Design of 3D-printed cranioplasty moulds using Neural Network
(2020) In Master’s Theses in Mathematical Sciences FMAM05 20202Mathematics (Faculty of Engineering)
- Abstract
- Cranioplasty is surgical repair of a skull bone defect due to a previous surgery or injury. Cranioplasty is most often performed with autologous bone flap, i.e the patient's own saved bone from previous surgery if this is available. If autologous bone is not available then custom procedure is to manually mould an implant using bone cement from the plastic polymethyl methacrylate (PMMA). Manually moulding implants during surgery has, albeit being a clinical routine, disadvantages and therefore Skåne University Hospital have developed a technique to 3D print patient-specific cranioplasty moulds based on a computed tomography (CT) scan of the skull bone. The shape of the mould is created by a combination of manual design process, mirroring... (More)
- Cranioplasty is surgical repair of a skull bone defect due to a previous surgery or injury. Cranioplasty is most often performed with autologous bone flap, i.e the patient's own saved bone from previous surgery if this is available. If autologous bone is not available then custom procedure is to manually mould an implant using bone cement from the plastic polymethyl methacrylate (PMMA). Manually moulding implants during surgery has, albeit being a clinical routine, disadvantages and therefore Skåne University Hospital have developed a technique to 3D print patient-specific cranioplasty moulds based on a computed tomography (CT) scan of the skull bone. The shape of the mould is created by a combination of manual design process, mirroring and interpolation. This partly limits the technique to unilateral defects and the method can be tricky and time consuming for complicated cases.
The purpose of the thesis was to develop a method based on a neural network to reconstruct missing parts of the skull bone and overcome the limitations with the current mirroring method when designing implants or moulds for cranioplasty.
The process included developing a method to extract data from CT images for training of the neural networks. During the process numerous neural network structures and models were developed and evaluated with the best performing network being a convolutional autoencoder with skip connections. The network was trained with data from a total of 240 patients with simulated defects. The results of the network shows that it is able to handle both unilateral and bilateral defects with a mean error of 1.07mm. In comparison to the currently used method it performed as well or even better in some cases. Overall the developed method showed good enough results for it to be implemented as clinical routine. (Less) - Popular Abstract (Swedish)
- Kranioplastik är ett kirurgiskt ingrepp där en del av skallen rekonstrueras på grund av estetiska eller medicinska skäl. I de flesta av fallen är kranioplastiken gjord med patientens egna skallben som har sparats från ett tidigare kirurgiskt ingrepp men om skallbenet saknas på grund av en defekt eller annan anledning så behövs ett implantat. Metoden för att designa implantatet har tidigare bestått av en kombination med subjektiva designbeslut och att man speglar den friska halvan av skallen. Det gör att metoden begränsas till unilaterala skador, det vill säga skador som inte sträcker sig över båda sidorna av skallen. Därför har jag i det här projektet utforskat möjligheten att använda ett neuralt nätverk i processen att designa... (More)
- Kranioplastik är ett kirurgiskt ingrepp där en del av skallen rekonstrueras på grund av estetiska eller medicinska skäl. I de flesta av fallen är kranioplastiken gjord med patientens egna skallben som har sparats från ett tidigare kirurgiskt ingrepp men om skallbenet saknas på grund av en defekt eller annan anledning så behövs ett implantat. Metoden för att designa implantatet har tidigare bestått av en kombination med subjektiva designbeslut och att man speglar den friska halvan av skallen. Det gör att metoden begränsas till unilaterala skador, det vill säga skador som inte sträcker sig över båda sidorna av skallen. Därför har jag i det här projektet utforskat möjligheten att använda ett neuralt nätverk i processen att designa skallimplantatet och på så vis övervinna de begränsningarna som fanns med den gamla metoden samt minska de subjektiva designbesluten.
Det neurala nätverket som presterade bäst var en autoencoder med faltning som tränades och utvärderades på totalt 300 skallar. Metoden gav så pass bra resultat att den kunde börja användas som klinisk rutin på Skånes universitetssjukhus i Lund där man alltså nu kan få ett skallimplantat designat med hjälp av artificiell intelligens. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9036345
- author
- Andersson, Jonathan LU
- supervisor
- organization
- alternative title
- Design av 3D-printade gjutformar för kranioplastik med hjälp av Neurala Nätverk
- course
- FMAM05 20202
- year
- 2020
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Cranioplasty, Neural Network, Convolutional Neural Network, autoencoder, 3D printing, data reconstruction
- publication/series
- Master’s Theses in Mathematical Sciences
- report number
- LUTFMA-3434-2020
- ISSN
- 1404-6342
- other publication id
- 2020:E86
- language
- English
- id
- 9036345
- date added to LUP
- 2021-01-25 16:47:30
- date last changed
- 2021-01-25 16:47:30
@misc{9036345,
abstract = {{Cranioplasty is surgical repair of a skull bone defect due to a previous surgery or injury. Cranioplasty is most often performed with autologous bone flap, i.e the patient's own saved bone from previous surgery if this is available. If autologous bone is not available then custom procedure is to manually mould an implant using bone cement from the plastic polymethyl methacrylate (PMMA). Manually moulding implants during surgery has, albeit being a clinical routine, disadvantages and therefore Skåne University Hospital have developed a technique to 3D print patient-specific cranioplasty moulds based on a computed tomography (CT) scan of the skull bone. The shape of the mould is created by a combination of manual design process, mirroring and interpolation. This partly limits the technique to unilateral defects and the method can be tricky and time consuming for complicated cases.
The purpose of the thesis was to develop a method based on a neural network to reconstruct missing parts of the skull bone and overcome the limitations with the current mirroring method when designing implants or moulds for cranioplasty.
The process included developing a method to extract data from CT images for training of the neural networks. During the process numerous neural network structures and models were developed and evaluated with the best performing network being a convolutional autoencoder with skip connections. The network was trained with data from a total of 240 patients with simulated defects. The results of the network shows that it is able to handle both unilateral and bilateral defects with a mean error of 1.07mm. In comparison to the currently used method it performed as well or even better in some cases. Overall the developed method showed good enough results for it to be implemented as clinical routine.}},
author = {{Andersson, Jonathan}},
issn = {{1404-6342}},
language = {{eng}},
note = {{Student Paper}},
series = {{Master’s Theses in Mathematical Sciences}},
title = {{Design of 3D-printed cranioplasty moulds using Neural Network}},
year = {{2020}},
}