Předmětem spolupráce je analýza současných trendů v moderních metodách hlubokého učení pro porozumění 3D datům ve formě obrazových i povrchových 3D dat. Tvorbě analýz vhodnosti využití nových metod pro potřeby a úlohy definované Partnerem a experimentálním ověření vlastností vybraných metod a konzultacích již existujících řešení Partnera a technické pomoci při jejich inovaci.

Description in English
This project aims at analysis of current trends in state-of-the-art deep learning methods for 3D image and surface data. Analysis of suitability of new approaches for the needs and tasks defined by the company, experimental implementation and verification of the promising methods, and consultations of existing solutions of the company and technical assistance in their innovation.

Keywords
hluboké učení, konvoluční neuronové sítě, CNN, analýza 3D tvaru, rozpoznávání obrazových dat

Key words in English
Deep learning, convolutional neural networks, CNN, 3D shape analysis, image recognition

Czech

Sadovský Petr, Ing., Ph.D. - principal person responsible

Research Centre of Information Technology
- responsible department (23.1.2020 - not assigned)
TESCAN 3DIM, s.r.o.
- client (23.1.2020 - 30.6.2021)
Department of Computer Graphics and Multimedia
- co-beneficiary (23.1.2020 - 30.6.2021)
Research Centre of Information Technology
- beneficiary (23.1.2020 - 30.6.2021)

KODYM, O.; ŠPANĚL, M.; HEROUT, A. Skull Shape Reconstruction Using Cascaded Convolutional Networks. Computers in Biology and Medicine, 2020, vol. 123, no. 103886, p. 1-9. ISSN: 0010-4825.
Detail

DREVICKÝ, D.; KODYM, O. Evaluating deep learning uncertainty measures in cephalometric landmark localization. In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 2: BIOIMAGING. Valetta: Institute for Systems and Technologies of Information, Control and Communication, 2020. p. 213-220. ISBN: 978-989-758-398-8.
Detail

KODYM, O.; ŠPANĚL, M.; HEROUT, A. Segmentation of Defective Skulls from CT Data for Tissue Modelling. In Towards the Automatization of Cranial Implant Design in Cranioplasty II. Lecture Notes in Computer Science. Lecture Notes in Computer Science. Strasbourg: Springer Nature Switzerland AG, 2021. no. 13123, p. 19-28. ISBN: 978-3-030-92652-6. ISSN: 0302-9743.
Detail

KODYM, O.; LI, J.; PEPE, A.; GSAXNER, C.; EGGER, J.; ŠPANĚL, M. SkullBreak/SkullFix - Dataset for automatic cranial implant design and a benchmark for volumetric shape learning tasks. Data in Brief, 2021, vol. 35, no. 106902, p. 1-7. ISSN: 2352-3409.
Detail

KODYM, O.; ŠPANĚL, M.; HEROUT, A. Deep Learning for Cranioplasty in Clinical Practice: Going from Synthetic to Real Patient Data. Computers in Biology and Medicine, 2021, vol. 137, no. 104766, p. 1-10. ISSN: 0010-4825.
Detail

KODYM, O. Souhrnná výzkumná zpráva k projektu TESCAN 3DIM - Metody hlubokého učení pro analýzu 3D dat. Brno: TESCAN 3DIM, s.r.o., 2020. 3 s.
Detail

LI, J.; ELLIS, D.; KODYM, O.; HEROUT, A.; ŠPANĚL, M.; EGGER, J. Towards clinical applicability and computational efficiency in automatic cranial implant design: An overview of the AutoImplant 2021 cranial implant design challenge. MEDICAL IMAGE ANALYSIS, 2023, vol. 88, no. 102865, p. 1-15. ISSN: 1361-8423.
Detail

KODYM, O.; ŠPANĚL, M.; HEROUT, A. Cranial Defect Reconstruction Using Cascaded CNN with Alignment. In Towards the Automatization of Cranial Implant Design in Cranioplasty. Lecture Notes in Computer Science. Lima: Springer Nature Switzerland AG, 2020. p. 56-64. ISBN: 978-3-030-64326-3.
Detail

Responsibility: Sadovský Petr, Ing., Ph.D.