3D Image Data Research Articles

Novel CNN-Based Approach for Reading Urban Form Data in 2D Images: An Application for Predicting Restaurant Location in Seoul, Korea

Artificial intelligence (AI) has demonstrated its ability to complete complex tasks in various fields. In urban studies, AI technology has been utilized in some limited domains, such as control of traffic and air quality. This study uses AI to better understand diverse urban studies data through a novel approach that uses a convolutional neural network (CNN). In this study, a building outline in the form of a two-dimensional image is used with its corresponding metadata to test the applicability of CNN in reading urban data. MobileNet, a high-efficiency CNN model, is trained to predict the location of restaurants in each building in Seoul, Korea. Consequently, using only 2D image data, the model satisfactorily predicts the locations of restaurants (AUC = 0.732); the model with 2D images and their metadata has higher performance but has an overfitting problem. In addition, the model using only 2D image data accurately predicts the regional distribution of restaurants and shows some typical urban forms with restaurants. The proposed model has several technical limitations but shows the potential to provide a further understanding of urban settings.

Non-Contact Face Temperature Measurement by Thermopile-Based Data Fusion.

Thermal imaging cameras and infrared (IR) temperature measurement devices act as state-of-the-art techniques for non-contact temperature determination of the skin surface. The former is cost-intensive in many cases for widespread application, and the latter requires manual alignment to the measuring point. Due to this background, this paper proposes a new method for automated, non-contact, and area-specific temperature measurement of the facial skin surface. It is based on the combined use of a low-cost thermopile sensor matrix and a 2D image sensor. The temperature values as well as the 2D image data are fused using a parametric affine transformation. Based on face recognition, this allows temperature values to be assigned to selected facial regions and used specifically to determine the skin surface temperature. The advantages of the proposed method are described. It is demonstrated by means of a participant study that the temperature absolute values, which are achieved without manual alignment in an automated manner, are comparable to a commercially available IR-based forehead thermometer.

A novel image deep learning-based sub-centimeter pulmonary nodule management algorithm to expedite resection of the malignant and avoid over-diagnosis of the benign.

With the popularization of chest computed tomography (CT) screening, there are more sub-centimeter (≤ 1cm) pulmonary nodules (SCPNs) requiring further diagnostic workup. This area represents an important opportunity to optimize the SCPN management algorithm avoiding "one-size fits all" approach. One critical problem is how to learn the discriminative multi-view characteristics and the unique context of each SCPN. Here, we propose a multi-view coupled self-attention module (MVCS) to capture the global spatial context of the CT image through modeling the association order of space and dimension. Compared with existing self-attention methods, MVCS uses less memory consumption and computational complexity, unearths dimension correlations that previous methods have not found, and is easy to integrate with other frameworks. In total, a public dataset LUNA16 from LIDC-IDRI, 1319 SCPNs from 1069 patients presenting to a major referral center, and 160 SCPNs from 137 patients from three other major centers were analyzed to pre-train, train, and validate the model. Experimental results showed that performance outperforms the state-of-the-art models in terms of accuracy and stability and is comparable to that of human experts in classifying precancerous lesions and invasive adenocarcinoma. We also provide a fusion MVCS network (MVCSN) by combining the CT image with the clinical characteristics and radiographic features of patients. This tool may ultimately aid in expediting resection of the malignant SCPNs and avoid over-diagnosis of the benign ones, resulting in improved management outcomes. In the diagnosis of sub-centimeter lung adenocarcinoma, fusion MVCSN can help doctors improve work efficiency and guide their treatment decisions to a certain extent. • Advances in computed tomography (CT) not only increase the number of nodules detected, but also the nodules that are identified are smaller, such as sub-centimeter pulmonary nodules (SCPNs). • We propose a multi-view coupled self-attention module (MVCS), which could model spatial and dimensional correlations sequentially for learning global spatial contexts, which is better than other attention mechanisms. • MVCS uses fewer huge memory consumption and computational complexity than the existing self-attention methods when dealing with 3D medical image data. Additionally, it reaches promising accuracy for SCPNs' malignancy evaluation and has lower training cost than other models.

Development of vehicle- recognition method on water surfaces using LiDAR data: SPD2 (spherically stratified point projection with diameter and distance)

Swarm robot systems are an important application of autonomous unmanned surface vehicles on water surfaces. For monitoring natural environments and conducting security activities within a certain range using a surface vehicle, the swarm robot system is more efficient than the operation of a single object as the former can reduce cost and save time. It is necessary to detect adjacent surface obstacles robustly to operate a cluster of unmanned surface vehicles. For this purpose, a LiDAR (light detection and ranging) sensor is used as it can simultaneously obtain 3D information for all directions, relatively robustly and accurately, irrespective of the surrounding environmental conditions. Although the GPS (global-positioning-system) error range exists, obtaining measurements of the surface-vessel position can still ensure stability during platoon maneuvering. In this study, a three-layer convolutional neural network is applied to classify types of surface vehicles. The aim of this approach is to redefine the sparse 3D point cloud data as 2D image data with a connotative meaning and subsequently utilize this transformed data for object classification purposes. Hence, we have proposed a descriptor that converts the 3D point cloud data into 2D image data. To use this descriptor effectively, it is necessary to perform a clustering operation that separates the point clouds for each object. We developed voxel-based clustering for the point cloud clustering. Furthermore, using the descriptor, 3D point cloud data can be converted into a 2D feature image, and the converted 2D image is provided as an input value to the network. We intend to verify the validity of the proposed 3D point cloud feature descriptor by using experimental data in the simulator. Furthermore, we explore the feasibility of real-time object classification within this framework.

Efficient single-pixel imaging encrypted transmission based on 3D Arnold transformation

Single-pixel imaging (SPI) can transform 2D or 3D image data into 1D light signals, which offers promising prospects for image compression and transmission. However, during data communication these light signals in public channels will easily draw the attention of eavesdroppers. Here, we introduce an efficient encryption method for SPI data transmission that uses the 3D Arnold transformation to directly disrupt 1D single-pixel light signals and utilizes the elliptic curve encryption algorithm for key transmission. This encryption scheme immediately employs Hadamard patterns to illuminate the scene and then utilizes the 3D Arnold transformation to permutate the 1D light signal of single-pixel detection. Then the transformation parameters serve as the secret key, while the security of key exchange is guaranteed by an elliptic curve-based key exchange mechanism. Compared with existing encryption schemes, both computer simulations and optical experiments have been conducted to demonstrate that the proposed technique not only enhances the security of encryption but also eliminates the need for complicated pattern scrambling rules. Additionally, this approach solves the problem of secure key transmission, thus ensuring the security of information and the quality of the decrypted images.

Hybrid machine-learning framework for volumetric segmentation and quantification of vacuoles in individual yeast cells using holotomography.

The precise, quantitative evaluation of intracellular organelles in three-dimensional (3D) imaging data poses a significant challenge due to the inherent constraints of traditional microscopy techniques, the requirements of the use of exogenous labeling agents, and existing computational methods. To counter these challenges, we present a hybrid machine-learning framework exploiting correlative imaging of 3D quantitative phase imaging with 3D fluorescence imaging of labeled cells. The algorithm, which synergistically integrates a random-forest classifier with a deep neural network, is trained using the correlative imaging data set, and the trained network is then applied to 3D quantitative phase imaging of cell data. We applied this method to live budding yeast cells. The results revealed precise segmentation of vacuoles inside individual yeast cells, and also provided quantitative evaluations of biophysical parameters, including volumes, concentration, and dry masses of automatically segmented vacuoles.

3D image reconstruction of architectural model based on 3D printing technology

ABSTRACT The goal of this study was to develop a 3D image reconstruction technique for an architectural model based on 3D printing technology with the objective to enhance the reconstruction effect of the architectural model. This method first analyzes the basic process of 3D printing architectural models and 3D reconstruction technology. On this basis, a 3D architectural model is obtained by sorting out the materials, drawing, printing the models, and removing the supporting materials. According to the 3D image data of the architectural model obtained after camera calibration, radial correction, noise reduction, and multi-view camera motion recovery are carried out on the image data. After the experiment, it was found that the absolute error of the location value of the feature points in the X, Y and Z directions of the given model by the method proposed in this study was 0.15, 0.065 and 0.017 respectively, while the absolute error of the corresponding location in the references [6] and [8] was 1.48 and 1.46, 1.601 and 0.995, 0.839 and 0.94 respectively. It is clear that the method developed in this study has a substantially greater reconstruction accuracy and that it accurately extracts 3D data from the architectural model. The degree of correspondence between the reconstruction outcomes and the architectural model has greatly increased.

Influence of quality of reduction using radiological criteria on kinematics and kinetics in ankle fractures with unstable syndesmotic injury

BackgroundIn ankle fractures with syndesmotic injury, the anatomic reduction of the ankle mortise is crucial for preventing osteoarthritis. Yet, no studies have analysed the effect of surgical reduction after unstable ankle fractures on patients' active functional outcome. MethodsThe Intraoperative 3D imaging data of patients surgically treated between 2012 and 2019 for ankle fracture with concomitant syndesmotic injury were reviewed. 58 patients were allocated to two groups depending on whether the criteria for radiologically optimal reduction were met (39 patients) or not (19 patients). Criteria for optimal reduction were composed of objectively measured and subjectively rated data. After undertaking the Olerud/Molander ankle score, a gait analysis and several active function tests using 3D motion capture were performed in order to evaluate kinetic and kinematic differences between both groups. FindingsPatients showed deficits of range of motion and balance parameters on the injured ankle, however, there were no significant differences between both groups. InterpretationAlthough, the data did not show that radiological reduction criteria have a statistically significant effect on active functional outcome after a mean follow up time of 5.7 years, tendencies for a better outcome of patients that met the criteria could be seen. It also must be taken into consideration that results are limited by case number and allocation ratio, which made a sub-analysis of the separate reduction criteria unfeasible.

BIDL: a brain-inspired deep learning framework for spatiotemporal processing.

Brain-inspired deep spiking neural network (DSNN) which emulates the function of the biological brain provides an effective approach for event-stream spatiotemporal perception (STP), especially for dynamic vision sensor (DVS) signals. However, there is a lack of generalized learning frameworks that can handle various spatiotemporal modalities beyond event-stream, such as video clips and 3D imaging data. To provide a unified design flow for generalized spatiotemporal processing (STP) and to investigate the capability of lightweight STP processing via brain-inspired neural dynamics, this study introduces a training platform called brain-inspired deep learning (BIDL). This framework constructs deep neural networks, which leverage neural dynamics for processing temporal information and ensures high-accuracy spatial processing via artificial neural network layers. We conducted experiments involving various types of data, including video information processing, DVS information processing, 3D medical imaging classification, and natural language processing. These experiments demonstrate the efficiency of the proposed method. Moreover, as a research framework for researchers in the fields of neuroscience and machine learning, BIDL facilitates the exploration of different neural models and enables global-local co-learning. For easily fitting to neuromorphic chips and GPUs, the framework incorporates several optimizations, including iteration representation, state-aware computational graph, and built-in neural functions. This study presents a user-friendly and efficient DSNN builder for lightweight STP applications and has the potential to drive future advancements in bio-inspired research.

Multivariate Process Control Chart Pattern Classification Using Multi-Channel Deep Convolutional Neural Networks

Statistical process control (SPC) charts are commonly used to monitor quality characteristics in manufacturing processes. When monitoring two or more related quality characteristics simultaneously, multivariate T2 control charts are often employed. Like univariate control charts, control chart pattern recognition (CCPR) plays a crucial role in multivariate SPC. The presence of non-random patterns in T2 control charts indicates that a process is influenced by one or more assignable causes and that corrective actions should be taken. In this study, we developed a deep learning-based classification model for recognizing control chart patterns in multivariate processes. To address the problem of the insufficient representation of one-dimensional (1D) data, we explore the advantages of using two-dimensional (2D) image data obtained from a threshold-free recurrence plot. A multi-channel deep convolutional neural network (MCDCNN) model was developed to incorporate both 1D and 2D representations of control chart data. This model was tested on multivariate processes with different covariance matrices and compared with other traditional algorithms. Moreover, the effects of imbalanced datasets and dataset size on classification performance were analyzed. Simulation studies revealed that the developed MCDCNN model outperforms other techniques in identifying multivariate non-random patterns. For the most significant one, our proposed MCDCNN method achieved a 10% improvement over traditional methods. The overall results suggest that the developed MCDCNN model can be beneficial for intelligent SPC.

Neural Rendering-Based 3D Scene Style Transfer Method via Semantic Understanding Using a Single Style Image

In the rapidly emerging era of untact (“contact-free”) technologies, the requirement for three-dimensional (3D) virtual environments utilized in virtual reality (VR)/augmented reality (AR) and the metaverse has seen significant growth, owing to their extensive application across various domains. Current research focuses on the automatic transfer of the style of rendering images within a 3D virtual environment using artificial intelligence, which aims to minimize human intervention. However, the prevalent studies on rendering-based 3D environment-style transfers have certain inherent limitations. First, the training of a style transfer network dedicated to 3D virtual environments demands considerable style image data. These data must align with viewpoints that closely resemble those of the virtual environment. Second, there was noticeable inconsistency within the 3D structures. Predominant studies often neglect 3D scene geometry information instead of relying solely on 2D input image features. Finally, style adaptation fails to accommodate the unique characteristics inherent in each object. To address these issues, we propose a novel approach: a neural rendering-based 3D scene-style conversion technique. This methodology employs semantic nearest-neighbor feature matching, thereby facilitating the transfer of style within a 3D scene while considering the distinctive characteristics of each object, even when employing a single style image. The neural radiance field enables the network to comprehend the geometric information of a 3D scene in relation to its viewpoint. Subsequently, it transfers style features by employing the unique features of a single style image via semantic nearest-neighbor feature matching. In an empirical context, our proposed semantic 3D scene style transfer method was applied to 3D scene style transfers for both interior and exterior environments. This application utilizes the replica, 3DFront, and Tanks and Temples datasets for testing. The results illustrate that the proposed methodology surpasses existing style transfer techniques in terms of maintaining 3D viewpoint consistency, style uniformity, and semantic coherence.

Unsupervised abnormality detection in neonatal MRI brain scans using deep learning

Analysis of 3D medical imaging data has been a large topic of focus in the area of Machine Learning/Artificial Intelligence, though little work has been done in algorithmic (particularly unsupervised) analysis of neonatal brain MRI’s. A myriad of conditions can manifest at an early age, including neonatal encephalopathy (NE), which can result in lifelong physical consequences. As such, there is a dire need for better biomarkers of NE and other conditions. The objective of the study is to improve identification of anomalies and prognostication of neonatal MRI brain scans. We introduce a framework designed to support the analysis and assessment of neonatal MRI brain scans, the results of which can be used as an aid to neuroradiologists. We explored the efficacy of the framework through iterations of several deep convolutional Autoencoder (AE) unsupervised modeling architectures designed to learn normalcy of the neonatal brain structure. We tested this framework on the developing human connectome project (dHCP) dataset with 97 patients that were previously categorized by severity. Our framework demonstrated the model’s ability to identify and distinguish subtle morphological signatures present in brain structures. Normal and abnormal neonatal brain scans can be distinguished with reasonable accuracy, correctly categorizing them in up to 83% of cases. Most critically, new brain anomalies originally missed during the radiological reading were identified and corroborated by a neuroradiologist. This framework and our modeling approach demonstrate an ability to improve prognostication of neonatal brain conditions and are able to localize new anomalies.

Laboratory investigation of railway-used ballast morphology using 3D imaging data analyses

Railway ballast deteriorates over time due to cyclic loading and environmental factors, necessitating its replacement to maintain track stability. However, the potential shortage of fresh ballast has posed challenges in replacing used ballast. One of the challenges in recycling used ballast is determining its useful life. This study proposes a method to characterize fresh and used ballast morphology and shape properties using digital images and advanced data analysis software. The aim is to establish a threshold that distinguishes when ballast can be considered “used.”. The study utilizes basic morphology parameters of flatness, elongation, roundness, sphericity, and convexity. Used ballast is generated through the Los Angeles abrasion (LAA) test method, while fresh ballast serves as the control sample. The analysis suggests that specific combinations of shape indices, such as roundness and convexity, offer insights into differentiating between the two types of ballast. Notably, significant differences are observed in the sphericity and convexity indices between fresh and used ballast samples. However, A systematic definition of used ballast based on the basic morphology indices alone was not feasible. Only the combination of sphericity index (> 0.825) and convexity index (> 0.9), by which both indices of used ballast were higher than those of fresh ballast, could roughly classify fresh and used ballast, although a minor overlap persisted. Further exploration of the data using advanced statistical techniques and machine learning could enhance the accuracy of defining used ballast.

1D Gradient-Weighted Class Activation Mapping, Visualizing Decision Process of Convolutional Neural Network-Based Models in Spectroscopy Analysis.

Being characterized by the self-adaption and high accuracy, the deep learning-based models have been widely applied in the 1D spectroscopy-related field. However, the "black-box" operation and "end-to-end" working style of the deep learning normally bring the low interpretability, where a reliable visualization is highly demanded. Although there are some well-developed visualization methods, such as Class Activation Mapping (CAM) and Gradient-weighted Class Activation Mapping (Grad-CAM), for the 2D image data, they cannot correctly reflect the weights of the model when being applied to the 1D spectral data, where the importance of position information is not considered. Here, aiming at the visualization of Convolutional Neural Network-based models toward the qualitative and quantitative analysis of 1D spectroscopy, we developed a novel visualization algorithm (1D Grad-CAM) to more accurately display the decision-making process of the CNN-based models. Different from the classical Grad-CAM, with the removal of the gradient averaging (GAP) and the ReLU operations, a significantly improved correlation between the gradient and the spectral location and a more comprehensive spectral feature capture were realized for 1D Grad-CAM. Furthermore, the introduction of difference (purity or linearity) and feature contribute in the CNN output in 1D Grad-CAM achieved a reliable evaluation of the qualitative accuracy and quantitative precision of CNN-based models. Facing the qualitative and adulteration quantitative analysis of vegetable oils by the combination of Raman spectroscopy and ResNet, the visualization by 1D Grad-CAM well reflected the origin of the high accuracy and precision brought by ResNet. In general, 1D Grad-CAM provides a clear vision about the judgment criterion of CNN and paves the way for CNN to a broad application in the field of 1D spectroscopy.

1665-P: Whole-Brain Activation Signatures of Weight-Lowering Drugs

Background: The development of effective anti-obesity therapeutics relies heavily on the ability to target CNS signaling mechanisms critically involved in the homeostatic control of body weight. To get insight into neurocircuits recruited by anti-obesity drug treatment, the present study aimed to determine whole-brain activation signatures of six different weight-lowering drug classes with documented efficacy in humans. Methods: Chow-fed mice received a single dose of lorcaserin (7 mg/kg, i.p.), rimonabant (10 mg/kg, i.p.), bromocriptine (D2 receptor agonist, 10 mg/kg, i.p.), sibutramine (10 mg/kg, p.o.), semaglutide (0.04 mg/kg, s.c.) or setmelanotide (4 mg/kg, s.c.). Brains were sampled two hours post-dosing, immunolabelled using c-Fos expression as a proxy for neuronal stimulation. Cleared whole-brains were imaged using light sheet fluorescence microscopy-deep learning pipeline enabling fully automated 3D mapping and quantitation of brain activation patterns at single-cell resolution. Results: The whole-brain analysis comprised 308 atlas-defined mouse brain areas. To enable fast and efficient data mining, an open-access web-based 3D imaging data viewer (NeuroPedia.dk) was developed. All weight-lowering drugs demonstrated brain-wide responses with notable similarities in c-Fos expression signatures. Overlapping c-Fos responses were detected in discrete homeostatic and non-homeostatic feeding centres located in the dorsal vagal complex and hypothalamus with concurrent activation of several limbic structures as well as the dopaminergic system. Conclusion: We pinpoint several overlapping brain activation signatures of various weight-lowering drugs. This shared feature suggests that weight-lowering drugs stimulate distinct homeostatic and non-homeostatic feeding centres. Future centrally acting anti-obesity compounds may be specifically designed to target key components of this neurocircuitry framework to provide more effective and sustained weight loss in obese patients. Disclosure L.S.Dalboege: Employee; Gubra aps. J.Hecksher-sørensen: Employee; Gubra ApS. M.Tozzi: None. C.G.Salinas: Employee; Gubra. T.Porsgaard: Employee; Gubra. D.D.Thorbek: Employee; Gubra APS. J.L.Skytte: Employee; Gubra ApS. U.Roostalu: Employee; Gubra. H.H.Hansen: None. J.Perens: Employee; Gubra ApS.

A data-driven modeling approach to quantify morphology effects on transport properties in nanostructured NMC particles

We present a data-driven modeling approach to quantify morphology effects on transport properties in nanostructured materials. Our approach is based on the combination of stochastic modeling of the 3D nanostructure and numerical modeling of effective transport properties, which is used to investigate process-structure–property relationships of hierarchically structured cathode materials for lithium-ion batteries. We focus on nanostructured LiNi1/3Mn1/3Co1/3O2 (NMC) particles, the nanoporous morphology of which has a crucial impact on their effective transport properties (i.e, effective ionic and electric conductivity) and thus on the performance of the cell. First, we develop a parametric stochastic model for the 3D morphology of the nanostructured NMC particles based on excursion sets of so-called χ2-fields. This model, which has only two parameters, is then fitted to FIB-SEM image data of the NMC particles manufactured with different calcination temperatures and different particle sizes. This way it is possible to generate digital twins of the NMC particles. In a second step, measured 3D image data and corresponding digital twins are used as input for the numerical simulation of effective transport properties. Based on the results obtained by these simulations, we can quantify process-structure–property relationships. Overall, we present a methodological framework that allows for an efficient optimization of the fabrication process of nanostructured NMC particles.

Two-Step CFAR-Based 3D Point Cloud Extraction Method for Circular Scanning Ground-Based Synthetic Aperture Radar

Ground-Based Synthetic Aperture Radar (GBSAR) has non-contact, all-weather, high resolution imaging and microdeformation sensing capabilities, which offers advantages in applications such as building structure monitoring and mine slope deformation retrieval. The Circular Scanning Ground-Based Synthetic Aperture Radar (CS-GBSAR) is one of its newest developed working mode, in which the radar rotates around an axis in a vertical plane. Such nonlinear observation geometry brings the unique advantage of three-dimensional (3D) imaging compared with traditional GBSAR modes. However, such nonlinear observation geometry causes strong sidelobes in SAR images, which makes it a difficult task to extract point cloud data. The Conventional Cell Averaging Constant False Alarm Rate (CA-CFAR) algorithm can extract 3D point cloud data layer-by-layer at different heights, which is time consuming and is easily influenced by strong sidelobes to obtain inaccurate results. To address these problems, this paper proposes a new two-step CFAR-based 3D point cloud extraction method for CS-GBSAR, which can extract accurate 3D point cloud data under the influence of strong sidelobes. It first utilizes maximum projection to obtain three-view images from 3D image data. Then, the first step CA-CFAR is applied to obtain the coarse masks of three-views. Then, the volume mask in the original 3D image is obtained via inverse projection. This can remove strong sidelobes outside the potential target region and obtain potential target area data by intersecting it with the SAR 3D image. Then, the second step CA-CFAR is applied to the potential target area data to obtain 3D point clouds. Finally, to further eliminate the residual strong sidelobes and output accurate 3D point clouds, the modified Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm is applied. The original DBSCAN method uses a spherical template to cluster. It covers more points, which is easily influenced by the strong sidelobe. Hence, the clustering results have more noise points. Meanwhile, modified DBSCAN clusters have a cylindrical template to accommodate the data’s features, which can reduce false clustering. The proposed method is validated via real data acquired by the North China University of Technology (NCUT)-developed CS-GBSAR system. The laser detection and ranging (LiDAR) data are used as the reference ground truth to demonstrate the method. The comparison experiment with conventional method shows that the proposed method can reduce 95.4% false clustered points and remove the strong sidelobes, which shows the better performance of the proposed method.

SELF-SUPERVISED LEARNING WITH RADIOLOGY REPORTS, A COMPARATIVE ANALYSIS OF STRATEGIES FOR LARGE VESSEL OCCLUSION AND BRAIN CTA IMAGES.

Scarcity of labels for medical images is a significant barrier for training representation learning approaches based on deep neural networks. This limitation is also present when using imaging data collected during routine clinical care stored in picture archiving communication systems (PACS), as these data rarely have attached the high-quality labels required for medical image computing tasks. However, medical images extracted from PACS are commonly coupled with descriptive radiology reports that contain significant information and could be leveraged to pre-train imaging models, which could serve as starting points for further task-specific fine-tuning. In this work, we perform a head-to-head comparison of three different self-supervised strategies to pre-train the same imaging model on 3D brain computed tomography angiogram (CTA) images, with large vessel occlusion (LVO) detection as the downstream task. These strategies evaluate two natural language processing (NLP) approaches, one to extract 100 explicit radiology concepts (Rad-SpatialNet) and the other to create general-purpose radiology reports embeddings (DistilBERT). In addition, we experiment with learning radiology concepts directly or by using a recent self-supervised learning approach (CLIP) that learns by ranking the distance between language and image vector embeddings. The LVO detection task was selected because it requires 3D imaging data, is clinically important, and requires the algorithm to learn outputs not explicitly stated in the radiology report. Pre-training was performed on an unlabeled dataset containing 1,542 3D CTA - reports pairs. The downstream task was tested on a labeled dataset of 402 subjects for LVO. We find that the pre-training performed with CLIP-based strategies improve the performance of the imaging model to detect LVO compared to a model trained only on the labeled data. The best performance was achieved by pre-training using the explicit radiology concepts and CLIP strategy.

Augmented Reality Support for Anterior Decompression and Fusion Using Floating Method for Cervical Ossification of the Posterior Longitudinal Ligament.

Anterior decompression and fusion (ADF) using the floating method for cervical ossification of the posterior longitudinal ligament (OPLL) is an ideal surgical technique, but it has a specific risk of insufficient decompression caused by the impingement of residual ossification. Augmented reality (AR) support is a novel technology that enables the superimposition of images onto the view of a surgical field. AR technology was applied to ADF for cervical OPLL to facilitate intraoperative anatomical orientation and OPLL identification. In total, 14 patients with cervical OPLL underwent ADF with microscopic AR support. The outline of the OPLL and the bilateral vertebral arteries was marked after intraoperative CT, and the reconstructed 3D image data were transferred and linked to the microscope. The AR microscopic view enabled us to visualize the ossification outline, which could not be seen directly in the surgical field, and allowed sufficient decompression of the ossification. Neurological disturbances were improved in all patients. No cases of serious complications, such as major intraoperative bleeding or reoperation due to the postoperative impingement of the floating OPLL, were registered. To our knowledge, this is the first report of the introduction of microscopic AR into ADF using the floating method for cervical OPLL with favorable clinical results.

Quantification of collective behaviour via causality analysis

Terms such as leader, mediator, and follower sound equal in the description of a pack of wolves, a street protest crowd, or a business team and have very similar meanings. This indicates the presence of some general law or structure that governs collective behaviour. To reveal this, we selected the most common parameter for most levels of the organisation—motion. A causality analysis of distance correlations was performed to obtain follow-up networks that show who follows whom and how strongly. These networks characterise an observed system in general and work as an automation bridge between the biological experiment and the broad field of network analysis. The proposed method was tested on 3D image data from a controlled experiment on a 6-member school of aquarium fish of Tiger Barb. The network patterns can be easily ethologically interpreted and agreed with expected behaviour.