Method of and system for displaying changes in a medical state of a patient with machine learning

Abstract

A method, system, and program applies machine learning in displaying a patient's medical state. A first set of data can be received corresponding to a medical state of the patient and defined by a first plurality of positives corresponding to deviations from a healthy state. The first positives are grouped to a second plurality of positives defining by a medical condition. Individual first positives can correlate to a different medical conditions, thus defining a plurality a problem bundles associated with the patient. A mismatch occurs when one of the first positives is not groupable to the second positives of a particular condition. Via machine learning, mismatches are reduced by repeatedly identifying the mismatch and revising a second data set of the relevant medical condition to include the relevant first positive, eliminating the mismatch from a subsequent groupings and displays.

1 . A method for applying machine learning in displaying a patient's medical state comprising:
 receiving, at a computing device having one or more processors, a first input of a first set of data corresponding to a first medical state of the patient and including a first plurality of positives, wherein each of the first plurality of positives corresponds to a deviation from a healthy state for a particular anatomical portion of the patient's body and is a numerical value;   grouping, at the computing device, at least some of the first plurality of positives to a second plurality of positives wherein a plurality of different medical conditions are each defined by a respective second data set of one of more the second plurality of positives, wherein the at least some of the first plurality of positives collectively correlate to at least one of the plurality of different medical conditions and the respective second plurality of positives, thus defining a first plurality a problem bundles associated with the patient wherein each problem bundle is at least one of the first plurality of positives of the first set of data grouped to at least one of the second plurality of positives of the second set of data of a particular medical condition;   identifying, at the computing device, at least one mismatch with the first plurality of problem bundles associated with the patient during said grouping, wherein said at least one mismatch is at least one of the first plurality of positives of the patient that is not groupable to one of the second plurality of positives of at least one particular medical condition, said at least one of the first plurality of positives thus in the first set of data and not in the second set of data of the at least one particular medical condition;   displaying, on a display controlled by the computing device, a first statgraph that includes a plurality of objects including a first plurality of nodes each corresponding to one of the first plurality of positives;   reducing mismatches, at the computing device, via a machine learning algorithm, including:
 repeatedly identifying, at the computing device, a first mismatch between a first positive and a first problem bundle involving a first medical condition, and 
 revising, at the computing device, in response to said repeatedly identifying, a second data set of the first medical condition to include the first positive and thereby eliminate the first mismatch from a subsequent grouping involving the first medical condition. 
   
     
     
         2 . The method of  claim 1  wherein said displaying step is further defined as:
 displaying the first statgraph to include the plurality of objects, wherein the plurality of objects also include:
 a homunculus plane in perspective view, wherein each of the first plurality of nodes is displayed in perspective view and each projects away from the homunculus plane at a location that corresponds to a particular anatomical portion of the patient's body, wherein a height of each of the first plurality of nodes from the homunculus plane corresponds to the respective numerical value, and 
 a first plurality of axes each interconnecting at least two of the first plurality of nodes. 
 
 
     
     
         3 . The method of  claim 2  further comprising:
 receiving, at the computing device, a second input corresponding to a second medical state of the patient and including a third plurality of positives, wherein each of the third plurality of positives corresponds to a deviation from a healthy state for a particular anatomical portion of the patient's body and is a numerical value. 
 
     
     
         4 . The method of  claim 3  further comprising:
 grouping, at the computing device, at least some of the third plurality of positives to a fourth plurality of positives wherein a plurality of different medical conditions are defined by one of more the fourth plurality of positives, wherein the at least some of the third plurality of positives collectively correlate to at least one of the plurality of different medical conditions and the respective fourth plurality of positives. 
 
     
     
         5 . The method of  claim 4  further comprising:
 morphing, on the display controlled by the computing device, the displayed first statgraph into a second statgraph that includes: 
 the homunculus plane in perspective view, 
 a second plurality of nodes each in perspective view, each corresponding to one of the third plurality of positives, and each projecting away from the homunculus plane at the cell that corresponds to the respective particular anatomical portion of the patient's body, wherein a height of each of the second plurality of nodes from the homunculus plane corresponds to the respective numerical value, the first plurality of nodes morphed into the second plurality of nodes, and 
 a second plurality of axes each interconnecting at least two of the second plurality of nodes, the first plurality of axes morphed into the second plurality of axes. 
 
     
     
         6 . The method of  claim 2  wherein each of the first plurality of nodes is displayed in a respective color and wherein at least two of the first plurality of nodes are displayed with different colors. 
     
     
         7 . The method of  claim 2  wherein each of the nodes is displayed with a respective opacity and wherein at least two of the first plurality of nodes are displayed with different opacities. 
     
     
         8 . The method of  claim 2  further comprising:
 controlling the display, with the computing device, to change the viewing perspective to a side view such that the homunculus plane is displayed as a line. 
 
     
     
         9 . A system for generating a display of a patient's medical state based on machine learning and comprising:
 a display; and   a computing device, comprising one or more processors and a non-transitory, computer readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform operations comprising:
 receiving, at a computing device having one or more processors, a first input of a first set of data corresponding to a first medical state of the patient and including a first plurality of positives, wherein each of the first plurality of positives corresponds to a deviation from a healthy state for a particular anatomical portion of the patient's body and is a numerical value; 
 grouping, at the computing device, at least some of the first plurality of positives to a second plurality of positives wherein a plurality of different medical conditions are each defined by a respective second data set of one of more the second plurality of positives, wherein the at least some of the first plurality of positives collectively correlate to at least one of the plurality of different medical conditions and the respective second plurality of positives, thus defining a first plurality a problem bundles associated with the patient wherein each problem bundle is at least one of the first plurality of positives of the first set of data grouped to at least one of the second plurality of positives of the second set of data of a particular medical condition; 
 identifying, at the computing device, at least one mismatch with the first plurality of problem bundles associated with the patient during said grouping, wherein said at least one mismatch is at least one of the first plurality of positives of the patient that is not groupable to one of the second plurality of positives of at least one particular medical condition, said at least one of the first plurality of positives thus in the first set of data and not in the second set of data of the at least one particular medical condition; 
 displaying, on the display, a plurality of objects including a first plurality of nodes each corresponding to one of the first plurality of positives; and 
 reducing mismatches, at the computing device, via a machine learning algorithm, including:
 repeatedly identifying, at the computing device, a first mismatch between a first positive and a first problem bundle involving a first medical condition, and
 revising, at the computing device, in response to said repeatedly identifying, a second data set of the first medical condition to include the first positive and thereby eliminate the first mismatch from a subsequent grouping involving the first medical condition. 
 
 
   
     
     
         10 . The system of  claim 9  wherein said computing device is further defined as comprising:
 a first component being a device wearable by the patient, said first component including transceiver, a speaker, and a microphone. 
 
     
     
         11 . The system of  claim 9  wherein said computing device is further defined as comprising:
 a second component being a stylus including one or more transducers and a transceiver. 
 
     
     
         12 . The system of  claim 9  wherein said computing device is further defined as comprising:
 a third component being a tablet including a transceiver and wherein said display is mounted in said tablet. 
 
     
     
         13 . A computer program product comprising program code stored on a non-transitory computer-readable medium, which when executed by a computing device having one or more processors, enables the computing device to apply machine learning in displaying a patient's medical state by performing actions including:
 receiving, at a computing device having one or more processors, a first input of a first set of data corresponding to a first medical state of the patient and including a first plurality of positives, wherein each of the first plurality of positives corresponds to a deviation from a healthy state for a particular anatomical portion of the patient's body and is a numerical value;   grouping, at the computing device, at least some of the first plurality of positives to a second plurality of positives wherein a plurality of different medical conditions are each defined by a respective second data set of one of more the second plurality of positives, wherein the at least some of the first plurality of positives collectively correlate to at least one of the plurality of different medical conditions and the respective second plurality of positives, thus defining a first plurality a problem bundles associated with the patient wherein each problem bundle is at least one of the first plurality of positives of the first set of data grouped to at least one of the second plurality of positives of the second set of data of a particular medical condition;   identifying, at the computing device, at least one mismatch with the first plurality of problem bundles associated with the patient during said grouping, wherein said at least one mismatch is at least one of the first plurality of positives of the patient that is not groupable to one of the second plurality of positives of at least one particular medical condition, said at least one of the first plurality of positives thus in the first set of data and not in the second set of data of the at least one particular medical condition;   displaying, on a display controlled by the computing device, a plurality of objects including a first plurality of nodes each corresponding to one of the first plurality of positives; and   reducing mismatches, at the computing device, via a machine learning algorithm, including:
 repeatedly identifying, at the computing device, a first mismatch between a first positive and a first problem bundle involving a first medical condition, and 
 revising, at the computing device, in response to said repeatedly identifying, a second data set of the first medical condition to include the first positive and thereby eliminate the first mismatch from a subsequent grouping involving the first medical condition.