Search

Your search keyword '"Valentinitsch A"' showing total 222 results
Start Over Author "Valentinitsch A"
222 results on '"Valentinitsch A"'

1. Geometrically Inspired Kernel Machines for Collaborative Learning Beyond Gradient Descent

Author

Kumar, Mohit, Valentinitsch, Alexander, Fuchs, Magdalena, Brucker, Mathias, Bowles, Juliana, Husakovic, Adnan, Abbas, Ali, and Moser, Bernhard A.

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

This paper develops a novel mathematical framework for collaborative learning by means of geometrically inspired kernel machines which includes statements on the bounds of generalisation and approximation errors, and sample complexity. For classification problems, this approach allows us to learn bounded geometric structures around given data points and hence solve the global model learning problem in an efficient way by exploiting convexity properties of the related optimisation problem in a Reproducing Kernel Hilbert Space (RKHS). In this way, we can reduce classification problems to determining the closest bounded geometric structure from a given data point. Further advantages that come with our solution is that our approach does not require clients to perform multiple epochs of local optimisation using stochastic gradient descent, nor require rounds of communication between client/server for optimising the global model. We highlight that numerous experiments have shown that the proposed method is a competitive alternative to the state-of-the-art.

Published
2024

2. VerSe: A Vertebrae Labelling and Segmentation Benchmark for Multi-detector CT Images

Author

Sekuboyina, Anjany, Husseini, Malek E., Bayat, Amirhossein, Löffler, Maximilian, Liebl, Hans, Li, Hongwei, Tetteh, Giles, Kukačka, Jan, Payer, Christian, Štern, Darko, Urschler, Martin, Chen, Maodong, Cheng, Dalong, Lessmann, Nikolas, Hu, Yujin, Wang, Tianfu, Yang, Dong, Xu, Daguang, Ambellan, Felix, Amiranashvili, Tamaz, Ehlke, Moritz, Lamecker, Hans, Lehnert, Sebastian, Lirio, Marilia, de Olaguer, Nicolás Pérez, Ramm, Heiko, Sahu, Manish, Tack, Alexander, Zachow, Stefan, Jiang, Tao, Ma, Xinjun, Angerman, Christoph, Wang, Xin, Brown, Kevin, Kirszenberg, Alexandre, Puybareau, Élodie, Chen, Di, Bai, Yiwei, Rapazzo, Brandon H., Yeah, Timyoas, Zhang, Amber, Xu, Shangliang, Hou, Feng, He, Zhiqiang, Zeng, Chan, Xiangshang, Zheng, Liming, Xu, Netherton, Tucker J., Mumme, Raymond P., Court, Laurence E., Huang, Zixun, He, Chenhang, Wang, Li-Wen, Ling, Sai Ho, Huynh, Lê Duy, Boutry, Nicolas, Jakubicek, Roman, Chmelik, Jiri, Mulay, Supriti, Sivaprakasam, Mohanasankar, Paetzold, Johannes C., Shit, Suprosanna, Ezhov, Ivan, Wiestler, Benedikt, Glocker, Ben, Valentinitsch, Alexander, Rempfler, Markus, Menze, Björn H., and Kirschke, Jan S.

Subjects
Computer Science - Computer Vision and Pattern Recognition, Electrical Engineering and Systems Science - Image and Video Processing
Abstract

Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision-support systems for diagnosis, surgery planning, and population-based analysis on spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms towards labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel-level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the the results of this evaluation and further investigate the performance-variation at vertebra-level, scan-level, and at different fields-of-view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The content and code concerning VerSe can be accessed at: https://github.com/anjany/verse., Comment: Challenge report for the VerSe 2019 and 2020. Published in Medical Image Analysis (DOI: https://doi.org/10.1016/j.media.2021.102166)

Published
2020
Full Text
View/download PDF

3. Probabilistic Point Cloud Reconstructions for Vertebral Shape Analysis

Author

Sekuboyina, Anjany, Rempfler, Markus, Valentinitsch, Alexander, Loeffler, Maximilian, Kirschke, Jan S., and Menze, Bjoern H.

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Computer Science - Computer Vision and Pattern Recognition
Abstract

We propose an auto-encoding network architecture for point clouds (PC) capable of extracting shape signatures without supervision. Building on this, we (i) design a loss function capable of modelling data variance on PCs which are unstructured, and (ii) regularise the latent space as in a variational auto-encoder, both of which increase the auto-encoders' descriptive capacity while making them probabilistic. Evaluating the reconstruction quality of our architectures, we employ them for detecting vertebral fractures without any supervision. By learning to efficiently reconstruct only healthy vertebrae, fractures are detected as anomalous reconstructions. Evaluating on a dataset containing $\sim$1500 vertebrae, we achieve area-under-ROC curve of $>$75%, without using intensity-based features., Comment: Accepted at Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019; JSK and BHM are joint supervising authors

Published
2019

4. Labelling Vertebrae with 2D Reformations of Multidetector CT Images: An Adversarial Approach for Incorporating Prior Knowledge of Spine Anatomy

Author

Sekuboyina, Anjany, Rempfler, Markus, Valentinitsch, Alexander, Menze, Bjoern H., and Kirschke, Jan S.

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Purpose: To use and test a labelling algorithm that operates on two-dimensional (2D) reformations, rather than three-dimensional (3D) data to locate and identify vertebrae. Methods: We improved the Btrfly Net (described by Sekuboyina et al) that works on sagittal and coronal maximum intensity projections (MIP) and augmented it with two additional components: spine-localization and adversarial a priori-learning. Furthermore, we explored two variants of adversarial training schemes that incorporated the anatomical a priori knowledge into the Btrfly Net. We investigated the superiority of the proposed approach for labelling vertebrae on three datasets: a public benchmarking dataset of 302 CT scans and two in-house datasets with a total of 238 CT scans. We employed Wilcoxon signed-rank test to compute the statistical significance of the improvement in performance observed due to various architectural components in our approach. Results: On the public dataset, our approach using the described Btrfly(pe-eb) network performed on par with current state-of-the-art methods achieving a statistically significant (p < .001) vertebrae identification rate of 88.5+/-0.2 % and localization distances of less than 7-mm. On the in-house datasets that had a higher inter-scan data variability, we obtained an identification rate of 85.1+/-1.2%. Conclusion: An identification performance comparable to existing 3D approaches was achieved when labelling vertebrae on 2D MIPs. The performance was further improved using the proposed adversarial training regime that effectively enforced local spine a priori knowledge during training. Lastly, spine-localization increased the generalizability of our approach by homogenizing the content in the MIPs., Comment: Published in Radiology:AI

Published
2019

5. Btrfly Net: Vertebrae Labelling with Energy-based Adversarial Learning of Local Spine Prior

Author

Sekuboyina, Anjany, Rempfler, Markus, Kukačka, Jan, Tetteh, Giles, Valentinitsch, Alexander, Kirschke, Jan S., and Menze, Bjoern H.

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Robust localisation and identification of vertebrae is essential for automated spine analysis. The contribution of this work to the task is two-fold: (1) Inspired by the human expert, we hypothesise that a sagittal and coronal reformation of the spine contain sufficient information for labelling the vertebrae. Thereby, we propose a butterfly-shaped network architecture (termed Btrfly Net) that efficiently combines the information across reformations. (2) Underpinning the Btrfly net, we present an energy-based adversarial training regime that encodes local spine structure as an anatomical prior into the network, thereby enabling it to achieve state-of-art performance in all standard metrics on a benchmark dataset of 302 scans without any post-processing during inference., Comment: Published as conference paper in Medical Image Computing and Computer Assisted Intervention - MICCAI 2018

Published
2018

6. A Localisation-Segmentation Approach for Multi-label Annotation of Lumbar Vertebrae using Deep Nets

Author

Sekuboyina, Anjany, Valentinitsch, Alexander, Kirschke, Jan S., and Menze, Bjoern H.

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Multi-class segmentation of vertebrae is a non-trivial task mainly due to the high correlation in the appearance of adjacent vertebrae. Hence, such a task calls for the consideration of both global and local context. Based on this motivation, we propose a two-staged approach that, given a computed tomography dataset of the spine, segments the five lumbar vertebrae and simultaneously labels them. The first stage employs a multi-layered perceptron performing non-linear regression for locating the lumbar region using the global context. The second stage, comprised of a fully-convolutional deep network, exploits the local context in the localised lumbar region to segment and label the lumbar vertebrae in one go. Aided with practical data augmentation for training, our approach is highly generalisable, capable of successfully segmenting both healthy and abnormal vertebrae (fractured and scoliotic spines). We consistently achieve an average Dice coefficient of over 90 percent on a publicly available dataset of the xVertSeg segmentation challenge of MICCAI 2016. This is particularly noteworthy because the xVertSeg dataset is beset with severe deformities in the form of vertebral fractures and scoliosis.

Published
2017

7. SurvivalNet: Predicting patient survival from diffusion weighted magnetic resonance images using cascaded fully convolutional and 3D convolutional neural networks

Author

Christ, Patrick Ferdinand, Ettlinger, Florian, Kaissis, Georgios, Schlecht, Sebastian, Ahmaddy, Freba, Grün, Felix, Valentinitsch, Alexander, Ahmadi, Seyed-Ahmad, Braren, Rickmer, and Menze, Bjoern

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Automatic non-invasive assessment of hepatocellular carcinoma (HCC) malignancy has the potential to substantially enhance tumor treatment strategies for HCC patients. In this work we present a novel framework to automatically characterize the malignancy of HCC lesions from DWI images. We predict HCC malignancy in two steps: As a first step we automatically segment HCC tumor lesions using cascaded fully convolutional neural networks (CFCN). A 3D neural network (SurvivalNet) then predicts the HCC lesions' malignancy from the HCC tumor segmentation. We formulate this task as a classification problem with classes being "low risk" and "high risk" represented by longer or shorter survival times than the median survival. We evaluated our method on DWI of 31 HCC patients. Our proposed framework achieves an end-to-end accuracy of 65% with a Dice score for the automatic lesion segmentation of 69% and an accuracy of 68% for tumor malignancy classification based on expert annotations. We compared the SurvivalNet to classical handcrafted features such as Histogram and Haralick and show experimentally that SurvivalNet outperforms the handcrafted features in HCC malignancy classification. End-to-end assessment of tumor malignancy based on our proposed fully automatic framework corresponds to assessment based on expert annotations with high significance (p>0.95)., Comment: Accepted at IEEE ISBI 2017

Published
2017

8. VerSe: A Vertebrae labelling and segmentation benchmark for multi-detector CT images

Author

Sekuboyina, Anjany, Husseini, Malek E., Bayat, Amirhossein, Löffler, Maximilian, Liebl, Hans, Li, Hongwei, Tetteh, Giles, Kukačka, Jan, Payer, Christian, Štern, Darko, Urschler, Martin, Chen, Maodong, Cheng, Dalong, Lessmann, Nikolas, Hu, Yujin, Wang, Tianfu, Yang, Dong, Xu, Daguang, Ambellan, Felix, Amiranashvili, Tamaz, Ehlke, Moritz, Lamecker, Hans, Lehnert, Sebastian, Lirio, Marilia, Olaguer, Nicolás Pérez de, Ramm, Heiko, Sahu, Manish, Tack, Alexander, Zachow, Stefan, Jiang, Tao, Ma, Xinjun, Angerman, Christoph, Wang, Xin, Brown, Kevin, Kirszenberg, Alexandre, Puybareau, Élodie, Chen, Di, Bai, Yiwei, Rapazzo, Brandon H., Yeah, Timyoas, Zhang, Amber, Xu, Shangliang, Hou, Feng, He, Zhiqiang, Zeng, Chan, Xiangshang, Zheng, Liming, Xu, Netherton, Tucker J., Mumme, Raymond P., Court, Laurence E., Huang, Zixun, He, Chenhang, Wang, Li-Wen, Ling, Sai Ho, Huỳnh, Lê Duy, Boutry, Nicolas, Jakubicek, Roman, Chmelik, Jiri, Mulay, Supriti, Sivaprakasam, Mohanasankar, Paetzold, Johannes C., Shit, Suprosanna, Ezhov, Ivan, Wiestler, Benedikt, Glocker, Ben, Valentinitsch, Alexander, Rempfler, Markus, Menze, Björn H., and Kirschke, Jan S.

Published
2021
Full Text
View/download PDF

13. Attention-Driven Deep Learning for Pathological Spine Segmentation

Author

Sekuboyina, Anjany, Kukačka, Jan, Kirschke, Jan S., Menze, Bjoern H., Valentinitsch, Alexander, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Glocker, Ben, editor, Yao, Jianhua, editor, Vrtovec, Tomaž, editor, Frangi, Alejandro, editor, and Zheng, Guoyan, editor

Published
2018
Full Text
View/download PDF

17. Multi-modal Image Classification Using Low-Dimensional Texture Features for Genomic Brain Tumor Recognition

Author

Alberts, Esther, Tetteh, Giles, Trebeschi, Stefano, Bieth, Marie, Valentinitsch, Alexander, Wiestler, Benedikt, Zimmer, Claus, Menze, Bjoern H., Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Cardoso, M. Jorge, editor, Arbel, Tal, editor, Ferrante, Enzo, editor, Pennec, Xavier, editor, Dalca, Adrian V., editor, Parisot, Sarah, editor, Joshi, Sarang, editor, Batmanghelich, Nematollah K., editor, Sotiras, Aristeidis, editor, Nielsen, Mads, editor, Sabuncu, Mert R., editor, Fletcher, Tom, editor, Shen, Li, editor, Durrleman, Stanley, editor, and Sommer, Stefan, editor

Published
2017
Full Text
View/download PDF

19. Regional variations in MR relaxation of hip joint cartilage in subjects with and without femoralacetabular impingement

Author

Subburaj, Karupppasamy, Valentinitsch, Alexander, Dillon, Alexander B, Joseph, Gabby B, Li, Xiaojuan, Link, Thomas M, Vail, Thomas P, and Majumdar, Sharmila

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Clinical Research, Biomedical Imaging, Pain Research, Chronic Pain, Musculoskeletal, Adult, Area Under Curve, Automation, Cartilage, Articular, Female, Femoracetabular Impingement, Hip, Hip Joint, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging, Male, Middle Aged, ROC Curve, Reproducibility of Results, Time Factors, Young Adult, MRI, Cartilage, T-1 rho and T-2, Femoral-acetabular impingement, T(1ρ) and T(2), Biomedical Engineering, Cognitive Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

The objective of this study was to analyze regional variations of magnetic resonance (MR) relaxation times (T1ρ and T2) in hip joint cartilage of healthy volunteers and subjects with femoral acetabular impingement (FAI). Morphological and quantitative images of the hip joints of 12 healthy volunteers and 9 FAI patients were obtained using a 3T MR scanner. Both femoral and acetabular cartilage layers in each joint were semi-automatically segmented on sagittal 3D high-resolution spoiled gradient echo (SPGR) images. These segmented regions of interest (ROIs) were automatically divided radially into twelve equal sub-regions (30(0) intervals) based on the fitted center of the femur head. The mean value of T1ρ/T2 was calculated in each sub-region after superimposing the divided cartilage contours on the MR relaxation (T1ρ/T2) maps to quantify the relaxation times. T1ρ and T2 relaxation times of the femoral cartilage were significantly higher in FAI subjects compared to healthy controls (39.9±3.3 msec in FAI vs. 35.4±2.3msec in controls for T1ρ (P=0.0020); 33.9±3.1 msec in FAI vs. 31.1±1.7 msec in controls for T2 (P=0.0160)). Sub-regional analysis showed significantly different T1ρ and T2 relaxation times in the anterior-superior region (R9) of the hip joint cartilage between subjects with FAI and healthy subjects, suggesting possible regional differences in cartilage matrix composition between these two groups. Receiver operating characteristic (ROC) analysis showed that sub-regional analysis in femoral cartilage was more sensitive in discriminating FAI joint cartilage from that of healthy joints than global analysis of the whole region (T1ρ: area under the curve (AUC)=0.981, P=0.0001 for R9 sub-region; AUC=0.901, P=0.002 for whole region; T2: AUC=0.976, P=0.0005 for R9 sub-region; AUC=0.808, P=0.0124 for whole region). The results of this study demonstrated regional variations in hip cartilage composition using MR relaxation times (T1ρ and T2) and suggested that analysis based on local regions was more sensitive than global measures in subjects with and without FAI.

Published
2013

20. Automated unsupervised multi‐parametric classification of adipose tissue depots in skeletal muscle

Author

Valentinitsch, Alexander, Karampinos, Dimitrios C, Alizai, Hamza, Subburaj, Karupppasamy, Kumar, Deepak, Link, Thomas M, and Majumdar, Sharmila

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Biomedical Imaging, Adipose Tissue, Aged, Algorithms, Body Fat Distribution, Body Mass Index, Cluster Analysis, Diabetes Mellitus, Type 2, Female, Humans, Image Enhancement, Image Interpretation, Computer-Assisted, Magnetic Resonance Imaging, Middle Aged, Muscle, Skeletal, Pattern Recognition, Automated, Predictive Value of Tests, Reference Values, Subcutaneous Fat, magnetic resonance imaging, water-fat imaging, subcutaneous adipose tissue, intermuscular adipose tissue, fat quantification, multi-parametric clustering, Physical Sciences, Engineering, Medical and Health Sciences, Nuclear Medicine & Medical Imaging, Clinical sciences
Abstract

PurposeTo introduce and validate an automated unsupervised multi-parametric method for segmentation of the subcutaneous fat and muscle regions to determine subcutaneous adipose tissue (SAT) and intermuscular adipose tissue (IMAT) areas based on data from a quantitative chemical shift-based water-fat separation approach.Materials and methodsUnsupervised standard k-means clustering was used to define sets of similar features (k = 2) within the whole multi-modal image after the water-fat separation. The automated image processing chain was composed of three primary stages: tissue, muscle, and bone region segmentation. The algorithm was applied on calf and thigh datasets to compute SAT and IMAT areas and was compared with a manual segmentation.ResultsThe IMAT area using the automatic segmentation had excellent agreement with the IMAT area using the manual segmentation for all the cases in the thigh (R(2): 0.96) and for cases with up to moderate IMAT area in the calf (R(2): 0.92). The group with the highest grade of muscle fat infiltration in the calf had the highest error in the inner SAT contour calculation.ConclusionThe proposed multi-parametric segmentation approach combined with quantitative water-fat imaging provides an accurate and reliable method for an automated calculation of the SAT and IMAT areas reducing considerably the total postprocessing time.

Published
2013

21. Multicenter precision of cortical and trabecular bone quality measures assessed by high‐resolution peripheral quantitative computed tomography

Author

Burghardt, Andrew J, Pialat, Jean‐Baptiste, Kazakia, Galateia J, Boutroy, Stephanie, Engelke, Klaus, Patsch, Janina M, Valentinitsch, Alexander, Liu, Danmei, Szabo, Eva, Bogado, Cesar E, Zanchetta, Maria Belen, McKay, Heather A, Shane, Elizabeth, Boyd, Steven K, Bouxsein, Mary L, Chapurlat, Roland, Khosla, Sundeep, and Majumdar, Sharmila

Subjects
Biomedical and Clinical Sciences, Clinical Sciences, Biomedical Imaging, Osteoporosis, Bioengineering, Transplantation, Clinical Trials and Supportive Activities, Clinical Research, Good Health and Well Being, Aged, Aged, 80 and over, Bone and Bones, Feasibility Studies, Female, Humans, Male, Middle Aged, Reproducibility of Results, Tomography, X-Ray Computed, HR-PQCT, OSTEOPOROSIS, PRECISION, BONE, MICROSTRUCTURE, BONE STRENGTH, MULTICENTER STUDIES, Biological Sciences, Engineering, Medical and Health Sciences, Anatomy & Morphology, Biological sciences, Biomedical and clinical sciences
Abstract

High-resolution peripheral quantitative computed tomography (HR-pQCT) has recently been introduced as a clinical research tool for in vivo assessment of bone quality. The utility of this technology to address important skeletal health questions requires translation to standardized multicenter data pools. Our goal was to evaluate the feasibility of pooling data in multicenter HR-pQCT imaging trials. Reproducibility imaging experiments were performed using structure and composition-realistic phantoms constructed from cadaveric radii. Single-center precision was determined by repeat scanning over short-term (

Published
2013

24. Hierarchical Multi-Organ Segmentation Without Registration in 3D Abdominal CT Images

Author

Zografos, Vasileios, Valentinitsch, Alexander, Rempfler, Markus, Tombari, Federico, Menze, Bjoern, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Menze, Bjoern, editor, Langs, Georg, editor, Montillo, Albert, editor, Kelm, Michael, editor, Müller, Henning, editor, Zhang, Shaoting, editor, Cai, Weidong, editor, and Metaxas, Dimitris, editor

Published
2016
Full Text
View/download PDF

36. Recombinant Human BMP6 Applied Within Autologous Blood Coagulum Accelerates Bone Healing: Randomized Controlled Trial in High Tibial Osteotomy Patients

Author

Elena Nemecek, Alexander Valentinitsch, Franz Kainberger, Markus M. Schreiner, Milan Milošević, Hermann Oppermann, Reinhard Windhager, Carmen Trost, Tatjana Bordukalo-Niksic, Kevin Staats, Sanja Pehar, Lovorka Grgurevic, Catharina Chiari, Mihaela Perić, Snjezana Martinovic, Alexander Kolb, T. Kuber Sampath, and Slobodan Vukicevic

Subjects
0301 basic medicine, Adult, Male, medicine.medical_specialty, INJURY/FRACTURE HEALING, Knee Joint, Bone Morphogenetic Protein 6, Endocrinology, Diabetes and Metabolism, 030209 endocrinology & metabolism, BMP, CLINICAL TRIALS, IMPLANTS, Osteoarthritis, Bone healing, Bone morphogenetic protein, 03 medical and health sciences, 0302 clinical medicine, High tibial osteotomy, Medicine, Humans, Orthopedics and Sports Medicine, Clinical Trials, Varus deformity, Bone mineral, Tibia, business.industry, Middle Aged, Osteoarthritis, Knee, medicine.disease, Clinical Trial, Recombinant Proteins, 3. Good health, Surgery, Osteotomy, 030104 developmental biology, Tolerability, Female, Implant, business
Abstract

Bone morphogenetic proteins (BMPs) are potent osteogenic proteins that induce new bone formation in vivo. However, their effect on bone healing in the trabecular bone surfaces remains challenging. We evaluated the safety and efficacy of recombinant human BMP6 (rhBMP6) applied within an autologous blood coagulum (ABC) in a surgically created wedge defect of the proximal tibia in patients undergoing high tibial osteotomy (HTO) for varus deformity and medial osteoarthritis of the knee. We enrolled 20 HTO patients in a randomized, placebo‐controlled, double‐blinded phase I/II clinical trial. RhBMP6/ABC (1.0 mg/10 mL ABC prepared from peripheral blood) or placebo (10 mL ABC containing excipients) was administered into the tibial wedge defects. Patients were followed for 0 to 24 months by clinical examination (safety) and computed tomography (CT) and serial radiographic analyses (efficacy). The results show that there were no detectable anti‐rhBMP6 antibodies in the blood of any of the 20 patients at 14 weeks after implantation. During the 24 months of follow‐up, there were no serious adverse reactions recorded. The CT scans from defects of patients treated with rhBMP6/ABC showed an accelerated bone healing compared with placebo at 9 weeks (47.8 ± 24.1 versus 22.2 ± 12.3 mg/cm3; p = 0.008) and at 14 weeks (89.7 ± 29.1 versus 53.6 ± 21.9 mg/cm3; p = 0.006) follow‐up. Radiographic analyses at weeks 6 and 24 and months 12 and 24 suggested the advanced bone formation and remodeling in rhBMP6/ABC‐treated patients. In conclusion, we show that rhBMP6/ABC at a dose of 100 μg/mL accelerated bone healing in patients undergoing HTO without serious adverse events and with a good tolerability compared with placebo alone. Overall, for the first time, a BMP‐based osteogenic implant was examined against a placebo for bone healing efficacy in the trabecular bone surface, using an objective bone mineral density measurement system. © 2020 The Authors. Journal of Bone and Mineral Research published by American Society for Bone and Mineral Research.

Published
2020

38. X-ray-based quantitative osteoporosis imaging at the spine

Author

Maximilian T. Löffler, Alexander Valentinitsch, Peter B. Noël, J. S. Kirschke, Thomas Baum, Kai Mei, and Nico Sollmann

Subjects
musculoskeletal diseases, 0301 basic medicine, medicine.medical_specialty, Endocrinology, Diabetes and Metabolism, Osteoporosis, Population, 030209 endocrinology & metabolism, Metabolic bone disease, 03 medical and health sciences, Absorptiometry, Photon, 0302 clinical medicine, Trabecular bone score, Bone Density, Internal medicine, medicine, Humans, Quantitative computed tomography, education, Dual-energy X-ray absorptiometry, Aged, education.field_of_study, medicine.diagnostic_test, business.industry, X-Rays, medicine.disease, Spine, Rheumatology, Orthopedic surgery, 030101 anatomy & morphology, Radiology, business
Abstract

Osteoporosis is a metabolic bone disease with a high prevalence that affects the population worldwide, particularly the elderly. It is often due to fractures associated with bone fragility that the diagnosis of osteoporosis becomes clinically evident. However, early diagnosis would be necessary to initiate therapy and to prevent occurrence of further fractures, thus reducing morbidity and mortality. X-ray-based imaging plays a key role for fracture risk assessment and monitoring of osteoporosis. Whereas over decades dual-energy X-ray absorptiometry (DXA) has been the main method used and still reflects the reference standard, another modality reemerges with quantitative computed tomography (QCT) because of its three-dimensional advantages and the opportunistic exploitation of routine CT scans. Against this background, this article intends to review and evaluate recent advances in the field of X-ray-based quantitative imaging of osteoporosis at the spine. First, standard DXA with the recent addition of trabecular bone score (TBS) is presented. Secondly, standard QCT, dual-energy BMD quantification, and opportunistic BMD screening in non-dedicated CT exams are discussed. Lastly, finite element analysis and microstructural parameter analysis are reviewed.

Published
2019

39. VerSe: a vertebrae labelling and segmentation benchmark for multi-detector CT images

Author

Sekuboyina, A, Husseini, ME, Bayat, A, Löffler, M, Liebl, H, Li, H, Tetteh, G, Kukačka, J, Payer, C, Štern, D, Urschler, M, Chen, M, Cheng, D, Lessmann, N, Hu, Y, Wang, T, Yang, D, Xu, D, Ambellan, F, Amiranashvili, T, Ehlke, M, Lamecker, H, Lehnert, S, Lirio, M, Olaguer, NPD, Ramm, H, Sahu, M, Tack, A, Zachow, S, Jiang, T, Ma, X, Angerman, C, Wang, X, Brown, K, Wolf, M, Kirszenberg, A, Puybareau, É, Chen, D, Bai, Y, Rapazzo, BH, Yeah, T, Zhang, A, Xu, S, Hou, F, He, Z, Zeng, C, Xiangshang, Z, Liming, X, Netherton, TJ, Mumme, RP, Court, LE, Huang, Z, He, C, Wang, L-W, Ling, SH, Huynh, LD, Boutry, N, Jakubicek, R, Chmelik, J, Mulay, S, Sivaprakasam, M, Paetzold, JC, Shit, S, Ezhov, I, Wiestler, B, Glocker, B, Valentinitsch, A, Rempfler, M, Menze, BH, and Kirschke, JS

Subjects
eess.IV, cs.CV
Abstract

Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision-support systems for diagnosis, surgery planning, and population-based analysis on spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms towards labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel-level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the the results of this evaluation and further investigate the performance-variation at vertebra-level, scan-level, and at different fields-of-view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The content and code concerning VerSe can be accessed at: https://github.com/anjany/verse.

Published
2021

40. VERSE: A Vertebrae labelling and segmentation benchmark for multi-detector CT images

Author

Mohanasankar Sivaprakasam, Timyoas Yeah, Tao Jiang, Xin Wang, Dalong Cheng, Manish Sahu, Maodong Chen, Sebastian Lehnert, Alexander Valentinitsch, Dong Yang, Nicolas Boutry, Shangliang Xu, Johannes C. Paetzold, Alexander Tack, Yujin Hu, Kevin W. Brown, Marilia Lirio, Malek El Husseini, Xu Liming, Darko Štern, Nikolas Lessmann, Suprosanna Shit, Tianfu Wang, Alexandre Kirszenberg, Martin Urschler, Daguang Xu, Feng Hou, Laurence E. Court, Raymond P. Mumme, Maximilian T. Löffler, Sai Ho Ling, Stefan Zachow, Zheng Xiangshang, Markus Rempfler, Yiwei Bai, Elodie Puybareau, Li-Wen Wang, Nicolás Pérez de Olaguer, Moritz Ehlke, Tamaz Amiranashvili, Di Chen, Christoph Angerman, Chan Zeng, Zixun Huang, Jiri Chmelik, Giles Tetteh, Hongwei Li, Jan S. Kirschke, Heiko Ramm, Amirhossein Bayat, Björn H. Menze, Ivan Ezhov, Jan Kukačka, Anjany Sekuboyina, Chenhang He, Ben Glocker, Tucker Netherton, Hans Liebl, Zhiqiang He, Roman Jakubicek, Christian Payer, Felix Ambellan, Supriti Mulay, Lê Duy Huỳnh, Brandon H. Rapazzo, Xinjun Ma, Amber Zhang, Hans Lamecker, and Benedikt Wiestler

Subjects
FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Population, Computer Science - Computer Vision and Pattern Recognition, Labelling, Segmentation, Spine, Vertebrae, Health Informatics, computer.software_genre, 09 Engineering, 11 Medical and Health Sciences, Voxel, medicine, FOS: Electrical engineering, electronic engineering, information engineering, Image Processing, Computer-Assisted, Humans, Radiology, Nuclear Medicine and imaging, education, education.field_of_study, Radiological and Ultrasound Technology, business.industry, Image and Video Processing (eess.IV), Medical image computing, Pattern recognition, Electrical Engineering and Systems Science - Image and Video Processing, Computer Graphics and Computer-Aided Design, Hybrid algorithm, Pipeline (software), Vertebra, Nuclear Medicine & Medical Imaging, Benchmarking, medicine.anatomical_structure, Benchmark (computing), Computer Vision and Pattern Recognition, Artificial intelligence, business, Tomography, X-Ray Computed, computer, Algorithms, Rare cancers Radboud Institute for Health Sciences [Radboudumc 9]
Abstract

Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision-support systems for diagnosis, surgery planning, and population-based analysis on spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms towards labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel-level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the the results of this evaluation and further investigate the performance-variation at vertebra-level, scan-level, and at different fields-of-view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The content and code concerning VerSe can be accessed at: https://github.com/anjany/verse., Comment: Challenge report for the VerSe 2019 and 2020. Published in Medical Image Analysis (DOI: https://doi.org/10.1016/j.media.2021.102166)

Published
2021

41. VerSe: A Vertebrae labelling and segmentation benchmark for multi-detector CT images

Author

Sekuboyina, A., Husseini, M.E., Bayat, A., Loffler, M., Liebl, H., Li, H., Tetteh, G., Kukacka, J., Payer, C., Stern, D., Urschler, M., Chen, M., Cheng, D., Lessmann, N., Hu, Y., Wang, T., Yang, D., Xu, D., Ambellan, F., Amiranashvili, T., Ehlke, M., Lamecker, H., Lehnert, S., Lirio, M., Olaguer, N.P., Ramm, H., Sahu, M., Tack, A., Zachow, S., Jiang, T., Ma, X., Angerman, C., Wang, X, Brown, K., Kirszenberg, A., Puybareau, E., Chen, D., Bai, Y., Rapazzo, B.H., Yeah, T., Zhang, A., Xu, S., Hou, F., He, Z., Zeng, C., Xiangshang, Z., Liming, X., Netherton, T.J., Mumme, R.P., Court, L.E., Huang, Z., He, C., Wang, L.W., Ling, S.H., Huynh, L.D., Boutry, N., Jakubicek, R., Chmelik, J., Mulay, S., Sivaprakasam, M., Paetzold, J.C., Shit, S., Ezhov, I., Wiestler, B., Glocker, B., Valentinitsch, A., Rempfler, M., Menze, B.H., Kirschke, J.S., Sekuboyina, A., Husseini, M.E., Bayat, A., Loffler, M., Liebl, H., Li, H., Tetteh, G., Kukacka, J., Payer, C., Stern, D., Urschler, M., Chen, M., Cheng, D., Lessmann, N., Hu, Y., Wang, T., Yang, D., Xu, D., Ambellan, F., Amiranashvili, T., Ehlke, M., Lamecker, H., Lehnert, S., Lirio, M., Olaguer, N.P., Ramm, H., Sahu, M., Tack, A., Zachow, S., Jiang, T., Ma, X., Angerman, C., Wang, X, Brown, K., Kirszenberg, A., Puybareau, E., Chen, D., Bai, Y., Rapazzo, B.H., Yeah, T., Zhang, A., Xu, S., Hou, F., He, Z., Zeng, C., Xiangshang, Z., Liming, X., Netherton, T.J., Mumme, R.P., Court, L.E., Huang, Z., He, C., Wang, L.W., Ling, S.H., Huynh, L.D., Boutry, N., Jakubicek, R., Chmelik, J., Mulay, S., Sivaprakasam, M., Paetzold, J.C., Shit, S., Ezhov, I., Wiestler, B., Glocker, B., Valentinitsch, A., Rempfler, M., Menze, B.H., and Kirschke, J.S.

Abstract

Contains fulltext : 238698.pdf (Publisher’s version ) (Open Access), Vertebral labelling and segmentation are two fundamental tasks in an automated spine processing pipeline. Reliable and accurate processing of spine images is expected to benefit clinical decision support systems for diagnosis, surgery planning, and population-based analysis of spine and bone health. However, designing automated algorithms for spine processing is challenging predominantly due to considerable variations in anatomy and acquisition protocols and due to a severe shortage of publicly available data. Addressing these limitations, the Large Scale Vertebrae Segmentation Challenge (VerSe) was organised in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2019 and 2020, with a call for algorithms tackling the labelling and segmentation of vertebrae. Two datasets containing a total of 374 multi-detector CT scans from 355 patients were prepared and 4505 vertebrae have individually been annotated at voxel level by a human-machine hybrid algorithm (https://osf.io/nqjyw/, https://osf.io/t98fz/). A total of 25 algorithms were benchmarked on these datasets. In this work, we present the results of this evaluation and further investigate the performance variation at the vertebra level, scan level, and different fields of view. We also evaluate the generalisability of the approaches to an implicit domain shift in data by evaluating the top-performing algorithms of one challenge iteration on data from the other iteration. The principal takeaway from VerSe: the performance of an algorithm in labelling and segmenting a spine scan hinges on its ability to correctly identify vertebrae in cases of rare anatomical variations. The VerSe content and code can be accessed at: https://github.com/anjany/verse.

Published
2021

43. Author response for 'Recombinant human <scp>BMP6</scp> applied within Autologous Blood Coagulum accelerates bone healing: Randomized controlled trial in High Tibial Osteotomy patients'

Author

Milan Milošević, Franz Kainberger, M. Schreiner, Alexander Valentinitsch, Reinhard Windhager, Mihaela Perić, Elena Nemecek, Hermann Oppermann, Kevin Staats, Sanja Pehar, Tatjana Bordukalo-Niksic, Catharina Chiari, Carmen Trost, Slobodan Vukičević, Lovorka Grgurević, Snjezana Martinovic, Alexander Kolb, and T. Kuber Sampath

Subjects
medicine.medical_specialty, Bone morphogenetic protein 6, High tibial osteotomy, Randomized controlled trial, business.industry, law, Autologous blood, Recombinant DNA, Medicine, Bone healing, business, Surgery, law.invention
Published
2020

46. Analysis and quantification of bone healing after open wedge high tibial osteotomy

Author

Reinhard Windhager, Franz Kainberger, Catharina Chiari, Elena Nemecek, Alexander Kolb, Slobodan Vukicevic, Gerhard M. Hobusch, Alexander Valentinitsch, Carmen Trost, and Lena Hirtler

Subjects
Tibia / surgery, Knee Joint, Radiography, medicine.medical_treatment, CT scan analysis, Bone healing, Osteotomy, Fractures, Bone, 03 medical and health sciences, 0302 clinical medicine, High tibial osteotomy, Open wedge, medicine, Humans, Bone ossification, Fracture Healing, Bone mineral, 030222 orthopedics, Tibia, Analysis of osteotomy gap, business.industry, Ossification, Radiographic analysis, Correction, Main Topic, 030229 sport sciences, General Medicine, Osteotomy / methods, medicine.symptom, Nuclear medicine, business
Abstract

Summary Background The aim of this study was to analyze radiographic imaging techniques and to quantify bone ossification in the osteotomy gap after high tibial osteotomy. Material and methods Study phase 1: high tibial osteotomy was performed on six lower extremities of human body donors and experimental X‑rays and computed tomography (CT) scans were applied. Different techniques were evaluated by three specialists for best representation of the osteotomy gap. Study phase 2: optimized radiological techniques were used for follow-up on 12 patients. The radiographs were examined by 3 specialists measuring 10 different parameters. The CT scans were analyzed with semiautomatic computer software for quantification of bone ossification. Results The osteotomy gap was best represented in 30° of flexion in the knee and 20° internal rotation of the leg. There were significant changes of the medial width over time (p 3) than at baseline (54.6 mg/cm3). The ossification of the gap was visualized by color coding. Conclusion Sclerosis and medial width of the osteotomy gap as well as area measurements were determined as reproducible parameters for evaluation of bone healing. Quantification of bone ossification can be calculated with CT scans using a semiautomatic computer program and should be used for research in bone healing.

Published
2019

47. Correction to: Analysis and quantification of bone healing after open wedge high tibial osteotomy

Author

Gerhard M. Hobusch, Elena Nemecek, Reinhard Windhager, Franz Kainberger, Alexander Kolb, Alexander Valentinitsch, Catharina Chiari, Slobodan Vukicevic, Lena Hirtler, and Carmen Trost

Subjects
Orthodontics, business.industry, MEDLINE, General Medicine, Bone healing, 030204 cardiovascular system & hematology, 03 medical and health sciences, 0302 clinical medicine, High tibial osteotomy, Open wedge, Medicine, 030212 general & internal medicine, Presentation (obstetrics), business
Abstract

Correction to: Wien Klin Wochenschr 2019 https://doi.org/10.1007/s00508-019-01541-8 The original version of this article unfortunately contained a mistake. The presentation of Tab. 4 was incorrect. The corrected table is given below. The original article has been …

Published
2019

48. Recombinant Human BMP6 Applied Within Autologous Blood Coagulum Accelerates Bone Healing: Randomized Controlled Trial in High Tibial Osteotomy Patients

Author

Chiari, Catharina, primary, Grgurevic, Lovorka, additional, Bordukalo‐Niksic, Tatjana, additional, Oppermann, Hermann, additional, Valentinitsch, Alexander, additional, Nemecek, Elena, additional, Staats, Kevin, additional, Schreiner, Markus, additional, Trost, Carmen, additional, Kolb, Alexander, additional, Kainberger, Franz, additional, Pehar, Sanja, additional, Milosevic, Milan, additional, Martinovic, Snjezana, additional, Peric, Mihaela, additional, Sampath, T Kuber, additional, Vukicevic, Slobodan, additional, and Windhager, Reinhard, additional

Published
2020
Full Text
View/download PDF

50. Diffusion tensor image features predict IDH genotype in newly diagnosed WHO grade II/III gliomas

Author

Eichinger, Paul, Alberts, Esther, Delbridge, Claire, Trebeschi, Stefano, Valentinitsch, Alexander, Bette, Stefanie, Huber, Thomas, Gempt, Jens, Meyer, Bernhard, Schlegel, Juergen, Zimmer, Claus, Kirschke, Jan S., Menze, Bjoern H., and Wiestler, Benedikt

Subjects
Adult, Male, Genotype, lcsh:R, lcsh:Medicine, Genomics, Glioma, Middle Aged, Article, Isocitrate Dehydrogenase, Workflow, Machine Learning, Diffusion Tensor Imaging, ROC Curve, Mutation, Image Processing, Computer-Assisted, Humans, Female, lcsh:Q, Neoplasm Grading, lcsh:Science, Neoplasm Staging
Abstract

We hypothesized that machine learning analysis based on texture information from the preoperative MRI can predict IDH mutational status in newly diagnosed WHO grade II and III gliomas. This retrospective study included in total 79 consecutive patients with a newly diagnosed WHO grade II or III glioma. Local binary pattern texture features were generated from preoperative B0 and fractional anisotropy (FA) diffusion tensor imaging. Using a training set of 59 patients, a single hidden layer neural network was then trained on the texture features to predict IDH status. The model was validated based on the prediction accuracy calculated in a previously unseen set of 20 gliomas. Prediction accuracy of the generated model was 92% (54/59 cases; AUC = 0.921) in the training and 95% (19/20; AUC = 0.952) in the validation cohort. The ten most important features were comprised of tumor size and both B0 and FA texture information, underlining the joint contribution of imaging data to classification. Machine learning analysis of DTI texture information and tumor size reliably predicts IDH status in preoperative MRI of gliomas. Such information may increasingly support individualized surgical strategies, supplement pathological analysis and highlight the potential of radiogenomics.

Published
2017

Catalog

Books, media, physical & digital resources
See catalog results