!!!    AutoImplant challenge will be held at MICCAI 2021 (https://autoimplant2021.grand-challenge.org/)! 

!!!    (Jan 14th 2021) Call for Sponsorship. We are looking for company partners/sponsorship. Drop us an email if interested.

Jan Egger (egger AT tugraz.at), Jianning Li (jianning.li AT icg.tugraz.at).

!!!   (October 8th 2020) AutoImplant 2020 AWARDS have been announced.  Congratulations to the Authors!

!!!  (September 25 2020) Our Programs and Leaderboard are online! 

!!!  (September 20 2020) Author notifications sent.

!!!  Results &  paper submission deadline extended to September 14th, 2020. 

!!! (August 10 2020)  Authors of the top three best papers will be awarded with the AutoImplant MICCAI 2020 Best Paper Award.

Best Paper Award!

!!!  (July 31 2020) 

Call for Papers!

 The test set is now released on the Dataset page!  Details for result submission and paper submission can be found on the Submission page.

!!!  (July 28 2020)   Challenge proceedings:  We plan to publish challenge papers (6-12 pages) in Lecture Notes in Computer Science (LNCS).  Please make early preparation if you want to submit your papers!  Detailed submission instructions will be announced shortly.   

!!!  (July 01 2020)   Follow our official Twitter account to get informed of the latest updates of the challenge in time!

!!! (June 23 2020)  Announcement: A Baseline Approach for the AutoImplant Challenge [pdf][codes]!

!!!  (June 03 2020)  The training set is now released on the Dataset page!

If you do not get a participation notification, please check your registration details (please indicate in your profile if you are an independent researcher without affiliation).  Inquires related to the challenge can be addressed to autoimplant.challenge@gmail.com.


Cranioplasty is the surgical process where a skull defect, caused by a brain tumor surgery or trauma, is repaired using a cranial implant, which must fit precisely against the borders of the skull defect as replacement to the removed cranial bone. The design of the cranial implant is a challenging task and involves several steps: (1) obtaining the 3D imaging data of the skull with the defect from CT or MRI, (2) converting the 3D imaging data into a 3D mesh model and, (3) creating an implant as 3D mesh model for 3D printing. The last step usually requires expensive commercial software, which clinical institutions often have limited access to.  A video shows an example of interactive cranial implant design using  Geomagic Sculpt.  Researchers have been working with common CAD software as an alternative to the commercial software for the design of cranial implants whereas these approaches still involve intensive human interaction, which is time-consuming and requires expertise in the specific medical domain (see also a video for an example of cranial implant design performed using MeVisLab). Therefore, a fast and automatic design of cranial implants is highly desired, which would also enable  in-operation room (in-OR) manufacturing of the implants for the patient when additive manufacturing (AM) is combined.

keywords:  craniotomy,  cranioplasty, cranial implant design,  skull reconstruction, deep learning, volumetric shape completion

Suggested problem formulation

Centered around this topic, our challenge provides 200  high-resolution (512x512xZ) skulls and seeks for data-driven approaches for the problem of automatic cranial implant design. We injected artificial defects into each healthy skull to simulate the process of skull opening in a surgery (craniotomy), thus creating training/evaluation  data pairs [6].  Participants are expected to design data-driven algorithms (such as deep learning) based on these data pairs for automatic skull defect restoration (skull reconstruction) and implant generation.  In this sense, the problem is being formulated as a 3D volumetric shape completion task , where a defective skull volume is automatically completed by the algorithms [1] [3] [5]. The restored defect, which is in fact the implant we want, can be obtained by the subtraction of the defective skull from the completed skull.  Another solution is to predict the implant directly from a defective skull  [4].  

How to participate

The AutoImplant challenge, which is a half day MICCAI event,  will be officially presented at  MICCAI 2020 virtual conference on 08 October 2020 AM.  The challenge website is currently based on the grand-challenge platform. Users can request participation on this website to get access to the training and test dataset.  Please note that we will only approve requests from users with a complete profile (i.e., real name, valid email address and valid affiliation).  The presentation of the submitted methods and  results will be organized as part of the MICCAI virtual conference.  


  • release of the training set:  June 03 2020 (completed!)
  • release of the test set :  Aug. 01 2020 July 31,2020 (completed!)
  • submission opens:  September 01 2020 --September 10, 2020     Aug. 03 2020
  • paper and result submission closes:  31 August 2020, 11:59 PM  14 September 2020, 11:59 PM
  • presentation of  the (selected) methods/results:  on 08 October 2020 AM, MICCAI virtual conference

  • [1] Jianning Li, Antonio Pepe, Christina Gsaxner, Jan Egger.  An Online Platform for Automatic Skull Defect Restoration and Cranial Implant Design (1 Jun 2020), arXiv:2006.00980
  • [2] Jan Egger, Jianning Li, Xiaojun Chen, Ute Schäfer, Gord von Campe, Marcell Krall, Ulrike Zefferer, Christina Gsaxner, Antonio Pepe, Dieter Schmalstieg. (2020, March 19).  Towards the Automatization of Cranial Implant Design in Cranioplasty. Zenodo. http://doi.org/10.5281/zenodo.3715953 Bibtex
  • [3] Ana Morais,  Jan Egger,  Victor Alves.   Automated computer-aided design of cranial implants using a deep volumetric convolutional denoising autoencoder, World Conference on Information Systems and Technologies 2019 (World-CIST’19):  New  Knowledge  in  Information  Systems  and Technologies, pp. 151–160, 2019
  • [4] Jianning Li, Antonio Pepe, Christina Gsaxner, Gord von Campe, Jan Egger.  A Baseline Approach for AutoImplant: the MICCAI 2020 Cranial Implant Design Challenge.  MICCAI CLIP 2020.  arxiv: 2006.12449
  • [5] Jianning Li. "Deep Learning for Cranial Defect Reconstruction".  Master Thesis, Institute of Computer Graphics and Vision, Graz University of Technology, Austria, pp. 1-68, January 2020.  Bibtex
  • [6] Jianning  Li  and  Jan  Egger. Towards  the  automatization  of  cranial  implant  design  for  3d printing. 3D-Printing Lab Opening and General Assembly Meeting (COMET K-Project CAMed), October, 2019.


Number of users: 196