Search

Myocardial infarction (MI) is one of the most prevalent cardiovascular diseases with associated clinical decision-making typically based on single-valued imaging biomarkers. However, such metrics only approximate the complex 3D structure and physiology of the heart and hence hinder a better understanding and prediction of MI outcomes. In this work, we investigate the utility of complete 3D cardiac shapes in the form of point clouds for an improved detection of MI events. To this end, we propose a fully automatic multi-step pipeline consisting of a 3D cardiac surface reconstruction step followed by a point cloud classification network. Our method utilizes recent advances in geometric deep learning on point clouds to enable direct and efficient multi-scale learning on high-resolution surface models of the cardiac anatomy. We evaluate our approach on 1068 UK Biobank subjects for the tasks of prevalent MI detection and incident MI prediction and find improvements of ~13% and ~5% respectively over clinical benchmarks. Furthermore, we analyze the role of each ventricle and cardiac phase for 3D shape-based MI detection and conduct a visual analysis of the morphological and physiological patterns typically associated with MI outcomes., Comment: Accepted at EMBC 2023

High-quality three-dimensional (3D) representations of cardiac anatomy and function are crucial for improving cardiac disease diagnosis beyond strictly volume-based biomarkers used in current clinical practice, as well as for the accurate simulation of cardiac electrophysiology and mechanics. However, current gold standard cardiac magnetic resonance imaging (MRI) protocols typically only acquire a set of 2D slices to approximate the true 3D morphology of the underlying heart. In this work, we propose a novel geometric deep learning method, the \emph{Mesh Deformation U-Net}, to reconstruct 3D cardiac surface meshes from 2D MRI slices as the key part of a fully automatic end-to-end pipeline. Its architecture combines spectral graph convolutions and mesh sampling operations in a hierarchical encoder-decoder structure to enable efficient multi-scale feature learning directly on mesh data. A targeted preprocessing step approximately fits a template mesh to the sparse MRI contours, before the Mesh Deformation U-Net corrects for motion-induced slice misalignment by simultaneously utilising information from multiple MRI views and the template-induced anatomical shape prior. We evaluate the Mesh Deformation U-Net on a large synthetic dataset of heart anatomies and outperform multiple benchmark approaches while achieving small reconstruction errors below the pixel size of the underlying image resolution for three different cardiac substructures. Furthermore, we apply the pre-trained Mesh Deformation U-Net as the key component of a 4-step reconstruction pipeline to cine magnetic resonance images of the UK Biobank and observe realistic heart reconstructions on both a local and global level. We calculate multiple widely used clinical metrics for the reconstructed meshes and obtain values in line with other large-scale population studies.

Human cardiac function is characterized by a complex interplay of mechanical deformation and electrophysiological conduction. Similar to the underlying cardiac anatomy, these interconnected physiological patterns vary considerably across the human population with important implications for the effectiveness of clinical decision-making and the accuracy of computerized heart models. While many previous works have investigated this variability separately for either cardiac anatomy or physiology, this work aims to combine both aspects in a single data-driven approach and capture their intricate interdependencies in a multi-domain setting. To this end, we propose a novel multi-domain Variational Autoencoder (VAE) network to capture combined Electrocardiogram (ECG) and Magnetic Resonance Imaging (MRI)-based 3D anatomy information in a single model. Each VAE branch is specifically designed to address the particular challenges of the respective input domain, enabling efficient encoding, reconstruction, and synthesis of multi-domain cardiac signals. Our method achieves high reconstruction accuracy on a United Kingdom Biobank dataset, with Chamfer Distances between reconstructed and input anatomies below the underlying image resolution and ECG reconstructions outperforming multiple single-domain benchmarks by a considerable margin. The proposed VAE is capable of generating realistic virtual populations of arbitrary size with good alignment in clinical metrics between the synthesized and gold standard anatomies and Maximum Mean Discrepancy (MMD) scores of generated ECGs below those of comparable single-domain approaches. Furthermore, we observe the latent space of our VAE to be highly interpretable with separate components encoding different aspects of anatomical and ECG variability. Finally, we demonstrate that the combined anatomy and ECG representation improves the performance in a cardiac disease classification task by 3.9% in terms of Area Under the Receiver Operating Characteristic (AUROC) curve over the best corresponding single-domain modeling approach.

Echolocation behavior, a navigation strategy based on acoustic signals, allows scientists to explore neural processing of behaviorally relevant stimuli. For the purpose of orientation, bats broadcast echolocation calls and extract spatial information from the echoes. Because bats control call emission and thus the availability of spatial information, the behavioral relevance of these signals is undiscussable. While most neurophysiological studies, conducted in the past, used synthesized acoustic stimuli that mimic portions of the echolocation signals, recent progress has been made to understand how naturalistic echolocation signals are encoded in the bat brain. Here, we review how does stimulus history affect neural processing, how spatial information from multiple objects and how echolocation signals embedded in a naturalistic, noisy environment are processed in the bat brain. We end our review by discussing the huge potential that state-of-the-art recording techniques provide to gain a more complete picture on the neuroethology of echolocation behavior.

The segmentation of arteries in invasive coronary angiography is necessary to build quantitative models and eventually improve the diagnosis of cardiovascular diseases. Standard segmentation algorithms suffer due to the lack of fully annotated datasets and tend to return disconnected vessels. Thus, we explore a semi-supervised segmentation framework to address these issues. Specifically, we use a student model and a teacher model as the main framework with Nested U-Nets (UNet++) as their backbones. The student model learns by minimizing a segmentation loss between the output and the ground truth, and a consistency loss guided by the uncertainty information. Additionally, a special loss function based on elastic interaction is used to improve the connectivity of arterial branches. We demonstrate the effectiveness of our proposed techniques over 42 labeled and 60 unlabeled samples and find relative improvement of 5.59% for Dice score and 69.99% for Betti number compared to a U-Net.

Understanding population-wide variability of the human heart is crucial to detect abnormalities and improve the assessment of both cardiac anatomy and function. While many computational modeling approaches have been developed to capture this variability separately for either cardiac anatomy or physiology, their complex interconnections have rarely been explored together. In this work, we propose a novel multi-modal variational autoencoder (VAE) capable of processing combined physiology and bitemporal anatomy information in the form of electrocardiograms (ECG) and 3D biventricular point clouds. Our method achieves high reconstruction accuracy on a UK Biobank dataset with Chamfer distances between predicted and input anatomies below the underlying image resolution and the ECG reconstructions outperforming a state-of-the-art benchmark approach specialized in ECG generation. We also evaluate its generative ability and find comparable populations of generated and gold standard anatomies, ECGs, and combined anatomy-ECG data in terms of common clinical metrics and maximum mean discrepancies.

Animals must extract relevant sensory information out of a multitude of non-informative and sometimes interfering stimuli. For orientation, bats rely on broadcasted calls and they must assign each echo to the corresponding call. When bats orient in acoustically enriched environments, call-echo assignment becomes challenging due to signal interference. Bats often adapt echolocation parameters which potentially improves signal extraction. However, they also adjust echolocation parameters with respect to target distance. To characterize adaptations that are exclusively elicited to minimize signal interference, we tested the effect of acoustic playback on the echolocation behavior of the fruit-eating bat, Carollia perspicillata . Hereby, distance-dependent changes were considered by swinging bats in a pendulum and directly measuring the object distance. Acoustic playback evoked different call adjustments in parameters such as bandwidth, peak-frequency, duration and call level. These adaptations were highly dynamic and could vary across individuals, days, trials, and even within trials. Our results demonstrate that bats do not only change one echolocation parameter when orienting in acoustically enriched environments. They rather have a tool-kit of different behavioral adaptations to cope with interfering acoustic stimuli. By dynamically switching between different adaptations, bats can maximize the extraction of their biosonar signals from the background.

Summary statement When echolocating under demanding conditions e.g. noisy, narrow space, or cluttered environments, frugivorous bats adapt their call pattern by increasing the call rate within biosonar groups. Abstract For orientation, echolocating bats emit biosonar calls and use echoes arising from call reflections. They often pattern their calls into groups which increases the rate of sensory feedback over time. Insectivorous bats emit call groups at a higher rate when orienting in cluttered compared to uncluttered environments. Frugivorous bats increase the rate of call group emission when they echolocate in noisy environments. Here, calls emitted by conspecifics potentially interfere with the bat’s biosonar signals and complicate the echolocation behavior. To minimize the information loss followed by signal interference, bats may profit from a temporally increased sensory acquisition rate, as it is the case for the call groups. In frugivorous bats, it remains unclear if call group emission represents an exclusive adaptation to avoid interference by signals from other bats or if it represents an adaptation that allows to orient under demanding environmental conditions. Here, we compared the emission pattern of the frugivorous bat Carollia perspicillata when the bats were flying in noisy versus silent, narrow versus wide or cluttered versus non-cluttered corridors. According to our results, the bats emitted larger call groups and they increased the call rate within the call groups when navigating in narrow, cluttered, or noisy environments. Thus, call group emission represents an adaptive behavior when the bats orient in complex environments.

Autonomous robots in unstructured and dynamically changing retail environments have to master complex perception, knowledgeprocessing, and manipulation tasks. To enable them to act competently, we propose a framework based on three core components: (o) a knowledge-enabled perception system, capable of combining diverse information sources to cope with occlusions and stacked objects with a variety of textures and shapes, (o) knowledge processing methods produce strategies for tidying up supermarket racks, and (o) the necessary manipulation skills in confined spaces to arrange objects in semi-accessible rack shelves. We demonstrate our framework in an simulated environment as well as on a real shopping rack using a PR2 robot. Typical supermarket products are detected and rearranged in the retail rack, tidying up what was found to be misplaced items., Comment: published in the proceedings of AAMAS 2016 as an extended abstract