System and method for refining nerve health assessment models
Inventors
Nay, David S. • Wybo, Christopher
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12390147-B1
Publication Date
2025-08-19
Expiration Date
Abstract
A method for refining a nerve health assessment model used during surgical procedures, includes accessing a database storing historical patient data; analyzing the historical patient data to generate or update a predictive model configured to estimate at least one of: a predicted post-operative outcome, an individualized target range for an intraoperative nerve function parameter, or a statistical risk profile for surgical complications; obtaining current intraoperative nerve function measurements; receiving actual post-operative outcome data for the current patient; updating the database to include new patient data associated with the current patient; and refining the predictive model based at least in part on the new patient data added to the database.
Core Innovation
The invention refines a nerve health assessment model used during surgical procedures by using historical patient data and subsequent patient-specific outcome data. The approach accesses a database storing historical patient data comprising pre-operative patient characteristics, previously obtained intraoperative nerve function measurements, and actual post-operative outcomes, and analyzes the historical patient data to generate or update a predictive model.
Current intraoperative nerve function measurements are obtained during a current surgical procedure by generating one or more electrical stimuli and transmitting the electrical stimuli to a nerve of a current patient via an electrode. Muscle responses evoked by the electrical stimuli are detected by a sensor in communication with the processor, and detected muscle responses are analyzed to determine at least one current intraoperative nerve function measurement.
After completion of the current surgical procedure, actual post-operative outcome data are received, and new patient data including pre-operative characteristics, the determined intraoperative nerve function measurement(s), and the actual post-operative outcome data are added to the database. The predictive model is then refined based at least in part on the new patient data added to the database.
In nerve function monitoring, the nerve function measurements include mechanical responses detected by a mechanomyography (MMG) sensor, and the determined intraoperative nerve function measurement includes at least one of a minimum stimulation threshold, a supramaximal stimulation threshold, or the magnitude of a supramaximal muscle response. The model refinement can generate confidence intervals for estimated outputs and can include comparing estimated output for a current patient with actual post-operative outcome data to adjust or retrain machine learning algorithm parameters.
Claims Coverage
Independent claim 1 and independent claim 11 define a method and a system for refining a nerve health assessment model used during surgical procedures. The independent claims include three inventive features: database-driven predictive model generation or update, acquisition of current intraoperative nerve function measurements via electrical stimulation and sensor-detected muscle responses, and outcome-based database update with predictive model refinement using new patient data.
Database-driven predictive model generation or update with historical patient data
Accessing a database storing historical patient data comprising pre-operative patient characteristics, previously obtained intraoperative nerve function measurements, and actual post-operative outcomes; analyzing the historical patient data to generate or update a predictive model configured to estimate at least one of a predicted post-operative outcome, an individualized target range for an intraoperative nerve function parameter, or a statistical risk profile for surgical complications, based at least in part on pre-operative patient characteristics and intraoperative nerve function measurements.
Electrical stimulation and sensor-based determination of current intraoperative nerve function measurements
Obtaining current intraoperative nerve function measurements during a current surgical procedure by generating one or more electrical stimuli; transmitting the stimuli to an electrode in contact with a nerve; detecting muscle responses evoked by the stimuli with a sensor in communication with the processor; and analyzing the detected muscle responses to determine at least one current intraoperative nerve function measurement.
Outcome-based database update and predictive model refinement using new patient data
Receiving actual post-operative outcome data after completion of the current surgical procedure; updating the database with new patient data including pre-operative characteristics, the determined intraoperative nerve function measurement(s), and the actual post-operative outcome data; and refining the predictive model based at least in part on the new patient data added to the database.
The independent claims cover a closed-loop framework in which historical patient data are used to generate or update a predictive model for nerve health assessment, current intraoperative nerve function measurements are obtained via electrical stimulation and sensor-detected muscle responses, actual post-operative outcome data are received and added to the database, and the predictive model is refined based on the newly added patient data.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?