Prediction of Tinnitus Treatment Outcomes Based on EEG Sensors and TFI Score Using Deep Learning

Abstract

Tinnitus is a hearing disorder that is characterized by the perception of sounds in the absence of an external source. Currently, there is no pharmaceutical cure for tinnitus, however, multiple therapies and interventions have been developed that improve or control associated distress and anxiety. We propose a new Artificial Intelligence (AI) algorithm as a digital prognostic health system that models electroencephalographic (EEG) data in order to predict patients’ responses to tinnitus therapies. The EEG data was collected from patients prior to treatment and 3-months following a sound-based therapy. Feature selection techniques were utilised to identify predictive EEG variables with the best accuracy. The patients’ EEG features from both the frequency and functional connectivity domains were entered as inputs that carry knowledge extracted from EEG into AI algorithms for training and predicting therapy outcomes. The AI models differentiated the patients’ outcomes into either therapy responder or non-responder, as defined by their Tinnitus Functional Index (TFI) scores, with accuracies ranging from 98%–100%. Our findings demonstrate the potential use of AI, including deep learning, for predicting therapy outcomes in tinnitus. The research suggests an optimal configuration of the EEG sensors that are involved in measuring brain functional changes in response to tinnitus treatments. It identified which EEG electrodes are the most informative sensors and how the EEG frequency and functional connectivity can better classify patients into the responder and non-responder groups. This has potential for real-time monitoring of patient therapy outcomes at home.

The potassium channels: Neurobiology and pharmacology of tinnitus

“…They can detect changes in brain activity and changes in connectivity (Adjamian, 2014). They have been used to evaluate the outcome of tinnitus treatment, and their association with ITFs has also been studied (Doborjeh et al., 2023; Guntensperger et al., 2019; Henderson‐Sabes et al., 2019; Zimmerman et al., 2019). Therefore, brain imaging technology may be a good choice to detect the effect of tinnitus treatment medications targeting potassium ion channels on tinnitus patients.…”

An implementation science systematic review of neurophysiological evidence indicates the tinnitus core network as a therapeutic target

“…Future implementation of neurophysiological evidence is currently being supported by innovation across three streams: commercially available low-cost mobile neuroimaging, artificial intelligence, and investment in secure data-sharing infrastructure. For example, sound-based tinnitus treatment efficacy prediction (Sanders et al, 2021;Doborjeh et al, 2023) and mindfulness efficacy were able to be predicted with 0.87 accuracy (F-score, 89%) using EEG data fed into a brain-inspired spiking neural network model (Doborjeh et al, 2019).…”

Prediction of acoustic tinnitus suppression using resting state EEG: An explainable AI approach

“…Furthermore, this approach could foster (objective) diagnostic options, tinnitus subtyping, and identification of individual treatment options like sound therapies [38]. Related to that, it was recently shown that certain EEG features such as frequency band power and functional connectivity could predict treatment response to a sound-based intervention with 98-100% accuracy [39].…”