to the medical information, that is in a structured format.   
     
     
         22 . The non-transitory computer readable medium of  claim 21 , wherein the software is further configured to cause the at least one processor to utilize a knowledge base comprising relationships between medical terms to identify one or more relationships between the initial input and one or more medical terms. 
     
     
         23 . The non-transitory computer readable medium of  claim 22 , wherein the one or more responses to the initial input are determined based on the one or more relationships that are identified. 
     
     
         24 . The non-transitory computer readable medium of  claim 22 , wherein the natural language format represents an absence of the one or more relationships between the initial input and the one or more medical terms. 
     
     
         25 . The non-transitory computer readable medium of  claim 22 , wherein the software is further configured to cause the at least one processor to apply a machine learning model to an output of the knowledge base to determine the one or more responses. 
     
     
         26 . The non-transitory computer readable medium of  claim 22 , wherein the conversational module applies natural language processing to the input. 
     
     
         27 . The non-transitory computer readable medium of  claim 22 , wherein the non-transitory computer readable medium is configured to receive multimodal input as the initial input. 
     
     
         28 . The non-transitory computer readable medium of  claim 27 , wherein the multimodal input comprises one or more of written input, spoken input, non-spoken audio input, static image input, and video input.