The University of Cambridge’s “Nature” sub-issue: the application of cohesive 3D printing and artificial intelligence in cochlear simulation

Cochlear implants have transformed the silent lives of hundreds of thousands of people with severe or severe hearing loss. However, cochlear implants currently on the market are hindered by ‘current spread’ and cannot precisely stimulate the auditory nerve. Until now, no proper test model has been able to accurately reproduce the “current spreading” problem that occurs in the human cochlea.Based on this, Professor Shery Huang’s Biointerface Research Group at the University of Cambridge and Professor Manohar Bance used 3D printing technology to create the first bionic model with the shape of the human cochlea and the conductivity of cochlear tissue, and combined it with machine learning to achieve artificial intelligence. Cochlear ‘Current spread’
clinical
predict. Relevant research results were published in the journal Nature Communications on October 29 with the title “3D printed biomimetic cochleae and machine learning co-modelling provides clinical informatics for cochlear implant patients”. The first author of the paper is Li Yiwen, a doctoral student in the research group.

The problem of “current propagation” in cochlear implants is caused by a conductive lymphatic fluid (perilymph) in the cochlear duct. This problem severely limits the accuracy of the cochlear implant’s response to the auditory nerve, and can result in severely distorted sounds (especially music) being perceived by the cochlear implant user. In addition, the cochlear tissue itself is located deep in the temporal bone and has a relatively complex anatomical structure, and the shape and conductivity of the human cochlea have significant individual differences, resulting in limited existing test models (including animal models, human specimen models and computer models). Meta-analytical model) cannot fully and accurately simulate the problem of “current propagation” in the human cochlea.