Quality control (QC) of structures segmentation in volumetric medical images is important for identifying segmentation errors in clinical practice and for facilitating model development by enhancing network performance in semi-supervised and active learning scenarios. This paper introduces SegQC, a novel framework for segmentation quality estimation and segmentation error detection. SegQC computes an estimate measure of the quality of a segmentation in volumetric scans and in their individual slices and identifies possible segmentation error regions within a slice. The key components of SegQC include: 1) SegQCNet, a deep network that inputs a scan and its segmentation mask and outputs segmentation error probabilities for each voxel in the scan; 2) three new segmentation quality metrics computed from the segmentation error probabilities; 3) a new method for detecting possible segmentation errors in scan slices computed from the segmentation error probabilities. We introduce a novel evaluation scheme to measure segmentation error discrepancies based on an expert radiologist's corrections of automatically produced segmentations that yields smaller observer variability and is closer to actual segmentation errors. We demonstrate SegQC on three fetal structures in 198 fetal MRI scans - fetal brain, fetal body and the placenta. To assess the benefits of SegQC, we compare it to the unsupervised Test Time Augmentation (TTA)-based QC and to supervised autoencoder (AE)-based QC. Our studies indicate that SegQC outperforms TTA-based quality estimation for whole scans and individual slices in terms of Pearson correlation and MAE for fetal body and fetal brain structures segmentation as well as for volumetric overlap metrics estimation of the placenta structure. Compared to both unsupervised TTA and supervised AE methods, SegQC achieves lower MAE for both 3D and 2D Dice estimates and higher Pearson correlation for volumetric Dice. Our segmentation error detection method achieved recall and precision rates of 0.77 and 0.48 for fetal body, and 0.74 and 0.55 for fetal brain segmentation error detection, respectively. Ranking derived from metrics estimation surpasses rankings based on entropy and sum for TTA and SegQCNet estimations, respectively. SegQC provides high-quality metrics estimation for both 2D and 3D medical images as well as error localization within slices, offering important improvements to segmentation QC.

Computer-assisted preoperative planning of pelvic fracture reduction surgery has the potential to increase the accuracy of the surgery and to reduce complications. However, the diversity of the pelvic fractures and the disturbance of small fracture fragments present a great challenge to perform reliable automatic preoperative planning. In this paper, we present a comprehensive and automatic preoperative planning pipeline for pelvic fracture surgery. It includes pelvic fracture labeling, reduction planning of the fracture, and customized screw implantation. First, automatic bone fracture labeling is performed based on the separation of the fracture sections. Then, fracture reduction planning is performed based on automatic extraction and pairing of the fracture surfaces. Finally, screw implantation is planned using the adjoint fracture surfaces. The proposed pipeline was tested on different types of pelvic fracture in 14 clinical cases. Our method achieved a translational and rotational accuracy of 2.56 mm and 3.31° in reduction planning. For fixation planning, a clinical acceptance rate of 86.7% was achieved. The results demonstrate the feasibility of the clinical application of our method. Our method has shown accuracy and reliability for complex multi-body bone fractures, which may provide effective clinical preoperative guidance and may improve the accuracy of pelvic fracture reduction surgery.

In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.

In-utero fetal MRI is emerging as an important tool in the diagnosis and analysis of the developing human brain. Automatic segmentation of the developing fetal brain is a vital step in the quantitative analysis of prenatal neurodevelopment both in the research and clinical context. However, manual segmentation of cerebral structures is time-consuming and prone to error and inter-observer variability. Therefore, we organized the Fetal Tissue Annotation (FeTA) Challenge in 2021 in order to encourage the development of automatic segmentation algorithms on an international level. The challenge utilized FeTA Dataset, an open dataset of fetal brain MRI reconstructions segmented into seven different tissues (external cerebrospinal fluid, gray matter, white matter, ventricles, cerebellum, brainstem, deep gray matter). 20 international teams participated in this challenge, submitting a total of 21 algorithms for evaluation. In this paper, we provide a detailed analysis of the results from both a technical and clinical perspective. All participants relied on deep learning methods, mainly U-Nets, with some variability present in the network architecture, optimization, and image pre- and post-processing. The majority of teams used existing medical imaging deep learning frameworks. The main differences between the submissions were the fine tuning done during training, and the specific pre- and post-processing steps performed. The challenge results showed that almost all submissions performed similarly. Four of the top five teams used ensemble learning methods. However, one team's algorithm performed significantly superior to the other submissions, and consisted of an asymmetrical U-Net network architecture. This paper provides a first of its kind benchmark for future automatic multi-tissue segmentation algorithms for the developing human brain in utero.

Pelvic fracture is a severe trauma with a high rate of morbidity and mortality. Accurate and automatic diagnosis and surgical planning of pelvic fracture require effective identification and localization of the fracture zones. This is a challenging task due to the complexity of pelvic fractures, which often exhibit multiple fragments and sites, large fragment size differences, and irregular morphology. We have developed a novel two-stage method for the automatic identification and localization of complex pelvic fractures. Our method is unique in that it allows to combine the symmetry properties of the pelvic anatomy and capture the symmetric feature differences caused by the fracture on both the left and right sides, thereby overcoming the limitations of existing methods which consider only image or geometric features. It implements supervised contrastive learning with a novel Siamese deep neural network, which consists of two weight-shared branches with a structural attention mechanism, to minimize the confusion of local complex structures of the pelvic bones with the fracture zones. A structure-focused attention (SFA) module is designed to capture the spatial structural features and enhances the recognition ability of fracture zones. Comprehensive experiments on 103 clinical CT scans from the publicly available dataset CTPelvic1K show that our method achieves a mean accuracy and sensitivity of 0.92 and 0.93, which are superior to those reported with three SOTA contrastive learning methods and five advanced classification networks, demonstrating the effectiveness of identifying and localizing various types of complex pelvic fractures from clinical CT images.

We describe a new method for the automatic detection of changes in repeat CT scanning with a reduced X-ray radiation dose. We present a theoretical formulation of the automatic change detection problem based on the on-line sparse-view repeat CT scanning dose optimization framework. We prove that the change detection problem is NP-hard and therefore cannot be efficiently solved exactly. We describe a new greedy change detection algorithm that is simple and robust and relies on only two key parameters. We demonstrate that the greedy algorithm accurately detects small, low contrast changes with only 12 scan angles. Our experimental results show that the new algorithm yields a mean changed region recall rate >89% and a mean precision rate >76%. It outperforms both our previous heuristic approach and a thresholding method using a low-dose prior image-constrained compressed sensing (PICCS) reconstruction of the repeat scan. The resulting changed region map may obviate the need for a high-quality repeat scan image when no major changes are detected and may streamline the radiologist's workflow by highlighting the regions of interest.

This paper describes a new method for imageless needle and patient tracking in interventional CT procedures based on fractional CT scanning. Our method accurately locates a needle with a spherical marker attached to it at a known distance from the tip with respect to the patient in the CT scanner coordinate frame with online sparse scan sampling and without reconstructing the CT image. The key principle of our method is to detect the needle and attached spherical marker in projection (sinogram) space based on the strongly attenuated X-ray signal due to the metallic composition of the needle and the needle's thin cylindrical geometry, and based on the marker's spherical geometry. A transformation from projection space to physical space uniquely determines the location and orientation of the needle and the needle tip position. Our method works directly in projection space and simultaneously performs patient registration and needle localization for every fractional CT scanning acquisition using the same sparse set of views. We performed registration and needle tip localization in five abdomen phantom scans using a rigid needle, and obtained a voxel-size tip localization error. Our experimental results indicate a voxel-sized deviation of the localization from a comparable method in 3-D image space, with the benefit of allowing X-ray dose reduction via fractional scanning at each localization. This benefit enables more frequent tip localizations during needle insertion for a similar total dose, or a reduced total dose for the same frequency of tip localization.

We present a new method for rigid registration of CT datasets in 3D Radon space based on sparse sampling of scanning projections. The inputs are the two 3D Radon transforms of the CT scans, one densely sampled and the other sparsely sampled (limited number of scan angles/ranges). The output is the rigid transformation that best matches them. The method first finds the best matching between each projection direction vector in the sparse transform and the corresponding direction vector in the dense transform. It then solves a system of linear equations derived from the direction vector pairs (parallel-beam projections) or finds a solution by non-linear optimization (fan-beam and cone-beam projections). Experimental studies show that our method for 3D parallel beam registration outperforms image space registration in terms of convergence range with significantly reduced X-ray dose compared to a full conventional CT scan.

We present a new method for on-line radiation dose optimization in repeat computer tomography (CT) scanning. Our method uses the information of the baseline scan during the repeat scanning to significantly reduce the radiation dose without compromising the repeat scan quality. It automatically registers the patient to the baseline scan using fractional scanning and detects in sinogram space the patient regions where changes have occurred without having to reconstruct the repeat scan image. It scans only these regions in the patient, thereby considerably reducing the necessary radiation dose. It then completes the missing values of the sparsely sampled repeat scan sinogram with those of the fully sampled baseline sinogram in regions where no changes were detected and computes the repeat scan image by standard filtered backprojection reconstruction. Experiments on a patient scan with simulated changes yield a mean recall of 98% using <19% of a full dose. Experiments on real CT scans of an abdomen phantom produce similar results, with a mean recall of 94.5% and only 14.4% of a full dose more than the theoretical optimum. As hardly any changed rays are missed, the reconstructed images are practically indistinguishable from a full dose scan. Our method successfully detects small, low contrast changes and produces an accurate repeat scan reconstruction using three times less radiation than an image space baseline method.

This paper presents new methods for the optimal selection of anatomical landmarks and optimal placement of fiducial markers in image-guided neurosurgery. These methods allow the surgeon to optimally plan fiducial marker locations on routine diagnostic images before preoperative imaging and to intraoperatively select the set of fiducial markers and anatomical landmarks that minimize the expected target registration error (TRE). The optimization relies on a novel empirical simulation-based TRE estimation method built on actual fiducial localization error (FLE) data. Our methods take the guesswork out of the registration process and can reduce localization error without additional imaging and hardware. Our clinical experiments on five patients who underwent brain surgery with a navigation system show that optimizing one marker location and the anatomical landmarks configuration reduced the TRE. The average TRE values using the usual fiducials setup and using the suggested method were 4.7 mm and 3.2 mm, respectively. We observed a maximum improvement of 4 mm. Reducing the target registration error has the potential to support safer and more accurate minimally invasive neurosurgical procedures.

Classical computational geometry algorithms handle geometric constructs whose shapes and locations are exact. However, many real-world applications require modeling and computing with geometric uncertainties, which are often coupled and mutually dependent. In this paper we address distance problems and orthogonal range queries in the plane, subject to geometric uncertainty. Point coordinates and range uncertainties are modeled with the Linear Parametric Geometric Uncertainty Model (LPGUM), a general and computationally efficient worst-case, first-order linear approximation of geometric uncertainty that supports dependence among uncertainties. We present algorithms for closest pair, diameter and bounding box problems, and efficient algorithms for uncertain range queries: uncertain range/nominal points, nominal range/uncertain points, uncertain range/uncertain points, with independent/dependent uncertainties.

Classical computational geometry algorithms handle geometric constructs whose shapes and locations are exact. However, many real-world applications require computing with geometric uncertainties, which are often coupled and mutually dependent. Existing uncertainty models cannot be used to handle dependencies among objects resulting in overestimation of the mutual errors. We have recently developed the Linear Parametric Geometric Uncertainty Model (LPGUM), a general and computationally efficient worst-case first-order linear approximation of geometric uncertainty that supports dependencies among uncertainties. In this paper, we present the properties of the uncertainty zones of a point and a line, and offer efficient algorithms to compute them. We also describe new efficient algorithms to handle relative position queries, e.g., the classification of an uncertain point with respect to an uncertain line. We show that, in all cases, the overhead of computing with dependent uncertainties is low.

The assembly planning problem has received significant attention due to its importance in autonomous manufacturing. Typical assembly planners assume that parts have nominal shapes, while in reality their geometry varies according to the tolerance specifications. To account for toleranced parts, an assembly plan must be feasible for all possible variations of its components. Despite its practical importance, very few works address this problem. This paper presents a general framework for mechanical assembly planning with toleranced planar parts and shows how to incorporate it into existing planners. Our framework uses a general tolerancing model for parts: vertices are standard elementary functions of the part dimensions, which are allowed to vary within tolerance intervals. The relative position of parts is uniquely determined by an assembly graph, which defines constraints between features of neighboring parts. The assembly graph supports placements of parts with rotational degrees of freedom, cyclic relations, and conditional constraints, which occur in non-nominal contacts between edges. Using this framework, we show how to augment existing algorithms for useful motion types, including single and multiple step translations, and infinitesimal rigid motions. We demonstrate the dramatic reduction in the number of valid assembly plans when tolerances are introduced.

This paper presents a novel image-guided robot-based system to assist orthopedic surgeons in performing distal locking of long bone intramedullary nails. The system consists of a bone-mounted miniature robot fitted with a drill guide that provides rigid mechanical guidance for hand-held drilling of the distal screws' pilot holes. The robot is automatically positioned so that the drill guide and nail distal locking axes coincide, using a single fluoroscopic X-ray image. Since the robot is rigidly attached to the intramedullary nail or bone, no leg immobilization or real-time tracking is required. We describe the system and protocol and present a method for accurate and robust drill guide and nail hole localization and registration. The in vitro system accuracy experiments for fronto-parallel viewing show a mean angular error of 1.3 degrees (std = 0.4 degrees ) between the computed drill guide axes and the actual locking holes axes, and a mean 3.0 mm error (std = 1.1 mm) in the entry and exit drill point, which is adequate for successfully locking the nail.

This paper presents a new method for creating a single panoramic image of a long bone from several individual fluoroscopic X-ray images. Panoramic images are useful preoperatively for diagnosis, and intraoperatively for long bone fragment alignment, for making anatomical measurements, and for documenting surgical outcomes. Our method composes individual overlapping images into an undistorted panoramic view that is the equivalent of a single X-ray image with a wide field of view. The correlations between the images are established from the graduations of a radiolucent ruler imaged alongside the long bone. Unlike existing methods, ours uses readily available hardware, requires a simple image acquisition protocol with minimal user input, and works with existing fluoroscopic C-arm units without modifications. It is robust and accurate, producing panoramas whose quality and spatial resolution is comparable to that of the individual images. The method has been successfully tested on in vitro and clinical cases.

We present a gradient-based method for rigid registration of a patient preoperative computed tomography (CT) to its intraoperative situation with a few fluoroscopic X-ray images obtained with a tracked C-arm. The method is noninvasive, anatomy-based, requires simple user interaction, and includes validation. It is generic and easily customizable for a variety of routine clinical uses in orthopaedic surgery. Gradient-based registration consists of three steps: 1) initial pose estimation; 2) coarse geometry-based registration on bone contours, and; 3) fine gradient projection registration (GPR) on edge pixels. It optimizes speed, accuracy, and robustness. Its novelty resides in using volume gradients to eliminate outliers and foreign objects in the fluoroscopic X-ray images, in speeding up computation, and in achieving higher accuracy. It overcomes the drawbacks of intensity-based methods, which are slow and have a limited convergence range, and of geometry-based methods, which depend on the image segmentation quality. Our simulated, in vitro, and cadaver experiments on a human pelvis CT, dry vertebra, dry femur, fresh lamb hip, and human pelvis under realistic conditions show a mean 0.5-1.7 mm (0.5-2.6 mm maximum) target registration accuracy.

This paper presents a new approach to robot-assisted spine and trauma surgery in which a miniature robot is directly mounted on the patient's bony structure near the surgical site. The robot is designed to operate in a semiactive mode to precisely position and orient a drill or a needle in various surgical procedures. Since the robot forms a single rigid body with the anatomy, there is no need for immobilization or motion tracking, which greatly enhances and simplifies the robot's registration to the target anatomy. To demonstrate this concept, we developed the MiniAture Robot for Surgical procedures (MARS), a cylindrical 5/spl times/7 cm/sup 3/, 200-g, six-degree-of-freedom parallel manipulator. We are currently developing two clinical applications to demonstrate the concept: 1) surgical tools guiding for spinal pedicle screws placement; and 2) drill guiding for distal locking screws in intramedullary nailing. In both cases, a tool guide attached to the robot is positioned at a planned location with a few intraoperative fluoroscopic X-ray images. Preliminary in-vitro experiments demonstrate the feasibility of this concept.

We present the first configuration space computation algorithm for pairs of rigid parts that move along fixed spatial axes. The motivation is contact analysis for mechanical design of spatial systems and of planar systems with axis misalignment. The part geometry is specified in a parametric boundary representation using planes, cylinders, and spheres. Our strategy is to exploit the specialized part geometry and the 2D structure of the configuration space to: 1) derive low-degree algebraic contact equations in the two part motion parameters, which can readily be solved to obtain contact curves, and 2) to use a practical planar configuration space construction algorithm. We demonstrate a preliminary implementation on three representative pairs, none of which is covered by other contact analysis algorithms. We show how the program is used in answering design questions.

This paper describes an ongoing project to develop a computer-integrated system to assist surgeons in revision total hip replacement (RTHR) surgery. In RTHR surgery, a failing orthopedic hip implant, typically cemented, is replaced with a new one by removing the old implant, removing the cement and fitting a new implant into an enlarged canal broached in the femur. RTHR surgery is a difficult procedure fraught with technical challenges and a high incidence of complications. The goals of the computer-based system are the significant reduction of cement removal labor and time, the elimination of cortical wall penetration and femur fracture, the improved positioning and fit of the new implant resulting from precise, high-quality canal milling and the reduction of bone sacrificed to fit the new implant. Our starting points are the ROBODOC system for primary hip replacement surgery and the manual RTHR surgical protocol. We first discuss the main difficulties of computer-integrated RTHR surgery and identify key issues and possible solutions. We then describe possible system architectures and protocols for preoperative planning and intraoperative execution. We present a summary of methods and preliminary results in CT image metal artifact removal, interactive cement cut-volume definition and cement machining, anatomy-based registration using fluoroscopic X-ray images and clinical trials using an extended RTHR version of ROBODOC. We conclude with a summary of lessons learned and a discussion of current and future work.

Presents a method for recognizing mechanisms and describing their behaviours from image sequences showing their relations. It uses a simple and expressive language for describing the behaviour of fixed-axes mechanisms. The language symbolically captures the important aspects of the kinematics and the simple dynamics of the mechanism. We show how this language combined with a vision system can automatically identify mechanisms and their behaviours from a sequence of images.

We address the problem of efficiently determining if the intersection of a given set of d-dimensional halfspaces is unbounded. It is shown that detecting unboundedness can be reduced to a single linear range computation followed by a single linear feasibility test. In contrast, detecting unboundedness is at least as hard as linear feasibility testing and maximization. Our analysis suggests that algorithms for establishing linear unboundedness can be used as a basis of simple and practical algorithms in motion planning, insertability analysis and assembly planning. We show that m-handed assembly planning can be reduced to testing for unboundedness. A valid motion sequence can be computed in polynomial time, if the parts are not already separated in their initial placement. No polynomial algorithms were previously known for this problem. We present experimental results obtained with an implementation of our algorithms.

We are studying difficult geometric problems in computer-aided mechanical design where visualization plays a key role. The research addresses the fundamental design task of contact analysis: deriving the part contacts and the ensuing motion constraints in a mechanical system. We have automated contact analysis of general planar systems via configuration space computation. Configuration space is a geometric representation of rigid-body interaction that encodes quantitative information, such as part motion paths, and qualitative information, such as system failure modes. The configuration space dimension equals the number of degrees of freedom in the system. Three-dimensional spaces are most important, but higher-dimensions are often useful. The qualitative aspects, which relate to the topology of the configuration space, are best understood by visualization. We explain what configuration space is, how it encodes contact information, and what research challenges it poses for visualization.

We present an algorithm for functional kinematic tolerance analysis of general planar mechanical systems with parametric tolerances. The algorithm performs worst-case analysis of systems of curved parts with contact changes, including open and closed kinematic chains. It computes quantitative variations and helps designers detect qualitative variations, such as blocking and under-cutting. The algorithm constructs a variation model for each interacting pair of parts: a mapping from the part tolerances and configurations to the kinematic variation of the pair. These models generalize the configuration space representation of nominal kinematics to toleranced parts. They are composed via sensitivity analysis and linear programming to derive the system variation at a given configuration. The variation relative to the nominal system function is computed by sampling the system variation. We demonstrate the algorithm on detailed parametric models of a movie camera film advance and of a micro-mechanical gear discriminator.

In this work, a new method for mesh simplification and surface reconstruction specifically designed for the needs of CAD/CAM engineering design and analysis is introduced. The method simplifies the original free-form face model by first constructing restricted curvature deviation regions, generating a boundary conforming finite element quadrilateral mesh of the regions, and then fitting a smooth surface over the quadrilateral mesh using the plate energy method. It is more general in scope than existing methods because it handles models with free-form faces and non-manifold geometry, not just triangular or polygonal faces. It produces a high-quality quadrilateral mesh which is suited for both finite element analysis and CAD/CAM. The smooth surface obtained by energy functional stabilization over limited curvature regions preserves the number of quadrilateral elements, and is best suited for surface modeling. The method is illustrated by building of several free-form surfaces from an arbitrary topology mesh.

We present an algorithm for dynamical simulation of rigid-body mechanical systems with changing contact topologies based on configuration spaces. The algorithm advances the state of the art in contact analysis, which is the main bottleneck in dynamical simulation. The task is to identify the touching parts and to compute the ensuing contact forces. Our algorithm computes the configuration spaces of all pairs of parts and uses them as contact models. It overcomes the limitations of mechanical systems simulators, which require precomputed contact models, and of general-body simulators, which perform contact analysis on the part models at every time step. Neither approach is practical for mechanisms with multiple contacts and complex contact geometry, such as clock escapements, chain gears, and part feeders. We describe a configuration space simulator for assemblies of planar parts with one degree of freedom apiece and demonstrate it on two mechanisms with many complex contacts.

No abstract available.

We describe the HIPAIR configuration space computation program for higher pairs and show how it automates reasoning about shape and motion for mechanism design. We describe an interactive parametric design module that combines configuration space computation with differential constraint satisfaction. HIPAIR handles pairs of 2.5D parts with two degrees of freedom, including pairs with intermittent, simultaneous, and degenerate contacts. This class contains 90% of 2.5D pairs and 80% of all higher pairs according to our survey of 2500 mechanisms. We have tested HIPAIR on over 100 pairs, including gears, cams, ratchets, and escapements. It analyzes pairs with thousands of contacts in under ten seconds. The configuration spaces encode the relations among part shapes, part motions, and overall behavior in a concise, complete, and explicit format. They help designers analyze part interactions, implement functions, identify failure modes, and modify designs.>