Koen Willemsen

Dr. Koen Willemsen

Coordinator 3D lab
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 January 2022 in Journal of Magnetic Resonance Imaging
Verified
Journal Article
Open access
PDF file
MFMateusz C. Florkow
Mateusz C. Florkow
KWKoen Willemsen
Koen Willemsen
VMVasco V. Mascarenhas
Vasco V. Mascarenhas
EOEdwin H.G. Oei
Edwin H.G. Oei
MSMarijn van Stralen
Marijn van Stralen
PSPeter R. Seevinck
Peter R. Seevinck
Magnetic resonance imaging (MRI) is increasingly utilized as a radiation‐free alternative to computed tomography (CT) for the diagnosis and treatment planning of musculoskeletal pathologies. MR imaging of hard tissues such as cortical bone remains challenging due to their low proton density and short transverse relaxation times, rendering bone tissues as nonspecific low signal structures on MR images obtained from most sequences. Developments in MR image acquisition and post‐processing have opened the path for enhanced MR‐based bone visualization aiming to provide a CT‐like contrast and, as such, ease clinical interpretation. The purpose of this review is to provide an overview of studies comparing MR and CT imaging for diagnostic and treatment planning purposes in orthopedic care, with a special focus on selective bone visualization, bone segmentation, and three‐dimensional (3D) modeling. This review discusses conventional gradient‐echo derived techniques as well as dedicated short echo time acquisition techniques and post‐processing techniques, including the generation of synthetic CT, in the context of 3D and specific bone visualization. Based on the reviewed literature, it may be concluded that the recent developments in MRI‐based bone visualization are promising. MRI alone provides valuable information on both bone and soft tissues for a broad range of applications including diagnostics, 3D modeling, and treatment planning in multiple anatomical regions, including the skull, spine, shoulder, pelvis, and long bones.Level of Evidence: 3Technical Efficacy: Stage 3

8 October 2019 in Magnetic Resonance in Medicine
Verified
Journal Article
Open access
PDF file
MFMateusz C. Florkow
Mateusz C. Florkow
FZFrank Zijlstra
Frank Zijlstra
KWKoen Willemsen
Koen Willemsen
MMMatteo Maspero
Matteo Maspero
CBCornelis A. T. van Den Berg
Cornelis A. T. van Den Berg
PSPeter R. Seevinck
Peter R. Seevinck
Purpose: To study the influence of gradient echo–based contrasts as input channels to a 3D patch‐based neural network trained for synthetic CT (sCT) generation in canine and human populations.Methods: Magnetic resonance images and CT scans of human and canine pelvic regions were acquired and paired using nonrigid registration. Magnitude MR images and Dixon reconstructed water, fat, in‐phase and opposed‐phase images were obtained from a single T1‐weighted multi‐echo gradient‐echo acquisition. From this set, 6 input configurations were defined, each containing 1 to 4 MR images regarded as input channels. For each configuration, a UNet‐derived deep learning model was trained for synthetic CT generation. Reconstructed Hounsfield unit maps were evaluated with peak SNR, mean absolute error, and mean error. Dice similarity coefficient and surface distance maps assessed the geometric fidelity of bones. Repeatability was estimated by replicating the training up to 10 times.Results: Seventeen canines and 23 human subjects were included in the study. Performance and repeatability of single‐channel models were dependent on the TE‐related water–fat interference with variations of up to 17% in mean absolute error, and variations of up to 28% specifically in bones. Repeatability, Dice similarity coefficient, and mean absolute error were statistically significantly better in multichannel models with mean absolute error ranging from 33 to 40 Hounsfield units in humans and from 35 to 47 Hounsfield units in canines.Conclusion: Significant differences in performance and robustness of deep learning models for synthetic CT generation were observed depending on the input. In‐phase images outperformed opposed‐phase images, and Dixon reconstructed multichannel inputs outperformed single‐channel inputs.

1 August 2019 in The Lancet Digital Health
Verified
Journal Article
Open access
KWKoen Willemsen
Koen Willemsen
RNRazmara Nizak
Razmara Nizak
HNHerke Jan Noordmans
Herke Jan Noordmans
RCRené M Castelein
René M Castelein
HWHarrie Weinans
Harrie Weinans
MKMoyo C Kruyt
Moyo C Kruyt
Additive manufacturing or three-dimensional (3D) printing of metal implants can provide novel solutions for difficult-to-treat conditions, yet legislation concerning patient-specific implants complicates the implementation of these techniques in daily practice. In this Article, we share our acquired knowledge of the logistical and legal challenges associated with the use of patient-specific 3D-printed implants to treat spinal instabilities. Two patients with semiurgent cases of spinal instability presented to our hospital in the Netherlands. In case 1, severe kyphotic deformity of the thoracic spine due to neurofibromatosis type 1 had led to incomplete paralysis, and a strong metallic strut extending from C6 to T11 was deemed necessary to provide long-term anterior support. In case 2, the patient presented with progressive paralysis caused by cervicothoracic dissociation due to vanishing bone disease. As the C5-T1 vertebral bodies had mostly vanished, an implant spanning the anterior spine from C4 to T2 was required. Because of the complex and challenging nature of both cases, conventional approaches were deemed inadequate; instead, patient-specific implants were designed with use of CT scans and computer-aided design software, and 3D printed in titanium with direct metal printing. For each implant, to ensure patient safety, a comprehensive technical file (describing the clinical substantiation, technical and design considerations, risk analysis, manufacturing process, and labelling) was produced in collaboration with a university department certified for the development and manufacturing of medical devices. Because the implants were categorised as custom-made or personalised devices under the EU Medical Device Regulation, the usual procedures for review and approval of medical devices by a notified body were not required. Finite-element analyses, compression strength tests, and cadaveric experiments were also done to ensure the devices were safe to use. The planning, design, production, and insertion of the 3D-printed personalised implant took around 6 months in the first patient, but, given the experience from the first case, only took around 6 weeks in the second patient. In both patients, the surgeries went as planned and good positioning of each implant was confirmed. Both patients were discharged home within 1 week after the surgery. In the first patient, a fatigue fracture occured in one of the conventional posterior fusion rods after 10 months, which we repaired, without any deformation of the spine or signs of failure of the personalised implant observed. No other adverse events occurred up to 25 months of follow-up in case 1 and 6 months of follow-up in case 2. Patient-specific treatment approaches incorporating 3D-printed implants can be helpful in carefully selected cases when conventional methods are not an option. Comprehensive and efficient interactions between medical engineers and physicians are essential to establish well designed frameworks to navigate the logistical and regulatory aspects of technology development to ensure the safety and legal validity of patient-specific treatments. The framework described here could encourage physicians to treat (once untreatable) patients with novel personalised techniques. Interreg VA Flanders-The Netherlands programme, Applied and Engineering Sciences research programme, the Netherlands Organisation for Scientific Research, and the Dutch Arthritis Foundation VIDEO

1 April 2021 in Acta Biomaterialia
Verified
Journal Article
Open access
HKH.M.A. Kolken
H.M.A. Kolken
CJC.P. de Jonge
C.P. de Jonge
TST. van der Sloten
T. van der Sloten
AGA. Fontecha Garcia
A. Fontecha Garcia
BPB. Pouran
B. Pouran
AZA.A. Zadpoor
A.A. Zadpoor
The unprecedented properties of meta-biomaterials could pave the way for the development of life-lasting orthopedic implants. Here, we used non-auxetic meta-biomaterials to address the shortcomings of the current treatment options in acetabular revision surgery. Due to the severe bone deficiencies and poor bone quality, it can be very challenging to acquire adequate initial implant stability and long-term fixation. More advanced treatments, such as patient-specific implants, do guarantee the initial stability, but are formidably expensive and may eventually fail due to stress shielding. We, therefore, developed meta-implants furnished with a deformable porous outer layer. Upon implantation, this layer plastically deforms into the defects, thereby improving the initial stability and homogeneously stimulating the surrounding bone. We first studied the space-filling behavior of additively manufactured pure titanium lattices, based on six different unit cells, in a compression test complemented with full-field strain measurements. The diamond, body-centered cubic, and rhombic dodecahedron unit cells were eventually selected for the design of the deformable porous outer layer. Each design came in three different relative density profiles, namely maximum (MAX), functionally graded (FG), and minimum (MIN). After their compression in bone-mimicking molds with simulated acetabular defects, the space-filling behavior of the implants was evaluated using load-displacement curves, micro-CT images, and 3D reconstructions. The meta-implants with an FG diamond infill exhibited the most promising space-filling behavior. However, the required push-in forces exceed the impact forces currently applied in surgery. Future research should, therefore, focus on design optimization, to improve the space-filling behavior and to facilitate the implantation process for orthopedic surgeons.

30 June 2021 in Journal of Orthopaedic Research
Verified
Journal Article
Open access
PDF file
MFMateusz C. Florkow
Mateusz C. Florkow
KWKoen Willemsen
Koen Willemsen
FZFrank Zijlstra
Frank Zijlstra
WFWouter Foppen
Wouter Foppen
BWBart C. H. van der Wal
Bart C. H. van der Wal
PSPeter R. Seevinck
Peter R. Seevinck
This study evaluated the accuracy of synthetic computed tomography (sCT), as compared to CT, for the 3D assessment of the hip morphology. Thirty male patients with asymptomatic hips, referred for magnetic resonance (MR) imaging and CT, were included in this retrospective study. sCT images were generated from three‐dimensional radiofrequency‐spoiled T1‐weighted multi‐echo gradient‐echo MR images using a commercially available deep learning‐enabled software and were compared with CT images through mean error and surface distance computation and by means of eight clinical morphometric parameters relevant for hip care. Parameters included center‐edge angle (CEA), sharp angle, acetabular index, extrusion index, femoral head center‐to‐midline distance, acetabular version (AV), and anterior and posterior acetabular sector angles. They were measured by two senior orthopedic surgeons and a radiologist in‐training on CT and sCT images. The reliability and agreement of CT‐ and sCT‐based measurements were assessed using intraclass correlation coefficients (ICCs) for absolute agreement, Bland–Altman plots, and two one‐sided tests for equivalence. The surface distance between CT‐ and sCT‐based bone models were on average submillimeter. CT‐ and sCT‐based measurements showed moderate to excellent interobserver and intraobserver correlation (0.56 < ICC < 0.99). In particular, the inter/intraobserver agreements were good for AV (ICC > 0.75). For CEA, the intraobserver agreement was good (ICC > 0.75) and the interobserver agreement was moderate (ICC > 0.69). Limits of agreements were similar between intraobserver CT and intermodal measurements. All measurements were found statistically equivalent, with average intermodal differences within the intraobserver limits of agreement. In conclusion, sCT and CT were equivalent for the assessment of the hip joint bone morphology.

24 September 2021 in Advanced Healthcare Materials
Verified
Journal Article
Open access
PDF file
NGNasim Golafshan
Nasim Golafshan
KWKoen Willemsen
Koen Willemsen
FKFiroz Babu Kadumudi
Firoz Babu Kadumudi
EVElke Vorndran
Elke Vorndran
Alireza Dolatshahi-PirouzAlireza Dolatshahi-Pirouz
Alireza Dolatshahi-Pirouz
Miguel CastilhoMiguel Castilho
Miguel Castilho
Osteoarthritis of the hip is a painful and debilitating condition commonly occurring in humans and dogs. One of the main causes that leads to hip osteoarthritis is hip dysplasia. Although the current surgical methods to correct dysplasia work satisfactorily in many circumstances, these are associated with serious complications, tissue resorption, and degeneration. In this study, a one‐step fabrication of a regenerative hip implant with a patient‐specific design and load‐bearing properties is reported. The regenerative hip implant is fabricated based on patient imaging files and by an extrusion assisted 3D printing process using a flexible, bone‐inducing biomaterial. The novel implant can be fixed with metallic screws to host bone and can be loaded up to physiological loads without signs of critical permanent deformation or failure. Moreover, after exposing the hip implant to accelerated in vitro degradation, it is confirmed that it is still able to support physiological loads even after losing ≈40% of its initial mass. In addition, the osteopromotive properties of the novel hip implant is demonstrated as shown by an increased expression of osteonectin and osteocalcin by cultured human mesenchymal stem cells after 21 days. Overall, the proposed hip implant provides an innovative regenerative and mechanically stable solution for hip dysplasia treatment.