Systems and methods for semi-automated medical processes

Abstract

Presented are systems and methods for the accurate acquisition of medical measurement data of a body part of patient. To assist in acquiring accurate medical measurement data, an automated diagnostic and treatment system provides instructions to the patient to allow the patient to precisely position a medical instrument in proximity to a target spot of a body part of patient. Based on a series of images acquired from a kiosk camera and an instrument camera, these instructions are generated. Subsequent images from instrument camera are analyzed by automated diagnostic and treatment system utilizing a database and deep convolution neural network (DCNN) to obtain measured medical data. An error threshold measurement by the automated diagnostic and treatment system may determine the accuracy of the instrument positioning and the medical measured data. The measured medical data may be communicated to a physician.

What is claimed is: 
     
         1 . A medical data system for generating patient treatment instructions, the system comprising:
 a diagnostic engine utilizing machine learning that processes patient information and calculates probabilities associated with illnesses that have been identified from a set of potential illnesses and generates diagnostic data, the diagnostic engine having an input that receives patient information from at least one of a patient, medical staff, doctor, a medical history database and automated system;   a treatment engine coupled to the diagnostic engine, the treatment engine receives diagnostic data from the diagnostic engine and performs the steps of:
 determining a treatment option for one or more of the illnesses; 
 analyzing a likelihood of success for the treatment option;
 if the likelihood of success is below a threshold, the treatment requests additional diagnostic data from the diagnostic engine; 
 if the likelihood of success is above the threshold, treatment instructions for one or more treatments options are generated based on the analysis; 
 
 receiving feedback regarding the treatment instructions; and 
 providing the feedback to the diagnostic engine to improve the machine learning processes. 
   
     
     
         2 . The medical data system according to  claim 1 , wherein the treatment engine assigns criticality scores to one or more of the illnesses. 
     
     
         3 . The medical data system according to  claim 1 , wherein the diagnostic engine populates an electronic health care record (EHR). 
     
     
         4 . The medical data system according to  claim 3 , further comprising a notes generation engine that generates exam notes to populate the EHR. 
     
     
         5 . The medical data system according to  claim 1 , wherein the medical data system comprises a coding engine that, based on at least one of drug information, a patient preference, and a patient medical history, generates an output that comprises a treatment code. 
     
     
         6 . The medical data system according to  claim 5 , wherein the treatment engine uses at least one of the treatment plan and the treatment code to update an EHR. 
     
     
         7 . A method for generating patient treatment instructions, the method comprising:
 receiving patient information at a diagnostic engine, the patient information being received from at least one of a patient, medical staff, doctor, a medical history database and automated system;   generating a set of potential illnesses from an analysis of the patient information;   assigning probabilities to the set of potential illnesses using machine learning processes;   producing diagnostic data from the set of potential illnesses and the assigned probabilities;   based on the diagnostic data, determining a treatment option for one or more of the illnesses;   analyzing a likelihood of success for the treatment option;
 if the likelihood of success is below a threshold, requesting additional diagnostic data from the diagnostic engine; 
 if the likelihood of success is above the threshold, generating treatment instructions for one or more treatments options based on the analysis; 
   receiving feedback regarding the treatment instructions; and   providing the feedback to the diagnostic engine to improve the machine learning processes.   
     
     
         8 . The method according to  claim 7 , wherein the one or more of illnesses may be assigned criticality scores. 
     
     
         9 . The method according to  claim 7 , further comprising receiving patient-related data from an electronic health care record (EHR). 
     
     
         10 . The method according to  claim 7 , further comprising, based on the diagnostic data, generating an output that comprises a treatment code that is based on at least one of drug information, a patient preference, and a patient medical history. 
     
     
         11 . The method according to  claim 10 , further comprising using at least one of the treatment instructions and the treatment code to update an EHR. 
     
     
         12 . The method according to  claim 10 , further comprising generating exam notes to populate the EHR. 
     
     
         13 . The method according to  claim 7 , further comprising recalculating the likelihood of success to adjust the treatment plan. 
     
     
         14 . The method according to  claim 7 , wherein the treatment instructions comprise one of patient treatment instructions and requests for additional examination. 
     
     
         15 . The method according to  claim 14 , wherein further comprising calculating a risk of a false negative for the one or more illnesses. 
     
     
         16 . A method for partial automated medical treatment, the method comprising:
 receiving patient information from a patient or medical staff;   receiving patient medical history from a medical history database;   generating a set of potential illnesses from an analysis of the patient information and the patient medical history;   assigning a first set of probabilities to the set of potential illnesses using machine learning processes, the first set of probabilities relating to estimated illness accuracies associated with first set of potential illnesses based on the machine learning process;   generating a set of diagnostic data for at least two of the potential illnesses within the plurality of potential illnesses;   based on the at least one potential illness within the set of potential illnesses and at least some of the set of diagnostic data, determining a treatment option for one or more of the illnesses within the set of potential illnesses;   analyzing a likelihood of success for the treatment option;
 if the likelihood of success is below a threshold, requesting additional diagnostic data from the diagnostic engine; 
 if the likelihood of success is above the threshold, generating treatment instructions for one or more treatments options based on the analysis; 
   receiving feedback regarding the treatment instructions; and   providing the feedback to the diagnostic engine to improve the machine learning processes.   
     
     
         17 . The method according to  claim 16 , further comprising, based on the diagnostic data, generating an output that comprises a treatment code that is based on at least one of drug information, a patient preference, and a patient medical history. 
     
     
         18 . The method according to  claim 17 , further comprising using at least one of the treatment instructions and the treatment code to update an EHR. 
     
     
         19 . The method according to  claim 17 , further comprising generating exam notes to populate the EHR. 
     
     
         20 . The method according to  claim 16 , wherein the treatment instructions comprise one of patient treatment instructions and requests for additional examination.