This was a question hearing implant company Cochlear BAS was facing with the introduction of their new Baha® Attract bone conduction implant system. To validate their design, they turned to Materialise’s Anatomical Data Mining (ADaM™) services. Using a database of CT scans, Materialise virtually implanted the system and analyzed the fit, giving Cochlear BAS the evidence it needed to support its product launch.

Developing a Subcutaneous Bone Conduction Implant System

Bone conduction implants mimic our natural ability to transfer sound through vibrations in the bone, rather than forcing sound through problem zones in the outer or middle ear. Skin-penetrating abutments are standard in bone conduction implant systems, but a drawback is the aesthetic aspect of the visible abutment. That is why Cochlear BAS launched its Baha® Attract System, a bone conduction implant system with a magnetic connection between the implant and the sound processor. The implant magnet is inserted subcutaneously and vibrations from the sound processor are transmitted through the skin via an external magnet.

Baha Attract System


Testing the Circular Magnet on 50 Virtual Patients 

Cochlear wanted to validate three aspects of their system design. First, they wanted to know the overlap between the magnet and cranium to predict the need for reaming during implantation. Second, they needed to quantify bone thickness to validate implant screw length. Third, they wanted to examine the thickness of the soft tissue to validate the strength of the external magnet.

Materialise acquired a set of 50 CT scans of crania with no visible conditions to the cranium, equally divided by gender. The sample consisted of 22 scans from the USA and 28 from Europe of people ranging between 16 and 83 years old.

The 50 crania were segmented from CT scans using Mimics® Research and converted into 3D models. Subsequently, they determined the cylindrical region of interest (ROI) using 3-matic® Research with a radius of 20 mm for each subject. They virtually implanted a circular magnet with a radius of 13 mm and analyzed the bone and skin surfaces at the location of the implant. Next, the distances between the different surfaces were determined and statistically analyzed. Materialise then provided Cochlear with a report with the findings from the study.


Illustration of the region of interest and virtual implantation of the magnet


"With data from Materialise we now have statistically confirmed information that gives us a better understanding of the anatomy where our implants are to be placed. With that knowledge we can optimize our implants and surgical instruments for an improved final result.”
– Jerry Frimanson, Cochlear BAS, Sweden

Population-Based Evidence to Support a Good Implant Design

The population analysis suggested that minor reaming would be needed in about 20% of the cases. Eleven subjects showed an overlap between the inner magnet and outer bone surface over an estimated area of more than 10 mm². The strength of the external magnet was adequate for the measured soft tissue thickness distribution. Finally, the bone thickness distribution gave valuable information for determining the length range for the screw that the surgeon would insert into the bone to fix the magnet.

Average bone thickness ranged between 5mm (red) and 25mm (green)

Minimum bone thickness ranged between 2mm (red) and 7mm (green).



Baha is a registered trademark of Cochlear Bone Anchored Solutions AB. Provided images: © Cochlear Bone Anchored Solutions AB 2015. All rights reserved

Regulatory Information:

The Research edition of the Mimics® Innovation Suite currently consists of the following software components: Mimics® Research version 18.0 and 3-matic®Research version 10.0 (released 2015). Mimics® Research is intended only for research purposes. It is intended as a software interface and image segmentation system for the transfer of imaging information from a variety of imaging sources to an output file. It is also used as software for simulating, measuring and modeling in the field of biomedical research. “Mimics® Research” must not be used, and is not intended to be used, for any medical purpose whatsoever. 3-matic® Research is intended for use as software for computer assisted design and engineering in the field of biomedical research. “3-matic® Research” must not be used, and is not intended to be used, for the design or manufacturing of medical devices of any kind.

Materialise Belgium – Technologielaan 15 – 3000 Leuven – Belgium 

Copyright 2015 Materialise N.V – L -10188 revision 1, 06/2015

Read other cases about

Industries Solutions