Repository logo
 

3D printed biomimetic cochleae and machine learning co-modelling provides clinical informatics for cochlear implant patients

Published version
Peer-reviewed

Type

Article

Change log

Authors

Lei, Iek Man 
Jiang, Chen 
Lei, Chon Lok 
Rijk, Simone Rosalie de 
Tam, Yu Chuen 

Abstract

Cochlear implants (CIs) restore hearing in patients with severe to profound deafness by delivering electrical stimuli inside the cochlea. Understanding CI stimulus spread, and how it correlates to patient-dependent factors, is hampered by the poor accessibility of the inner ear and by the lack of suitable in vitro, in vivo or in silico models. Here, we present 3D printing-neural network co-modelling for interpreting clinical electric field imaging (EFI) profiles of CI patients. With tuneable electro-anatomy, the 3D printed cochleae were shown to replicate clinical scenarios of EFI profiles at the off-stimuli positions. The co-modelling framework demonstrated autonomous and robust predictions of patient EFI or cochlear geometry, unfolded the electro-anatomical factors causing CI stimulus spread, assisted on-demand printing for CI testing, and inferred patients’ in vivo cochlear tissue resistivity (estimated mean = 6.6 kΩcm) by CI telemetry. We anticipate our framework will facilitate physical modelling and digital twin innovations for electrical prostheses in healthcare.

Description

Keywords

4201 Allied Health and Rehabilitation Science, 32 Biomedical and Clinical Sciences, 3202 Clinical Sciences, 42 Health Sciences, Rehabilitation, Assistive Technology, Bioengineering

Journal Title

Nature Communications

Conference Name

Journal ISSN

2041-1723

Volume Title

Publisher

Nature Research
Sponsorship
European Research Council (758865)
Wellcome Trust (204845/Z/16/Z)
This work was supported by the European Research Council (ERC-StG, 758865), the Cambridge Hearing Trust and the Evelyn Trust. I.M.L. acknowledges the financial support from the W D Armstrong Trust and the Macao Postgraduate Scholarship Fund. C.J. acknowledges the support from the Wellcome Trust (204845/Z/16/Z). C.L.L. acknowledges the support from the University of Macau via a UM Macao Fellowship and the Clarendon Scholarship Fund. S.R.D.R. acknowledges the financial support from the Baroness de Turckheim Fund, Trinity College Cambridge. The authors acknowledge the Henry Royce Institute Cambridge Equipment (EP/P024947/1).