Publication detail

Skull Shape Reconstruction Using Cascaded Convolutional Networks

KODYM, O. ŠPANĚL, M. HEROUT, A.

Original Title

Skull Shape Reconstruction Using Cascaded Convolutional Networks

English Title

Skull Shape Reconstruction Using Cascaded Convolutional Networks

Type

journal article in Web of Science

Language

en

Original Abstract

Designing a cranial implant to restore the protective and aesthetic function of the patient's skull is a challenging process that requires a substantial amount of manual work, even for an experienced clinician. While computer-assisted approaches with various levels of required user interaction exist to aid this process, they are usually only validated on either a single type of simple synthetic defect or a very limited sample of real defects. The work presented in this paper aims to address two challenges: (i) design a fully automatic 3D shape reconstruction method that can address diverse shapes of real skull defects in various stages of healing and (ii) to provide an open dataset for optimization and validation of anatomical reconstruction methods on a set of synthetically broken skull shapes. We propose an application of the multi-scale cascade architecture of convolutional neural networks to the reconstruction task. Such an architecture is able to tackle the issue of trade-off between the output resolution and the receptive field of the model imposed by GPU memory limitations. Furthermore, we experiment with both generative and discriminative models and study their behavior during the task of anatomical reconstruction. The proposed method achieves an average surface error of 0.59 for our synthetic test dataset with as low as 0.48 for unilateral defects of parietal and temporal bone, matching state-of-the-art performance while being completely automatic. We also show that the model trained on our synthetic dataset is able to reconstruct real patient defects.

English abstract

Designing a cranial implant to restore the protective and aesthetic function of the patient's skull is a challenging process that requires a substantial amount of manual work, even for an experienced clinician. While computer-assisted approaches with various levels of required user interaction exist to aid this process, they are usually only validated on either a single type of simple synthetic defect or a very limited sample of real defects. The work presented in this paper aims to address two challenges: (i) design a fully automatic 3D shape reconstruction method that can address diverse shapes of real skull defects in various stages of healing and (ii) to provide an open dataset for optimization and validation of anatomical reconstruction methods on a set of synthetically broken skull shapes. We propose an application of the multi-scale cascade architecture of convolutional neural networks to the reconstruction task. Such an architecture is able to tackle the issue of trade-off between the output resolution and the receptive field of the model imposed by GPU memory limitations. Furthermore, we experiment with both generative and discriminative models and study their behavior during the task of anatomical reconstruction. The proposed method achieves an average surface error of 0.59 for our synthetic test dataset with as low as 0.48 for unilateral defects of parietal and temporal bone, matching state-of-the-art performance while being completely automatic. We also show that the model trained on our synthetic dataset is able to reconstruct real patient defects.

Keywords

Cranial implant design, Anatomical reconstruction, 3D shape completion, Convolutional neural networks, Generative adversarial networks

Released

26.08.2020

Publisher

NEUVEDEN

Location

NEUVEDEN

ISBN

0010-4825

Periodical

COMPUTERS IN BIOLOGY AND MEDICINE

Year of study

123

Number

103886

State

US

Pages from

1

Pages to

9

Pages count

9

URL

Documents

BibTex


@article{BUT168170,
  author="Oldřich {Kodym} and Michal {Španěl} and Adam {Herout}",
  title="Skull Shape Reconstruction Using Cascaded Convolutional Networks",
  annote="Designing a cranial implant to restore the protective and aesthetic function of
the patient's skull is a challenging process that requires a substantial amount
of manual work, even for an experienced clinician. While computer-assisted
approaches with various levels of required user interaction exist to aid this
process, they are usually only validated on either a single type of simple
synthetic defect or a very limited sample of real defects. The work presented in
this paper aims to address two challenges: (i) design a fully automatic 3D shape
reconstruction method that can address diverse shapes of real skull defects in
various stages of healing and (ii) to provide an open dataset for optimization
and validation of anatomical reconstruction methods on a set of synthetically
broken skull shapes.
We propose an application of the multi-scale cascade architecture of
convolutional neural networks to the reconstruction task. Such an architecture is
able to tackle the issue of trade-off between the output resolution and the
receptive field of the model imposed by GPU memory limitations. Furthermore, we
experiment with both generative and discriminative models and study their
behavior during the task of anatomical reconstruction.
The proposed method achieves an average surface error of 0.59 for our synthetic
test dataset with as low as 0.48 for unilateral defects of parietal and temporal
bone, matching state-of-the-art performance while being completely automatic. We
also show that the model trained on our synthetic dataset is able to reconstruct
real patient defects.",
  address="NEUVEDEN",
  chapter="168170",
  doi="10.1016/j.compbiomed.2020.103886",
  edition="NEUVEDEN",
  howpublished="online",
  institution="NEUVEDEN",
  number="103886",
  volume="123",
  year="2020",
  month="august",
  pages="1--9",
  publisher="NEUVEDEN",
  type="journal article in Web of Science"
}