US 12,440,134 B2
Apparatuses and methods for individualized polygraph testing
Bradford R. Everman, Haddonfield, NJ (US); and Brian Bradke, Brookfield, VT (US)
Filed by GMECI, LLC, Beavercreek, OH (US)
Filed on Mar. 5, 2024, as Appl. No. 18/595,836.
Application 18/595,836 is a continuation of application No. 18/072,397, filed on Nov. 30, 2022, granted, now 11,950,909.
Application 18/072,397 is a continuation in part of application No. 17/387,491, filed on Jul. 28, 2021, granted, now 11,604,513, issued on Mar. 14, 2023.
Prior Publication US 2025/0000410 A1, Jan. 2, 2025
Int. Cl. A61B 5/16 (2006.01); A61B 5/00 (2006.01); A61B 5/024 (2006.01)
CPC A61B 5/164 (2013.01) [A61B 5/7221 (2013.01); A61B 5/7246 (2013.01); A61B 5/7267 (2013.01); A61B 5/7405 (2013.01); A61B 5/02416 (2013.01); A61B 5/742 (2013.01)] 16 Claims
OG exemplary drawing
 
1. An apparatus for individualized polygraph testing, the apparatus comprising:
at least an interface configured to communicate questions to a user;
at least a sensor configured to detect a parameter, wherein the parameter comprises a biofeedback of the user, wherein the at least a sensor comprises at least a ground electrode and at least an electromyography (EMG) electrode; and
at least a computing device comprising:
at least a processor; and
a memory communicatively connected to the at least a processor, the memory containing instructions configuring the at least a processor to:
detect, using the at least a sensor, at least a biofeedback signal comprising at least a speech parameter as a function of the biofeedback of the user associated with the communicated question, wherein the at least a biofeedback signal comprises at least an EMG signal detected using the at least an EMG electrode;
classify the at least a biofeedback signal to a user state as a function of a user state classifier, wherein the user state classifier is generated using a user state machine learning algorithm, wherein training the user state machine learning algorithm comprises:
training the user state machine learning algorithm using a first training set comprising cohort data associated with a cohort to which the user belongs; and
retraining the user state machine learning algorithm using a second training set comprising user data;
determine at least a speech pattern as a function of the at least a speech parameter by implementing a machine learning model trained on historic information comprising at least speech-related study data; and
generate a confidence metric as a function of the classified user state.