The Transimpedance Matrix (TIM) is an objective electrophysiological measurement that enables the analysis of the distribution of electrical potential within the cochlea. It is obtained by sequentially stimulating the individual electrodes of the cochlear implant and recording the electrical potential generated via the other electrodes in the array. The relationship between the recorded potential and the delivered current enables the construction of a matrix describing the spatial distribution of transimpedance along the electrode array. Recent studies suggest that the TIM pattern may provide information about the intracochlear environment and exhibit different characteristics depending on the underlying pathology. The aim of this study was to characterise TIM patterns in patients with Cochlear cochlear implants and different aetiologies of hearing loss, using the GJB2 group as a reference. A total of 191 patients were analysed, comprising 259 implants and 382 TIM sessions. For each aetiological group, descriptive statistics were calculated and mean 22×22 TIM matrices and corresponding heatmaps were constructed, in order to compare both the values and the spatial distribution. The study included a cross-sectional analysis across the different aetiologies and a longitudinal analysis in patients with repeated measurements. The relationship between TIM patterns and facial nerve stimulation was also explored. The GJB2 group, comprising 32 patients and 65 TIM sessions, had a mean extra-diagonal value of 1260.3 Ω. The cross-sectional analysis revealed marked variability across the different aetiologies, both in the overall values and in the spatial distribution of TIM. The CHARGE group showed the highest values (2,867.6 Ω), followed by Neurinoma (1,728.9 Ω) and Generalised Malformation (1,591.8 Ω). Longitudinal analysis also revealed a trend towards an increase in TIM over time, with an average change of +16.8 Ω/year, albeit with differences between the subgroups analysed. In patients with facial nerve stimulation, higher TIM values were observed compared with the GJB2 reference; this difference diminishes, but remains present, after excluding CHARGE patients. The results highlight the presence of quantitative and spatial differences in TIM patterns across different aetiologies, suggesting that the interpretation of TIM may benefit from analysis of the entire matrix rather than just individual impedance values. The data obtained do not yet allow for the definition of a specific TIM for each aetiology, but they provide an initial basis for identifying reference TIM patterns. The study therefore aims to broaden the scope of interpretation of the Transimpedance Matrix, moving from a predominantly qualitative assessment of the heatmap to a quantitative and comparative approach. The identification of TIM patterns characteristic of different aetiologies could, in the long term, form the basis for the development of algorithms to support diagnosis and clinical management in patients with cochlear implants.
La Transimpedance Matrix (TIM) è una misura elettrofisiologica oggettiva che permette di analizzare la distribuzione del potenziale elettrico all’interno della coclea. Viene ottenuta stimolando in sequenza i singoli elettrodi dell’impianto cocleare e registrando il potenziale elettrico generato attraverso gli altri elettrodi dell’array. Il rapporto tra il potenziale registrato e la corrente erogata consente di costruire una matrice che descrive la distribuzione spaziale della transimpedenza lungo l’array elettrodico. Studi recenti suggeriscono che la configurazione della TIM possa fornire informazioni sull’ambiente intracocleare e presentare caratteristiche differenti in relazione alla patologia sottostante. Obiettivo del presente studio è stato caratterizzare i pattern TIM in pazienti portatori di impianto cocleare Cochlear con differenti eziologie dell’ipoacusia, utilizzando il gruppo GJB2 come riferimento. Sono stati analizzati 191 pazienti, per un totale di 259 impianti e 382 sessioni TIM. Per ciascun gruppo eziologico sono state calcolate statistiche descrittive e costruite matrici TIM medie 22×22 e relative heatmap, al fine di confrontare sia i valori sia la distribuzione spaziale. Lo studio ha incluso un’analisi trasversale tra le differenti eziologie e un’analisi longitudinale nei pazienti con misurazioni ripetute. È stata inoltre esplorata la relazione tra pattern TIM e stimolazione del nervo facciale. Il gruppo GJB2, costituito da 32 pazienti e 65 sessioni TIM, presentava un valore medio extra-diagonale di 1260,3 Ω. L’analisi trasversale ha evidenziato una marcata variabilità tra le diverse eziologie, sia nei valori complessivi sia nella distribuzione spaziale della TIM. Il gruppo CHARGE mostrava i valori più elevati (2867,6 Ω), seguito da Neurinoma (1728,9 Ω) e Malformativa Generica (1591,8 Ω). L’analisi longitudinale ha inoltre evidenziato una tendenza all’aumento della TIM nel tempo, con una variazione media di +16,8 Ω/anno, pur con differenze tra i sottogruppi analizzati. Nei pazienti con stimolazione del nervo facciale sono stati osservati valori TIM superiori rispetto al riferimento GJB2; tale differenza si riduce, ma rimane presente, dopo l’esclusione dei pazienti CHARGE. I risultati evidenziano la presenza di differenze quantitative e spaziali nei pattern TIM delle diverse eziologie, suggerendo che l’interpretazione della TIM possa beneficiare dell’analisi dell’intera matrice e non soltanto dei singoli valori di impedenza. I dati ottenuti non consentono ancora di definire una TIM specifica per ciascuna eziologia, ma costituiscono una prima base per l’individuazione di pattern TIM di riferimento. Lo studio si propone quindi di ampliare le possibilità di interpretazione della Transimpedance Matrix, passando da una valutazione prevalentemente qualitativa della heatmap a un approccio quantitativo e comparativo. L’identificazione di pattern TIM caratteristici delle differenti eziologie potrebbe rappresentare, in prospettiva, la base per lo sviluppo di algoritmi di supporto diagnostico e clinico nei pazienti portatori di impianto cocleare.
ANALISI COMPUTAZIONALE DELLA TRANSIMPEDANCE MATRIX MEDIANTE MACHINE LEARNING: STUDIO PILOTA DI FENOTIPIZZAZIONE EZIOLOGICA
CESARIN, ALESSIA ELEONORA
2025/2026
Abstract
The Transimpedance Matrix (TIM) is an objective electrophysiological measurement that enables the analysis of the distribution of electrical potential within the cochlea. It is obtained by sequentially stimulating the individual electrodes of the cochlear implant and recording the electrical potential generated via the other electrodes in the array. The relationship between the recorded potential and the delivered current enables the construction of a matrix describing the spatial distribution of transimpedance along the electrode array. Recent studies suggest that the TIM pattern may provide information about the intracochlear environment and exhibit different characteristics depending on the underlying pathology. The aim of this study was to characterise TIM patterns in patients with Cochlear cochlear implants and different aetiologies of hearing loss, using the GJB2 group as a reference. A total of 191 patients were analysed, comprising 259 implants and 382 TIM sessions. For each aetiological group, descriptive statistics were calculated and mean 22×22 TIM matrices and corresponding heatmaps were constructed, in order to compare both the values and the spatial distribution. The study included a cross-sectional analysis across the different aetiologies and a longitudinal analysis in patients with repeated measurements. The relationship between TIM patterns and facial nerve stimulation was also explored. The GJB2 group, comprising 32 patients and 65 TIM sessions, had a mean extra-diagonal value of 1260.3 Ω. The cross-sectional analysis revealed marked variability across the different aetiologies, both in the overall values and in the spatial distribution of TIM. The CHARGE group showed the highest values (2,867.6 Ω), followed by Neurinoma (1,728.9 Ω) and Generalised Malformation (1,591.8 Ω). Longitudinal analysis also revealed a trend towards an increase in TIM over time, with an average change of +16.8 Ω/year, albeit with differences between the subgroups analysed. In patients with facial nerve stimulation, higher TIM values were observed compared with the GJB2 reference; this difference diminishes, but remains present, after excluding CHARGE patients. The results highlight the presence of quantitative and spatial differences in TIM patterns across different aetiologies, suggesting that the interpretation of TIM may benefit from analysis of the entire matrix rather than just individual impedance values. The data obtained do not yet allow for the definition of a specific TIM for each aetiology, but they provide an initial basis for identifying reference TIM patterns. The study therefore aims to broaden the scope of interpretation of the Transimpedance Matrix, moving from a predominantly qualitative assessment of the heatmap to a quantitative and comparative approach. The identification of TIM patterns characteristic of different aetiologies could, in the long term, form the basis for the development of algorithms to support diagnosis and clinical management in patients with cochlear implants.| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI CESARIN- TIM 2.pdf
accesso aperto
Dimensione
6.57 MB
Formato
Adobe PDF
|
6.57 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/119529