Refine
Year of publication
Document Type
- In Proceedings (83) (remove)
Is part of the Bibliography
- no (83)
Keywords
- statistical shape and intensity models (2)
- 3d-reconstruction from 2d X-rays (1)
- 3d-reconstruction from 2d Xrays (1)
- Articulated Models, Statistical Shape And Intensity Models, 2D/3D Anatomy Reconstruction, Pelvic Parameters Measurement, Total Hip Arthroplasty (1)
- Kniegelenk (1)
- Medizinische Planung (1)
- Osteoarthrose (1)
- Sparse Geometry Reconstruction (1)
- Statistical Shape Models (1)
- Total Knee Arthoplasty (1)
Institute
- Visual and Data-centric Computing (80)
- Visual Data Analysis (77)
- Therapy Planning (68)
- Numerical Mathematics (5)
- Computational Medicine (4)
- Visual Data Analysis in Science and Engineering (4)
- Geometric Data Analysis and Processing (3)
- Computational Systems Biology (1)
- Mathematics for Life and Materials Science (1)
Manual processing of tomographic data volumes, such as interactive image segmentation in medicine or paleontology, is considered a time-consuming and cumbersome endeavor. Immersive volume sculpting stands as a potential solution to improve its efficiency and intuitiveness. However, current open-source software solutions do not yield the required performance and functionalities. We address this issue by contributing a novel open-source game engine voxel library that supports real-time immersive volume sculpting. Our design leverages GPU instancing, parallel computing, and a chunk-based data structure to optimize collision detection and rendering. We have implemented features that enable fast voxel interaction and improve precision. Our benchmark evaluation indicates that our implementation offers a significant improvement over the state-of-the-art and can render and modify millions of visible voxels while maintaining stable performance for real-time interaction in virtual reality.
Fossil preparation is the activity of processing paleontological specimens for research and exhibition purposes. In addition to traditional mechanical extraction of fossils, preparation presently comprises non-destructive digital methods that are part of a relatively new field, namely virtual paleontology. Despite significant technological advances, both traditional and digital preparation remain cumbersome and time-consuming endeavors. However, this field has received scarce attention from a human-computer interaction perspective. The present study aims to elucidate the state-of-the-art for paleontological fossil preparation in order to determine its main challenges and start a conversation regarding opportunities for creating novel designs that tackle the field's current issues. We conducted a qualitative study involving both technical preparators and virtual paleontologists. The study was divided into two parts: First, we assembled technical preparators and paleontology researchers in a focus group session to discuss their workflows, obtain a preliminary understanding of their issues, and ideate solutions based on their counterparts' workflows. Next, we conducted a series of contextual inquiries involving direct observation and semi-structured in-depth interviews. We transcribed our recordings and examined the data through theoretical and inductive thematic analysis, clustering emerging themes and applying concepts from human-computer interaction and related fields. Our findings report on challenges faced by traditional and digital fossil preparators and potential opportunities to improve their tools and workflows. We contribute with a novel analysis of fossil preparation from an HCI perspective.
Statistical shape modeling aims at capturing shape variations of an anatomical structure that occur within a given population. Shape models are employed in many tasks, such as shape reconstruction and image segmentation, but also shape generation and classification. Existing shape priors either require dense correspondence between training examples or lack robustness and topological guarantees. We present FlowSSM, a novel shape modeling approach that learns shape variability without requiring dense correspondence between training instances. It relies on a hierarchy of continuous deformation flows, which are parametrized by a neural network. Our model outperforms state-of-the-art methods in providing an expressive and robust shape prior for distal femur and liver. We show that the emerging latent representation is discriminative by separating healthy from pathological shapes. Ultimately, we demonstrate its effectiveness on two shape reconstruction tasks from partial data. Our source code is publicly available (https://github.com/davecasp/flowssm).
Virtual paleontology studies digital fossils through data analysis and visualization systems. The discipline is growing in relevance for the evident advantages of non-destructive imaging techniques over traditional paleontological methods, and it has made significant advancements during the last few decades. However, virtual paleontology still faces a number of technological challenges, amongst which are interaction shortcomings of image segmentation applications. Whereas automated segmentation methods are seldom applicable to fossil datasets, manual exploration of these specimens is extremely time-consuming as it impractically delves into three-dimensional data through two-dimensional visualization and interaction means. This paper presents an application that employs virtual reality and haptics to virtual paleontology in order to evolve its interaction paradigms and address some of its limitations. We provide a brief overview of the challenges faced by virtual paleontology practitioners, a description of our immersive virtual paleontology prototype, and the results of a heuristic evaluation of our design.
Reconstructing anatomical shapes from sparse or partial measurements relies on prior knowledge of shape variations that occur within a given population. Such shape priors are learned from example shapes, obtained by segmenting volumetric medical images. For existing models, the resolution of a learned shape prior is limited to the resolution of the training data. However, in clinical practice, volumetric images are often acquired with highly anisotropic voxel sizes, e.g. to reduce image acquisition time in MRI or radiation exposure in CT imaging. The missing shape information between the slices prohibits existing methods to learn a high-resolution shape prior. We introduce a method for high-resolution shape reconstruction from sparse measurements without relying on high-resolution ground truth for training. Our method is based on neural implicit shape representations and learns a continuous shape prior only from highly anisotropic segmentations. Furthermore, it is able to learn from shapes with a varying field of view and can reconstruct from various sparse input configurations. We demonstrate its effectiveness on two anatomical structures: vertebra and femur, and successfully reconstruct high-resolution shapes from sparse segmentations, using as few as three orthogonal slices.
We present a method based on a generative model for detection of disturbances such as prosthesis, screws, zippers, and metals in 2D radiographs. The generative model is trained in an unsupervised fashion using clinical radiographs as well as simulated data, none of which contain disturbances. Our approach employs a latent space consistency loss which has the benefit of identifying similarities, and is enforced to reconstruct X-rays without disturbances. In order to detect images with disturbances, an anomaly score is computed also employing the Frechet distance between the input X-ray and the reconstructed one using our generative model. Validation was performed using clinical pelvis radiographs. We achieved an AUC of 0.77 and 0.83 with clinical and synthetic data, respectively. The results demonstrated a good accuracy of our method for detecting outliers as well as the advantage of utilizing synthetic data.
Three-dimensional medical imaging enables detailed understanding of osteoarthritis structural status. However, there remains a vast need for automatic, thus, reader-independent measures that provide reliable assessment of subject-specific clinical outcomes. To this end, we derive a consistent generalization of the recently proposed B-score to Riemannian shape spaces. We further present an algorithmic treatment yielding simple, yet efficient computations allowing for analysis of large shape populations with several thousand samples. Our intrinsic formulation exhibits improved discrimination ability over its Euclidean counterpart, which we demonstrate for predictive validity on assessing risks of total knee replacement. This result highlights the potential of the geodesic B-score to enable improved personalized assessment and stratification for interventions.
Surgical tool segmentation in endoscopic videos is an important component of computer assisted interventions systems. Recent success of image-based solutions using fully-supervised deep learning approaches can be attributed to the collection of big labeled datasets. However, the annotation of a big dataset of real videos can be prohibitively expensive and time consuming. Computer simulations could alleviate the manual labeling problem, however, models trained on simulated data do not generalize to real data. This work proposes a consistency-based framework for joint learning of simulated and real (unlabeled) endoscopic data to bridge this performance generalization issue. Empirical results on two data sets (15 videos of the Cholec80 and EndoVis'15 dataset) highlight the effectiveness of the proposed Endo-Sim2Real method for instrument segmentation. We compare the segmentation of the proposed approach with state-of-the-art solutions and show that our method improves segmentation both in terms of quality and quantity.
Volumetry of cartilage of the knee is needed for knee osteoarthritis (KOA) assessment. It is typically performed manually in a tedious and subjective process. We developed a method for an automated, segmentation-based quantification of cartilage volume by employing 3D Convolutional Neural Networks (CNNs). CNNs were trained in a supervised manner using magnetic resonance imaging data and cartilage volumetry readings performed by clinical experts for 1378 subjects provided by the Osteoarthritis Initiative. It was shown that 3D CNNs are able to achieve volume measures comparable to the magnitude of variation between expert readings and
the real in vivo situation. In the future, accurate automated cartilage volumetry might support both, diagnosis of KOA as well as longitudinal analysis of KOA progression.
In dentistry, software-based medical image analysis and visualization provide efficient and accurate diagnostic and therapy planning capabilities. We present an approach for the automatic recognition of tooth types and positions in digital volume tomography (DVT). By using deep learning techniques in combination with dimensionality reduction through non-planar reformatting of the jaw anatomy, DVT data can be efficiently processed and
teeth reliably recognized and classified, even in the presence of imaging artefacts, missing or dislocated teeth.
We evaluated our approach, which is based on 2D Convolutional Neural Networks (CNNs), on 118 manually
annotated cases of clinical DVT datasets. Our proposed method correctly classifies teeth with an accuracy of
94% within a limit of 2mm distance to ground truth labels.
Statistical Shape Models (SSMs) are a proven means for model-based 3D anatomy reconstruction from medical
image data. In orthopaedics and biomechanics, SSMs are increasingly employed to individualize measurement
data or to create individualized anatomical models to which implants can be adapted to or functional tests can be performed on. For modeling and analysis of articulated structures, so called articulated SSMs (aSSMs) have been developed. However, a missing feature of aSSMs is the consideration of collisions in the course of individual fitting and articulation. The aim of our work was to develop aSSMs that handle collisions between components correctly. That way it becomes possible to adjust shape and articulation in view of a physically and geometrically plausible individualization. To be able to apply collision-aware aSSMs in simulation and optimisation, our approach is based on an ecient collision detection method employing Graphics Processing Units (GPUs).
We describe a novel nonlinear statistical shape model basedon differential coordinates viewed as elements of GL+(3). We adopt an as-invariant-as possible framework comprising a bi-invariant Lie group mean and a tangent principal component analysis based on a unique GL+(3)-left-invariant, O(3)-right-invariant metric. Contrary to earlier work that equips the coordinates with a specifically constructed group structure, our method employs the inherent geometric structure of the group-valued data and therefore features an improved statistical power in identifying shape differences. We demonstrate this in experiments on two anatomical datasets including comparison to the standard Euclidean as well as recent state-of-the-art nonlinear approaches to statistical shape modeling.
We present a novel approach for nonlinear statistical shape modeling that is invariant under Euclidean motion and thus alignment-free. By analyzing metric distortion and curvature of shapes as elements of Lie groups in a consistent Riemannian setting, we construct a framework that reliably handles large deformations. Due to the explicit character of Lie group operations, our non-Euclidean method is very efficient allowing for fast and numerically robust processing. This facilitates Riemannian analysis of large shape populations accessible through longitudinal and multi-site imaging studies providing increased statistical power. We evaluate the performance of our model w.r.t. shape-based classification of pathological malformations of the human knee and show that it outperforms the standard Euclidean as well as a recent nonlinear approach especially in presence of sparse training data. To provide insight into the model’s ability of capturing natural biological shape variability, we carry out an analysis of specificity and generalization ability.
We present a method for the automated segmentation of knee bones and cartilage from magnetic resonance imaging, that combines a priori knowledge of anatomical shape with Convolutional Neural Networks (CNNs). The proposed approach incorporates 3D Statistical Shape Models (SSMs) as well as 2D and 3D CNNs to achieve a robust and accurate segmentation of even highly pathological knee structures. The method is evaluated on data of the MICCAI grand challenge "Segmentation of Knee Images 2010". For the first time an accuracy equivalent to the inter-observer variability of human readers has been achieved in this challenge. Moreover, the quality of the proposed method is thoroughly assessed using various measures for 507 manual segmentations of bone and cartilage, and 88 additional manual segmentations of cartilage. Our method yields sub-voxel accuracy. In conclusion, combining of anatomical knowledge using SSMs with localized classification via CNNs results in a state-of-the-art segmentation method.
Due to very short T2 relaxation times, imaging of tendons is typically performed using ultra-short echo-time (UTE) acquisition techniques. In this work, we combined an echo-train shifted multi-echo 3D UTE imaging sequence with a 3D curved surface reconstruction to virtually extract the patellar tendon from an acquired 3D UTE dataset. Based on the analysis of the acquired multi-echo data, a T2* relaxation time parameter map was calculated and interpolated to the curved surface of the patellar tendon.
Degenerative Gelenkerkrankungen, wie die Osteoarthrose, sind ein häufiges Krankheitsbild unter älteren Erwachsenen. Hierbei verringert sich u.a. der Gelenkspalt aufgrund degenerierten Knorpels oder geschädigter Menisci. Ein in den Gelenkspalt eingebrachter interpositionaler Spacer soll die mit der Osteoarthrose einhergehende verringerte Gelenkkontaktfläche erhöhen und so der teilweise oder vollständige Gelenkersatz hinausgezögert oder vermieden werden.
In dieser Arbeit präsentieren wir eine Planungssoftware für die Auswahl und Positionierung eines interpositionalen Spacers am Patientenmodell.
Auf einer MRT-basierten Bildsegmentierung aufbauend erfolgt eine geometrische Rekonstruktion der 3D-Anatomie des Kniegelenks. Anhand dieser wird der Gelenkspalt bestimmt, sowie ein Spacer ausgewählt und algorithmisch vorpositioniert. Die Positionierung des Spacers ist durch den Benutzer jederzeit interaktiv anpassbar.
Für jede Positionierung eines Spacers wird ein Fitness-Wert zur Knieanatomie des jeweiligen Patienten berechnet und den Nutzern Rückmeldung hinsichtlich Passgenauigkeit gegeben. Die Software unterstützt somit als Entscheidungshilfe die behandelnden Ärzte bei der patientenspezifischen Spacerauswahl.
In this study we investigate methods for fitting a Statistical Shape Model (SSM) to intraoperatively acquired point cloud data from a surgical navigation system. We validate the fitted models against the pre-operatively acquired Magnetic Resonance Imaging (MRI) data from the same patients.
We consider a cohort of 10 patients who underwent navigated total knee arthroplasty. As part of the surgical protocol the patients’ distal femurs were partially digitized. All patients had an MRI scan two months pre-operatively. The MRI data were manually segmented and the reconstructed bone surfaces used as ground truth against which the fit was compared. Two methods were used to fit the SSM to the data, based on (1) Iterative Closest Points (ICP) and (2) Gaussian Mixture Models (GMM).
For both approaches, the difference between model fit and ground truth surface averaged less than 1.7 mm and excellent correspondence with the distal femoral morphology can be demonstrated.
Deformable model-based approaches to 3D image segmentation have been shown to be highly successful. Such methodology requires an appearance model that drives the deformation of a geometric model to the image data. Appearance models are usually either created heuristically or through supervised learning. Heuristic methods have been shown to work effectively in many applications but are hard to transfer from one application (imaging modality/anatomical structure) to another. On the contrary, supervised learning approaches can learn patterns from a collection of annotated training data. In this work, we show that the supervised joint dictionary learning technique is capable of overcoming the traditional drawbacks of the heuristic approaches. Our evaluation based on two different applications (liver/CT and knee/MR) reveals that our approach generates appearance models, which can be used effectively and efficiently in a deformable model-based segmentation framework.
We present a novel framework for fully automated and highly accurate determination of facial landmarks and dense correspondence, e.g. a topologically identical mesh of arbitrary resolution, across the entire surface of 3D face models. For robustness and reliability of the proposed approach, we are combining 2D landmark detectors and 3D statistical shape priors with a variational matching method. Instead of matching faces in the spatial domain only, we employ image registration to align the 2D parametrization of the facial surface to a planar template we call the Unified Facial Parameter Domain (ufpd). This allows us to simultaneously match salient photometric and geometric facial features using robust image similarity measures while reasonably constraining geometric distortion in regions with less significant features. We demonstrate the accuracy of the dense correspondence established by our framework on the BU3DFE database with 2500 facial surfaces and show, that our framework outperforms current state-of-the-art methods with respect to the fully automated location of facial landmarks.