Michel Beyer Michel Beyer

Mr. Michel Beyer Michel Beyer

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
30 September 2024 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
BMBilal Msallem
Bilal Msallem
JVJoel Josef Vavrina
Joel Josef Vavrina
MBMichel Beyer Michel Beyer
Michel Beyer Michel Beyer
FHFlorian S. Halbeisen
Florian S. Halbeisen
GLGuenter Lauer
Guenter Lauer
Florian Markus ThieringerFlorian Markus Thieringer
Florian Markus Thieringer
Background: With the rise of new 3D printers, assessing accuracy is crucial for obtaining the best results in patient care. Previous studies have shown that the highest accuracy is achieved with SLS printing technology; however, SLA printing technology has made significant improvements in recent years. Methods: In this study, a realistic anatomical model of a mandible and skull, a cutting guide for mandibular osteotomy, and a splint for orthognathic surgery were replicated five times each using two different 3D printing technologies: SLA and SLS. Results: The SLA group had a median trueness RMS value of 0.148 mm and a precision RMS value of 0.117 mm. The SLS group had a median trueness RMS value of 0.144 mm and a precision RMS value of 0.096 mm. There was no statistically significant difference in RMS values between SLS and SLA technologies regarding trueness. Regarding precision, however, the RMS values for SLS technology were significantly lower in the splint and cutting guide applications than those printed with SLA technology. Conclusions: Both 3D printing technologies produce modern models and applications with equally high dimensional accuracy. Considering current cost pressures experienced by hospitals, the lower-cost SLA 3D printer is a reliable choice for point-of-care 3D printing.

26 April 2023 in Bioengineering
Verified
Journal Article
Open access
PDF file
Paul FaragoPaul Farago
Paul Farago
SSSebastian-Aurelian Stefaniga
Sebastian-Aurelian Stefaniga
CBClaudia-Georgiana Barbura
Claudia-Georgiana Barbura
LMLaura-Ioana Mihăilă
Laura-Ioana Mihăilă
SHSorin Hintea
Sorin Hintea
RIRobert Radu Ilesan
Robert Radu Ilesan
Parkinson’s disease is a progressive neurodegenerative disorder caused by dopaminergic neuron degeneration. Parkinsonian speech impairment is one of the earliest presentations of the disease and, along with tremor, is suitable for pre-diagnosis. It is defined by hypokinetic dysarthria and accounts for respiratory, phonatory, articulatory, and prosodic manifestations. The topic of this article targets artificial-intelligence-based identification of Parkinson’s disease from continuous speech recorded in a noisy environment. The novelty of this work is twofold. First, the proposed assessment workflow performed speech analysis on samples of continuous speech. Second, we analyzed and quantified Wiener filter applicability for speech denoising in the context of Parkinsonian speech identification. We argue that the Parkinsonian features of loudness, intonation, phonation, prosody, and articulation are contained in the speech, speech energy, and Mel spectrograms. Thus, the proposed workflow follows a feature-based speech assessment to determine the feature variation ranges, followed by speech classification using convolutional neural networks. We report the best classification accuracies of 96% on speech energy, 93% on speech, and 92% on Mel spectrograms. We conclude that the Wiener filter improves both feature-based analysis and convolutional-neural-network-based classification performances.

17 May 2023 in Bioengineering
Verified
Journal Article
Open access
PDF file
RIRobert Radu Ilesan
Robert Radu Ilesan
MBMichel Beyer Michel Beyer
Michel Beyer Michel Beyer
CKChristoph Kunz
Christoph Kunz
Florian Markus ThieringerFlorian Markus Thieringer
Florian Markus Thieringer
Medical image segmentation, whether semi-automatically or manually, is labor-intensive, subjective, and needs specialized personnel. The fully automated segmentation process recently gained importance due to its better design and understanding of CNNs. Considering this, we decided to develop our in-house segmentation software and compare it to the systems of established companies, an inexperienced user, and an expert as ground truth. The companies included in the study have a cloud-based option that performs accurately in clinical routine (dice similarity coefficient of 0.912 to 0.949) with an average segmentation time ranging from 3′54″ to 85′54″. Our in-house model achieved an accuracy of 94.24% compared to the best-performing software and had the shortest mean segmentation time of 2′03″. During the study, developing in-house segmentation software gave us a glimpse into the strenuous work that companies face when offering clinically relevant solutions. All the problems encountered were discussed with the companies and solved, so both parties benefited from this experience. In doing so, we demonstrated that fully automated segmentation needs further research and collaboration between academics and the private sector to achieve full acceptance in clinical routines.

7 November 2024 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
LSLukas Benedikt Seifert
Lukas Benedikt Seifert
MBMichel Beyer Michel Beyer
Michel Beyer Michel Beyer
VCVanessa Czok
Vanessa Czok
AAAlexander Aigner
Alexander Aigner
SASead Abazi
Sead Abazi
RSRobert Sader
Robert Sader
Background: Three-dimensional facial scan technologies, such as stereophotogrammetry, are frequently used in oral and maxillofacial surgery, dentistry, and plastic surgery to assess patient outcomes and plan surgical procedures. Most facial scanners rely on stationary clinical systems, which provide high accuracy and reliability in generating 3D representations of the human face; however, they are cost-intensive and immobile. Recently, smartphone-based facial scan applications that use stereophotogrammetry have been developed. These applications are cost-effective and more accessible than existing stationary systems. Methods: In this study, we compared the accuracy of three different smartphone applications—EM3D, Polycam, and ScandyPro—on an iPhone 14 Pro, which utilizes Apple’s LiDAR (Light Detection and Ranging) technology, to a stationary system by 3DMD, which is considered a gold standard in many studies. Results: The applications included in the study have demonstrated the capability to perform accurately in clinical settings, with mean surface distances of 1.46 mm for EM3D, 1.66 mm for Polycam, and 1.61 mm for ScandyPro. Conclusions: These results suggest that smartphone-based facial scan systems could be a viable alternative, particularly in resource-limited settings. However, further research and collaboration between academia and industry are necessary to fully integrate these technologies into clinical practice.

16 October 2025 in Dentistry Journal
Verified
Journal Article
Open access
PDF file
MBMichel Beyer Michel Beyer
Michel Beyer Michel Beyer
LSLena Scheller
Lena Scheller
BABurde Alexandru
Burde Alexandru
SASead Abazi
Sead Abazi
ASAdelita Sommacal
Adelita Sommacal
Florian Markus ThieringerFlorian Markus Thieringer
Florian Markus Thieringer
Background: Three-dimensional (3D) printing technologies such as Stereolithography (SLA) and Digital Light Processing (DLP) are widely used in dental implantology for the fabrication of surgical guides. While both methods offer clinical viability, their comparative accuracy, efficiency, and material consumption remain subjects of debate. Objectives: To compare the dimensional accuracy, printing time, and material consumption of dental surgical guides fabricated using an SLA printer (Formlabs Form 3B) and a DLP printer (NextDent 5100) at various printing orientations. Methods: A standardized surgical guide was designed and printed on both printers across seven orientations (0–90°). Five guides per angle were fabricated per technology (n = 35 per printer), scanned, and compared with the CAD reference to evaluate dimensional accuracy. Printing time and resin consumption were recorded. Statistical analyses included the Shapiro–Wilk test and Mann–Whitney U test (α = 0.05). Results: Within the evaluated printers and resins, SLA-printed guides demonstrated slightly lower Root Mean Square (RMS) values in most regions, especially in occlusal and drill hole surfaces, while DLP guides tended to undersize Optimal accuracy was observed at 45° for SLA and 60° for DLP. Material consumption was lower for the SLA printer compared with the DLP printer, but SLA required longer printing time (90–200 min vs. 25–75 min for DLP). Conclusions: Both technologies produced clinically acceptable guides under the tested conditions. The tested SLA printer tended to offer slightly higher accuracy and material efficiency, whereas the DLP printer achieved shorter printing times, supporting its use in high-throughput workflows. Printing orientation significantly influenced accuracy and resource use.

13 December 2024 in International Journal of Computer Assisted Radiology and Surgery
Verified
Journal Article
Open access
PDF file
MMMichaela Maintz
Michaela Maintz
NDNora Desan
Nora Desan
Neha SharmaNeha Sharma
Neha Sharma
JBJörg Beinemann
Jörg Beinemann
MBMichel Beyer Michel Beyer
Michel Beyer Michel Beyer
Florian Markus ThieringerFlorian Markus Thieringer
Florian Markus Thieringer
Purpose: The use of computer-assisted virtual surgical planning (VSP) for craniosynostosis surgery is gaining increasing implementation in the clinics. However, accurately transferring the preoperative planning data to the operating room remains challenging. We introduced and investigated a fully digital workflow to perform fronto-orbital advancement (FOA) surgery using 3D-printed patient-specific implants (PSIs) and cold-ablation robot-guided laser osteotomy. This novel approach eliminates the need for traditional surgical templates while enhancing precision and customization, offering a more streamlined and efficient surgical process. Methods: Computed tomography data of a patient with craniosynostosis were used to digitally reconstruct the skull and to perform VSP of the FOA. In total, six PSIs per skull were 3D-printed with a medical-grade bioresorbable composite using the Arburg Plastic Freeforming technology. The planned osteotomy paths and the screw holes, including their positions and axis angles, were digitally transferred to the cold-ablation robot-guided osteotome interface. The osteotomies were performed on 3D-printed patient skull models. The implants, osteotomy and final FOA results were scanned and compared to the VSP data. Results: The osteotomy deviations for the skulls indicated an overall maximum distance of 1.7 mm, a median deviation of 0.44 mm, and a maximum root mean square (RMS) error of 0.67 mm. The deviation of the point-to-point surface comparison of the FOA with the VSP data resulted in a median accuracy of 1.27 mm. Accessing the orbital cavity with the laser remained challenging. Conclusion: This in vitro study showcases a novel FOA technique by effectively combining robot-guided laser osteotomy with 3D-printed patient-specific implants, eliminating the need for surgical templates and achieving high accuracy in bone cutting and positioning. The workflow holds promise for reducing preoperative planning time and increasing surgical efficiency. Further studies on bone tissue are required to validate the safety and effectiveness of this approach, especially in addressing the challenges of pediatric craniofacial surgery.