Towards clinical applicability and computational efficiency in automatic cranial implant design: An overview of the AutoImplant 2021 cranial implant design challenge.

Cranial implants are commonly used for surgical repair of craniectomy-induced skull defects. These implants are usually generated offline and may require days to weeks to be available. An automated implant design process combined with onsite manufacturing facilities can guarantee immediate implant availability and avoid secondary intervention. To address this need, the AutoImplant II challenge was organized in conjunction with MICCAI 2021, catering for the unmet clinical and computational requirements of automatic cranial implant design. The first edition of AutoImplant (AutoImplant I, 2020) demonstrated the general capabilities and effectiveness of data-driven approaches, including deep learning, for a skull shape completion task on synthetic defects. The second AutoImplant challenge (i.e., AutoImplant II, 2021) built upon the first by adding real clinical craniectomy cases as well as additional synthetic imaging data. The AutoImplant II challenge consisted of three tracks. Tracks 1 and 3 used skull images with synthetic defects to evaluate the ability of submitted approaches to generate implants that recreate the original skull shape. Track 3 consisted of the data from the first challenge (i.e., 100 cases for training, and 110 for evaluation), and Track 1 provided 570 training and 100 validation cases aimed at evaluating skull shape completion algorithms at diverse defect patterns. Track 2 also made progress over the first challenge by providing 11 clinically defective skulls and evaluating the submitted implant designs on these clinical cases. The submitted designs were evaluated quantitatively against imaging data from post-craniectomy as well as by an experienced neurosurgeon. Submissions to these challenge tasks made substantial progress in addressing issues such as generalizability, computational efficiency, data augmentation, and implant refinement. This paper serves as a comprehensive summary and comparison of the submissions to the AutoImplant II challenge. Codes and models are available at https://github.com/Jianningli/Autoimplant_II.

Estimating Growth in Height from Limited Longitudinal Growth Data Using Full-Curves Training Dataset: A Comparison of Two Procedures of Curve Optimization-Functional Principal Component Analysis and SITAR.

A variety of models are available for the estimation of parameters of the human growth curve. Several have been widely and successfully used with longitudinal data that are reasonably complete. On the other hand, the modeling of data for a limited number of observation points is problematic and requires the interpolation of the interval between points and often an extrapolation of the growth trajectory beyond the range of empirical limits (prediction). This study tested a new approach for fitting a relatively limited number of longitudinal data using the normal variation of human empirical growth curves. First, functional principal components analysis was done for curve phase and amplitude using complete and dense data sets for a reference sample (Brno Growth Study). Subsequently, artificial curves were generated with a combination of 12 of the principal components and applied for fitting to the newly analyzed data with the Levenberg-Marquardt optimization algorithm. The approach was tested on seven 5-points/year longitudinal data samples of adolescents extracted from the reference sample. The samples differed in their distance from the mean age at peak velocity for the sample and were tested by a permutation leave-one-out approach. The results indicated the potential of this method for growth modeling as a user-friendly application for practical applications in pediatrics, auxology and youth sport.

Deep learning for cranioplasty in clinical practice: Going from synthetic to real patient data.

Correct virtual reconstruction of a defective skull is a prerequisite for successful cranioplasty and its automatization has the potential for accelerating and standardizing the clinical workflow. This work provides a deep learning-based method for the reconstruction of a skull shape and cranial implant design on clinical data of patients indicated for cranioplasty. The method is based on a cascade of multi-branch volumetric CNNs that enables simultaneous training on two different types of cranioplasty ground-truth data: the skull patch, which represents the exact shape of the missing part of the original skull, and which can be easily created artificially from healthy skulls, and expert-designed cranial implant shapes that are much harder to acquire. The proposed method reaches an average surface distance of the reconstructed skull patches of 0.67 mm on a clinical test set of 75 defective skulls. It also achieves a 12% reduction of a newly proposed defect border Gaussian curvature error metric, compared to a baseline model trained on synthetic data only. Additionally, it produces directly 3D printable cranial implant shapes with a Dice coefficient 0.88 and a surface error of 0.65 mm. The outputs of the proposed skull reconstruction method reach good quality and can be considered for use in semi- or fully automatic clinical cranial implant design workflows.

AutoImplant 2020-First MICCAI Challenge on Automatic Cranial Implant Design.

The aim of this paper is to provide a comprehensive overview of the MICCAI 2020 AutoImplant Challenge. The approaches and publications submitted and accepted within the challenge will be summarized and reported, highlighting common algorithmic trends and algorithmic diversity. Furthermore, the evaluation results will be presented, compared and discussed in regard to the challenge aim: seeking for low cost, fast and fully automated solutions for cranial implant design. Based on feedback from collaborating neurosurgeons, this paper concludes by stating open issues and post-challenge requirements for intra-operative use. The codes can be found at https://github.com/Jianningli/tmi.

SkullBreak / SkullFix - Dataset for automatic cranial implant design and a benchmark for volumetric shape learning tasks.

The article introduces two complementary datasets intended for the development of data-driven solutions for cranial implant design, which remains to be a time-consuming and laborious task in current clinical routine of cranioplasty. The two datasets, referred to as the SkullBreak and SkullFix in this article, are both adapted from a public head CT collection CQ500 (http://headctstudy.qure.ai/dataset) with CC BY-NC-SA 4.0 license. The SkullBreak contains 114 and 20 complete skulls, each accompanied by five defective skulls and the corresponding cranial implants, for training and evaluation respectively. The SkullFix contains 100 triplets (complete skull, defective skull and the implant) for training and 110 triplets for evaluation. The SkullFix dataset was first used in the MICCAI 2020 AutoImplant Challenge (https://autoimplant.grand-challenge.org/) and the ground truth, i.e., the complete skulls and the implants in the evaluation set are held private by the organizers. The two datasets are not overlapping and differ regarding data selection and synthetic defect creation and each serves as a complement to the other. Besides cranial implant design, the datasets can be used for the evaluation of volumetric shape learning algorithms, such as volumetric shape completion. This article gives a description of the two datasets in detail.

Skull shape reconstruction using cascaded convolutional networks.

Designing a cranial implant to restore the protective and aesthetic function of the patient's skull is a challenging process that requires a substantial amount of manual work, even for an experienced clinician. While computer-assisted approaches with various levels of required user interaction exist to aid this process, they are usually only validated on either a single type of simple synthetic defect or a very limited sample of real defects. The work presented in this paper aims to address two challenges: (i) design a fully automatic 3D shape reconstruction method that can address diverse shapes of real skull defects in various stages of healing and (ii) to provide an open dataset for optimization and validation of anatomical reconstruction methods on a set of synthetically broken skull shapes. We propose an application of the multi-scale cascade architecture of convolutional neural networks to the reconstruction task. Such an architecture is able to tackle the issue of trade-off between the output resolution and the receptive field of the model imposed by GPU memory limitations. Furthermore, we experiment with both generative and discriminative models and study their behavior during the task of anatomical reconstruction. The proposed method achieves an average surface error of 0.59mm for our synthetic test dataset with as low as 0.48mm for unilateral defects of parietal and temporal bone, matching state-of-the-art performance while being completely automatic. We also show that the model trained on our synthetic dataset is able to reconstruct real patient defects.

Indoor and Outdoor Backpack Mapping with Calibrated Pair of Velodyne LiDARs.

This paper presents a human-carried mapping backpack based on a pair of Velodyne LiDAR scanners. Our system is a universal solution for both large scale outdoor and smaller indoor environments. It benefits from a combination of two LiDAR scanners, which makes the odometry estimation more precise. The scanners are mounted under different angles, thus a larger space around the backpack is scanned. By fusion with GNSS/INS sub-system, the mapping of featureless environments and the georeferencing of resulting point cloud is possible. By deploying SoA methods for registration and the loop closure optimization, it provides sufficient precision for many applications in BIM (Building Information Modeling), inventory check, construction planning, etc. In our indoor experiments, we evaluated our proposed backpack against ZEB-1 solution, using FARO terrestrial scanner as the reference, yielding similar results in terms of precision, while our system provides higher data density, laser intensity readings, and scalability for large environments.

Intensity-Based Nonoverlapping Area Registration Supporting "Drop-Outs" in Terms of Model-Based Radiostereometric Analysis.

A model-based radiostereometric analysis (MBRSA) is a method for precise measurement of prosthesis migration, which does not require marking the implant with tantalum beads. Instead, the prosthesis pose is typically recovered using a feature-based 2D-3D registration of its virtual model into a stereo pair of radiographs. In this study, we evaluate a novel intensity-based formulation of previously published nonoverlapping area (NOA) approach. The registration is capable of performing with both binary radiographic segmentations and nonsegmented X-ray images. In contrast with the feature-based version, it is capable of dealing with unreliable parts of prosthesis. As the straightforward formulation allows efficient acceleration using modern graphics adapters, it is possible to involve precise high-poly virtual models. Moreover, in case of binary segmentations, the nonoverlapping area is simply interpretable and useful for indicating the accuracy of the registration outcome. In silico and phantom evaluations were performed using a cementless Zweymüller femoral stem and its reverse engineered (RE) model. For initial pose estimates with difference from the ground-truth limited to ±4 mm and ±4°, respectively, the mean absolute translational error was not higher than 0.042 ± 0.035 mm. The error in rotation around the proximodistal axis was 0.181 ± 0.265°, and the error for the remaining axes was not higher than 0.035 ± 0.037°.

Network collaborative environment supporting 3D medicine.

This paper is focused on the virtual collaborative consultation system which is intended for support of 3D geometrical modelling applications in the field of clinical human medicine. The system allows uploading the CT/MR data and 3D tissue geometry models (prepared in advance). The data define a 3D scene, which allows for viewing of the data and consulting them between technicians and physicians over the medium of computer network. The system is conceived as a three layer client-server architecture. For communication between the server and a client, the HTTPS protocol is used. Test results in Czech republic and the world-wide tests as well confirm, that the system is practically applicable and beneficial.

Consultation virtual collaborative environment for 3D medicine.

This article focuses on the problems of consultation virtual collaborative environment, which is designed to support 3D medical applications. This system allows loading CT/MR data from PACS system, segmentation and 3D models of tissues. It allows distant 3D consultations of the data between technicians and surgeons. System is designed as three-layer client-server architecture. Communication between clients and server is done via HTTP/HTTPS protocol. Results and tests have confirmed, that today's standard network latency and dataflow do not affect the usability of our system.

Network collaborative environment for human tissues 3D modeling.

This paper deals with the new concept of network based virtual collaborative environment to support clinical applications of 3D models of human tissues, created from CT/MR data. It is a topic lying between 3D tissue modeling and PACS systems. Designed system allows clinical realizations of 3D applications as a service to clinical workplaces, provided by specialized 3D laboratory, even over great distances. Problem lies within the need of doing necessary consultations, corrections and verifications distantly. This is solved by our system in the form of virtual collaborative environment. This system is built upon three-layer client-server architecture. Our application is focused on 3D tissue modeling. Generally it can be used as a basis for other similar applications.