Automated program optimization

Abstract

An example of a system may include a processor; and a memory device comprising instructions, which when executed by the processor, cause the processor to access at least one of: patient input, clinician input, or automatic input; use the patient input, clinician input, or automatic input in a search method, the search method designed to evaluate a plurality of candidate neuromodulation parameter sets to identify an optimal neuromodulation parameter set; and program a neuromodulator using the optimal neuromodulation parameter set to stimulate a patient.

What is claimed is: 
     
         1 . A system comprising:
 a processor; and   a memory device comprising instructions, which then executed by the processor, cause the processor to:
 access at least one of: patient input, clinician input, or automatic input; 
 use the patient input, clinician input, or automatic input in a search method, the search method designed to evaluate a plurality of candidate neuromodulation parameter sets to identify an optimal neuromodulation parameter set of the plurality of candidate neuromodulation parameter sets; and 
 program a neuromodulator using the optimal neuromodulation parameter set to stimulate a patient. 
   
     
     
         2 . The system of  claim 1 , wherein the patient input comprises subjective data. 
     
     
         3 . The system of  claim 2 , wherein the subjective data comprises a visual analog scale (VAS), numerical rating scale (NRS), a satisfaction score, a global impression of change, or an activity level. 
     
     
         4 . The system of  claim 1 , wherein the clinician input comprises a selected neuromodulation parameter set, a selected neuromodulation parameter set dimension, or a search method configuration option. 
     
     
         5 . The system of  claim 4 , wherein the selected neuromodulation parameter set dimension comprises a spatial location, a frequency, a pulse width, a number of pulses within a burst or train of pulses, the train-to-train interval, the burst frequency of these trains, a pulse duty cycle, or a burst duty cycle. 
     
     
         6 . The system of  claim 4 , wherein the search method configuration option comprises a test range for a neuromodulation parameter set dimension, a termination criteria for a neuromodulation parameter set test, an amount of time to test a neuromodulation parameter set, a minimum evaluation time for a candidate neuromodulation parameter set, or a survival threshold for a neuromodulation parameter set under test. 
     
     
         7 . The system of  claim 1 , wherein the automatic input comprises data received from a patient device. 
     
     
         8 . The system of  claim 7 , wherein the patient device comprises an accelerometer and the automatic input comprises activity data. 
     
     
         9 . The system of  claim 7 , wherein the patient device comprises a heart rate monitor and the automatic input comprises heart rate or heart rate variability. 
     
     
         10 . The system of  claim 7 , wherein the patient device comprises an implantable pulse generator and the automatic input comprises field potentials. 
     
     
         11 . The system of  claim 1 , wherein the search method comprises a sorting algorithm that uses scoring from the patient to sort the plurality of candidate parameter sets and remove parameter sets from the plurality of candidate parameter sets that fail to meet a threshold score. 
     
     
         12 . The system of  claim 1 , wherein the search method comprises a gradient descent system that progresses through the plurality of candidate parameter sets to optimize a dimension of the candidate parameter sets. 
     
     
         13 . The system of  claim 1 , wherein the search method comprises a sensitivity analysis that builds a model from stimulation variables and outcome variables, and uses a regression model to identify a vector of coefficients. 
     
     
         14 . A method comprising:
 accessing, at a computerized system, at least one of: patient input, clinician input, or automatic input;   using the patient input, clinician input, or automatic input in a search method, the search method designed to evaluate a plurality of candidate neuromodulation parameter sets to identify an optimal neuromodulation parameter set of the plurality of candidate neuromodulation parameter sets; and   programming a neuromodulator using the optimal neuromodulation parameter set to stimulate a patient.   
     
     
         15 . The method of  claim 14 , wherein the patient input comprises subjective data, the subjective data comprising a visual analog scale (VAS), numerical rating scale (NRS), a satisfaction score, a global impression of change, or an activity level. 
     
     
         16 . The method of  claim 14 , wherein the clinician input comprises a selected neuromodulation parameter set, a selected neuromodulation parameter set dimension, or a search method configuration option. 
     
     
         17 . The method of  claim 14 , wherein the search method comprises a sorting algorithm that uses scoring from the patient to sort the plurality of candidate parameter sets and remove parameter sets from the plurality of candidate parameter sets that fail to meet a threshold score. 
     
     
         18 . The method of  claim 14 , wherein the search method comprises a gradient descent method that progresses through the plurality of candidate parameter sets to optimize a dimension of the candidate parameter sets. 
     
     
         19 . The method of  claim 14 , wherein the search method comprises a sensitivity analysis that builds a model from stimulation variables and outcome variables, and uses a regression model to identify a vector of coefficients. 
     
     
         20 . A non-transitory machine-readable medium including instructions, which when executed by a machine, cause the machine to:
 access at least one of: patient input, clinician input, or automatic input;   use the patient input, clinician input, or automatic input in a search method, the search method designed to evaluate a plurality of candidate neuromodulation parameter sets to identify an optimal neuromodulation parameter set of the plurality of candidate neuromodulation parameter sets; and   program a neuromodulator using the optimal neuromodulation parameter set to stimulate a patient.