System for Surgical Decisions Using Deep Learning

Abstract

A method of monitoring and treating a patient using physiological data, a computer learning system, and a predictive model. The method may include generating predicted physiological data for a patient that is compared to a predictive model. When the predicted physiological data is comparable to the predictive model, preemptive care is administered.

I claim: 
     
         1 . A method of monitoring and treating a patient using physiological data, a computer learning system, and a predictive model, the method comprising:
 inputting into the computer learning system reference physiological data for previous patients that experienced a surgical complication;   generating, in the computer learning system, a predictive model corresponding to the reference physiological data for patients that experienced the surgical complication;   inputting into the computer learning system a patient's current physiological data that is contemporaneously obtained from a physiological monitor during a surgery;   generating predicted future physiological data for the patient based on the patient's current physiological data and the predictive model;   comparing the patient's future predicted physiological data to the predictive model; and   if the patient's predicted future physiological data is comparable to the predictive model, then administering preemptive care for the surgical complication.   
     
     
         2 . The method of  claim 1 , wherein the computer learning system comprises a deep learning processor. 
     
     
         3 . The method of  claim 1 , wherein administering preemptive care for the surgical complication is selected from the group consisting of omitting surgical actions, administering medication, conducting diagnostic testing, and performing surgical actions. 
     
     
         4 . The method of  claim 1 , further comprising predicting a date and time when the surgical complication will occur based on the comparison of the patient's predicted physiological data with the predictive model. 
     
     
         5 . The method of  claim 4 , further comprising, if the surgical complication is predicted to occur during a surgery, then administering preemptive care during the surgery. 
     
     
         6 . The method of  claim 5 , wherein administering preemptive care for the surgical complication during surgery is selected from the group consisting of administering medication, avoiding surgical actions, conducting diagnostic testing, and performing surgical actions. 
     
     
         7 . The method of  claim 4 , further comprising providing the surgical complication and the predicted date and time on a display for viewing by a medical provider. 
     
     
         8 . The method of  claim 7 , wherein the predicted date and time are provided on the display in the form of a countdown clock. 
     
     
         9 . The method of  claim 1 , wherein the patient's physiological data is selected from the group consisting of pulse, oxygen level, brain wave patterns, electrocardiogram data, blood test results, and organ scans. 
     
     
         10 . The method of  claim 5 , further comprising providing to a medical surgical participant, on a display, identification of the preemptive care to be administered during surgery. 
     
     
         11 . The method of  claim 5 , further comprising generating and providing to a medical surgical participant, on a display and based on at least the surgical complication, identification of contraindicated actions that should not be taken during surgery. 
     
     
         12 . The method of  claim 1 , wherein administering preemptive care comprises the implementation of non-surgical actions and the avoidance of further surgical actions. 
     
     
         13 . The method of  claim 1 , wherein the surgical complication is organ failure. 
     
     
         14 . The method of  claim 1 , wherein the surgical complication is stroke.