Low-resolution Maps Research Articles

Degradation-Guided Multi-Modal Fusion Network for Depth Map Super-Resolution

Depth map super-resolution (DSR) is a technique aimed at restoring high-resolution (HR) depth maps from low-resolution (LR) depth maps. In this process, color images are commonly used as guidance to enhance the restoration procedure. However, the intricate degradation of LR depth poses a challenge, and previous image-guided DSR approaches, which implicitly model the degradation in the spatial domain, often fall short of producing satisfactory results. To address this challenge, we propose a novel approach called the Degradation-Guided Multi-modal Fusion Network (DMFNet). DMFNet explicitly characterizes the degradation and incorporates multi-modal fusion in both spatial and frequency domains to improve the depth quality. Specifically, we first introduce the deep degradation regularization loss function, which enables the model to learn the explicit degradation from the LR depth maps. Simultaneously, DMFNet converts the color images and depth maps into spectrum representations to provide comprehensive multi-domain guidance. Consequently, we present the multi-modal fusion block to restore the depth maps by leveraging both the RGB-D spectrum representations and the depth degradation. Extensive experiments demonstrate that DMFNet achieves state-of-the-art (SoTA) performance on four benchmarks, namely the NYU-v2, Middlebury, Lu, and RGB-D-D datasets.

Discovering prominent differences in structural and functional connectomes using a multinomial stochastic block model

ABSTRACT Understanding the differences between functional and structural human brain connectivity has been a focus of an extensive amount of neuroscience research. We employ a novel approach using the multinomial stochastic block model (MSBM) to explicitly extract components that characterize prominent differences across graphs. We analyze structural and functional connectomes derived from high-resolution diffusion-weighted MRI and fMRI scans of 250 Human Connectome Project subjects, analyzed at group connectivity level across 50 subjects. The inferred brain partitions revealed consistent, spatially homogeneous clustering patterns across inferred resolutions demonstrating the MSBM’s reliability in identifying brain areas with prominent structure-function differences. Prominent differences in low-resolution brain maps (K = {3, 4} clusters) were attributed to weak functional connectivity in the bilateral anterior temporal lobes, while higher resolution results (K ≥ 25) revealed stronger interhemispheric functional than structural connectivity. Our findings emphasize significant differences in high-resolution functional and structural connectomes, revealing challenges in extracting meaningful connectivity measurements from both modalities, including tracking fibers through the corpus callosum and attenuated functional connectivity in anterior temporal lobe fMRI data, which we attribute to increased noise levels. The MSBM emerges as a valuable tool for understanding differences across graphs, with potential future applications and avenues beyond the current focus on characterizing modality-specific distinctions in connectomics data.

Lightweight medical image segmentation network with multi-scale feature-guided fusion

In the field of computer-aided medical diagnosis, it is crucial to adapt medical image segmentation to limited computing resources. There is tremendous value in developing accurate, real-time vision processing models that require minimal computational resources. When building lightweight models, there is always a trade-off between computational cost and segmentation performance. Performance often suffers when applying models to meet resource-constrained scenarios characterized by computation, memory, or storage constraints. This remains an ongoing challenge. This paper proposes a lightweight network for medical image segmentation. It introduces a lightweight transformer, proposes a simplified core feature extraction network to capture more semantic information, and builds a multi-scale feature interaction guidance framework. The fusion module embedded in this framework is designed to address spatial and channel complexities. Through the multi-scale feature interaction guidance framework and fusion module, the proposed network achieves robust semantic information extraction from low-resolution feature maps and rich spatial information retrieval from high-resolution feature maps while ensuring segmentation performance. This significantly reduces the parameter requirements for maintaining deep features within the network, resulting in faster inference and reduced floating-point operations (FLOPs) and parameter counts. Experimental results on ISIC2017 and ISIC2018 datasets confirm the effectiveness of the proposed network in medical image segmentation tasks. For instance, on the ISIC2017 dataset, the proposed network achieved a segmentation accuracy of 82.33 % mIoU, and a speed of 71.26 FPS on 256 × 256 images using a GeForce GTX 3090 GPU. Furthermore, the proposed network is tremendously lightweight, containing only 0.524M parameters. The corresponding source codes are available at https://github.com/CurbUni/LMIS-lightweight-network.

Investigating the Use of Traveltime and Reflection Tomography for Deep Learning-Based Sound-Speed Estimation in Ultrasound Computed Tomography.

Ultrasound computed tomography (USCT) quantifies acoustic tissue properties such as the speed-of-sound (SOS). Although full-waveform inversion (FWI) is an effective method for accurate SOS reconstruction, it can be computationally challenging for large-scale problems. Deep learning-based image-to-image learned reconstruction (IILR) methods can offer computationally efficient alternatives. This study investigates the impact of the chosen input modalities on IILR methods for high-resolution SOS reconstruction in USCT. The selected modalities are traveltime tomography (TT) and reflection tomography (RT), which produce a low-resolution SOS map and a reflectivity map, respectively. These modalities have been chosen for their lower computational cost relative to FWI and their capacity to provide complementary information: TT offers a direct SOS measure, while RT reveals tissue boundary information. Systematic analyses were facilitated by employing a virtual USCT imaging system with anatomically realistic numerical breast phantoms. Within this testbed, a supervised convolutional neural network (CNN) was trained to map dual-channel (TT and RT images) to a high-resolution SOS map. Single-input CNNs were trained separately using inputs from each modality alone (TT or RT) for comparison. The accuracy of the methods was systematically assessed using normalized root mean squared error (NRMSE), structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR). For tumor detection performance, receiver operating characteristic analysis was employed. The dual-channel IILR method was also tested on clinical human breast data. Ensemble average of the NRMSE, SSIM, and PSNR evaluated on this clinical dataset were 0.2355, 0.8845, and 28.33 dB, respectively.

CMB-PAInT: An inpainting tool for the cosmic microwave background

The presence of astrophysical emissions in microwave observations forces us to perform component separation to extract the Cosmic Microwave Background (CMB) signal. However, even in the most optimistic cases, there are still strongly contaminated regions, such as the Galactic plane or those with emission from extragalactic point sources, which require the use of a mask. Since many CMB analyses, especially the ones working in harmonic space, need the whole sky map, it is crucial to develop a reliable inpainting algorithm that replaces the values of the excluded pixels by others statistically compatible with the rest of the sky. This is especially important when working with Q and U sky maps in order to obtain E- and B-mode maps which are free from E-to-B leakage. In this work we study a method based on Gaussian Constrained Realizations (GCR), that can deal with both intensity and polarization. Several tests have been performed to asses the validation of the method, including the study of the one-dimensional probability distribution function (1-PDF), E- and B-mode map reconstruction, and power spectra estimation. We have considered two scenarios for the input simulation: one case with only CMB signal and a second one including also Planck PR4 semi-realistic noise. Even if we are limited to low resolution maps, Nside = 64 if T, Q and U are considered, we believe that this is a useful approach to be applied to future missions such as LiteBIRD, where the target are the largest scales.

CONCERTO at APEX On-sky performance in continuum

Context. CarbON CII line in post-rEionisation and ReionisaTiOn epoch (CONCERTO) instrument is a low-resolution mapping spectrometer based on lumped element kinetic inductance detector (LEKIDs) technology, operating at 130-310 GHz. It was installed on the 12-metre APEX telescope in Chile in April 2021 and was in operation until May 2023. CONCERTO’s main goals were the observation of [CII]-emission line fluctuations at high redshift and of the Sunyaev–Zel’dovich (SZ) signal from galaxy clusters. Aims. We present the data processing algorithms and the performance of CONCERTO in continuum by analysing the data from the commissioning and scientific observations. Methods. We developed a standard data processing pipeline to proceed from the raw data to continuum maps. Using a large dataset of calibrators (Uranus, Mars, and quasars) acquired in 2021 and 2022 at the APEX telescope across a wide range of atmospheric conditions, we measured the CONCERTO continuum performance and tested its stability against observing conditions. Further, using observations on the COSMOS field and observations targeting a distant sub-millimetre galaxy in the UDS field, we assessed the robustness of the CONCERTO performance on faint sources and compared our measurements with expectations. Results. The beam pattern is characterised by an effective full width at half maximum (FWHM) of 31.9 ± 0.6″ and 34.4 ± 1.0″ for high-frequency (HF) and low-frequency (LF) bands, respectively. The main beam is slightly elongated with a mean eccentricity of 0.46. Two error beams of ~65″ and ~130″ are characterised, allowing us to estimate a main beam efficiency of ~0.52. The field of view is accurately reconstructed and presents coherent distortions between the HF and LF arrays. LEKID parameters were robustly determined for 80% of the read tones. Cross-talks between LEKIDs are the first cause of flagging, followed by an excess of eccentricity for ~10% of the LEKIDs, all located in a given region of the field of view. Of the 44 scans of Uranus selected for the absolute photometric calibration, 72.5% and 78.2% of the LEKIDs were selected as valid detectors with a probability >70%. By comparing the Uranus measurements with a model, we obtain calibration factors of 19.5±0.6 Hz Jy−1 and 25.6±0.9 Hz Jy−1 for HF and LF, respectively. The point-source continuum measurement uncertainties are 3.0% and 3.4% for the HF and LF bands, ignoring the uncertainty in the model (which is <2%). This demonstrates the accuracy of the methods we deployed to process the data. Finally, the RMS of CONCERTO maps is verified to evolve as proportional to the inverse square root of the integration time. The measured noise-equivalent flux densities (NEFDs) for HF and LF are 115±2 mJy beam−1 s1/2 and 95±1 mJy beam−1 s1/2, respectively, obtained using CONCERTO data on the COSMOS field for a mean precipitable water vapour (pwv) and elevation of 0.81 mm and 55.7 deg. Conclusions. CONCERTO has unique capabilities in fast dual-band spectral mapping at ~30 arcsec resolution and with a ~18.5 arcmin instantaneous field of view. CONCERTO’s performance in continuum is perfectly in line with expectations.

Comparison of super-resolution deep learning models for flow imaging

The primary goal of this study is to introduce deep learning (DL) methods as a cost-effective alternative to the computationally intensive Direct Numerical Simulation (DNS) simulations. We show that one can obtain a parametric field from a low-resolution input and map it to a fine grid output, significantly reducing the computational burden. We assess five super-resolution models for up-scaling low-resolution flow data into fine-grid numerical simulations’ output for accuracy and efficiency. The proposed architectures employ convolutional neural networks interconnected in encoder/decoder branches. We investigate these models using turbulent velocity fields inside a suddenly expanded channel characterized by complex features, including turbulence, instabilities, asymmetries, separation, and reattachment. Our results reveal that an encoder/decoder model with residual connections delivers the fastest results, a U-Net-based model with skip connections excels at producing sharper edges in regions prone to blurring, while deeper models incorporating maximum and average pooling layers show superior performance in reconstructing velocity profiles. These findings significantly contribute to our understanding of the potential of deep learning in fluid mechanics. The models presented in this study are trained and validated on standard computer hardware and can be easily adapted to other problems. The findings are promising for discovering and analyzing flow physics, highlighting the potential for DL techniques to improve the accuracy of the available fluid mechanics computational tools.

Signaling by a bacterial phytochrome histidine kinase involves a conformational cascade reorganizing the dimeric photoreceptor

Phytochromes (Phys) are a divergent cohort of bili-proteins that detect light through reversible interconversion between dark-adapted Pr and photoactivated Pfr states. While our understandings of downstream events are emerging, it remains unclear how Phys translate light into an interpretable conformational signal. Here, we present models of both states for a dimeric Phy with histidine kinase (HK) activity from the proteobacterium Pseudomonas syringae, which were built from high-resolution cryo-EM maps (2.8–3.4-Å) of the photosensory module (PSM) and its following signaling (S) helix together with lower resolution maps for the downstream output region augmented by RoseTTAFold and AlphaFold structural predictions. The head-to-head models reveal the PSM and its photointerconversion mechanism with strong clarity, while the HK region is interpretable but relatively mobile. Pr/Pfr comparisons show that bilin phototransformation alters PSM architecture culminating in a scissoring motion of the paired S-helices linking the PSMs to the HK bidomains that ends in reorientation of the paired catalytic ATPase modules relative to the phosphoacceptor histidines. This action apparently primes autophosphorylation enroute to phosphotransfer to the cognate DNA-binding response regulator AlgB which drives quorum-sensing behavior through transient association with the photoreceptor. Collectively, these models illustrate how light absorption conformationally translates into accelerated signaling by Phy-type kinases.

Dual protection by Bcp1 and Rkm1 ensures incorporation of uL14 into pre-60S ribosomal subunits.

Eukaryotic ribosomal proteins contain extended regions essential for translation coordination. Dedicated chaperones stabilize the associated ribosomal proteins. We identified Bcp1 as the chaperone of uL14 in Saccharomyces cerevisiae. Rkm1, the lysine methyltransferase of uL14, forms a ternary complex with Bcp1 and uL14 to protect uL14. Rkm1 is transported with uL14 by importins to the nucleus, and Bcp1 disassembles Rkm1 and importin from uL14 simultaneously in a RanGTP-independent manner. Molecular docking, guided by crosslinking mass spectrometry and validated by a low-resolution cryo-EM map, reveals the correlation between Bcp1, Rkm1, and uL14, demonstrating the protection model. In addition, the ternary complex also serves as a surveillance point, whereas incorrect uL14 is retained on Rkm1 and prevented from loading to the pre-60S ribosomal subunits. This study reveals the molecular mechanism of how uL14 is protected and quality checked by serial steps to ensure its safe delivery from the cytoplasm until its incorporation into the 60S ribosomal subunit.

Lightning fast video anomaly detection via multi-scale adversarial distillation

We propose a very fast frame-level model for anomaly detection in video, which learns to detect anomalies by distilling knowledge from multiple highly accurate object-level teacher models. To improve the fidelity of our student, we distill the low-resolution anomaly maps of the teachers by jointly applying standard and adversarial distillation, introducing an adversarial discriminator for each teacher to distinguish between target and generated anomaly maps. We conduct experiments on three benchmarks (Avenue, ShanghaiTech, UCSD Ped2), showing that our method is over 7 times faster than the fastest competing method, and between 28 and 62 times faster than object-centric models, while obtaining comparable results to recent methods. Our evaluation also indicates that our model achieves the best trade-off between speed and accuracy, due to its previously unheard-of speed of 1480 FPS. In addition, we carry out a comprehensive ablation study to justify our architectural design choices. Our code is freely available at: https://github.com/ristea/fast-aed.

Digging into Depth and Color Spaces: A Mapping Constraint Network for Depth Super-Resolution

Scene depth super-resolution (DSR) poses an inherently ill-posed problem due to the extremely large space of one-to-many mapping functions from a given low-resolution (LR) depth map, which possesses limited depth information, to multiple plausible high-resolution (HR) depth maps. This characteristic renders the task highly challenging, as identifying an optimal solution becomes significantly intricate amidst this multitude of potential mappings. While simplistic constraints have been proposed to address the DSR task, the relationship between LR and HR depth maps and the color image has not been thoroughly investigated. In this paper, we introduce a novel mapping constraint network (MCNet) that incorporates additional constraints derived from both LR depth maps and color images. This integration aims to optimize the space of mapping functions and enhance the performance of DSR. Specifically, alongside the primary DSR network (DSRNet) dedicated to learning LR-to-HR mapping, we have developed an auxiliary degradation network (ADNet) that operates in reverse, generating the LR depth map from the reconstructed HR depth map to obtain depth features in LR space. To enhance the learning process of DSRNet in LR-to-HR mapping, we introduce two mapping constraints in LR space: 1) the cycle-consistent constraint, which offers additional supervision by establishing a closed loop between LR-to-HR and HR-to-LR mappings, and 2) the region-level contrastive constraint, aimed at reinforcing region-specific HR representations by explicitly modeling the consistency between LR and HR spaces. To leverage the color image effectively, we introduce a feature screening module (FSM) to adaptively fuse color features at different layers, which can simultaneously maintain strong structural context and suppress texture distraction through subspace generation and image projection. Comprehensive experimental results across synthetic and real-world benchmark datasets unequivocally demonstrate the superiority of our proposed method over state-of-the-art DSR methods. Our MCNet achieves an average MAD reduction of 3.7% and 7.5% over state-of-the-art DSR method for \(\times\) 8 and \(\times\) 16 cases on Milddleburry dataset, respectively, without incurring additional costs during inference.

Light field image super-resolution using a Content-Aware Spatial–Angular Interaction network

Light field images record a series of viewpoints of a scene and thus have many attractive applications. However, the trade-off between angular and spatial resolution in the imaging process makes light field image super-resolution necessary. In this paper, we propose a Content-Aware Spatial–Angular Interaction (dubbed CASAI) network for light field image super-resolution. The gradient branch of CASAI makes full use of the context information of the low-resolution gradient map and the multi-level features of the super-resolution branch to generate the high-resolution gradient map, which enables the awareness of structure, texture and detail, and provides effective prior knowledge for the super-resolution process. The super-resolution branch of CASAI generates high-quality super-resolution images by making full use of intra-view (i.e., spatial) and inter-view (i.e., angular) information through a spatial–angular adaptive interaction block and using the high-resolution gradient prior as guidance. The spatial–angular adaptive interaction block enables the awareness of the different importance of spatial features and angular features, so as to better integrate intra-view and inter-view information to improve the performance of light field image super-resolution. The experimental results indicate that when the upsampling factor is 4, our method outperforms our baseline (LF-InterNet) with an average PSNR increase of 0.48 dB, while also achieving the best SSIM. Visualization results demonstrate the advantage of our method in simultaneously generating natural SR images and restoring structures.

An enhanced approach for few-shot segmentation via smooth downsampling mask and label smoothing loss

Few-shot semantic segmentation aims to segment new categories with only a small number of annotated images. Previous methods mainly focused on exploiting the pixel-level correlation between the support image and the query image, combined with attention-based methods, resulting in significant advancements. In this paper, we introduce a new perspective to enhance few-shot segmentation. We identify that utilizing the bilinear interpolation method to downsample the mask leads to the loss of fine-grained information from the target features. To address this issue, we propose a Smooth Downsampling Mask (SDM) method. The SDM method is designed to retain more effective target semantic features by employing a cascaded downsampling approach with a smooth kernel for mask processing. Additionally, we propose a label smoothing loss to further enhance the performance, which provides direct guidance for low-resolution feature map optimization. Both methods can be used as plug-and-play modules for existing methods. Notably, our proposed method does not involve additional learnable parameters and is computationally efficient, thus achieving painless gains. To validate the effectiveness of our method, we take three publicly available models as baselines and conduct extensive experiments on three public benchmarks PASCAL-5i, COCO-20i and FSS-1000, and achieve considerable improvement.

USV Path Planning in a Hybrid Map Using a Genetic Algorithm with a Feedback Mechanism

Unmanned surface vehicles (USVs) often operate in real-world environments with long voyage distances and complex routes. The use of a single-grid map model presents challenges, such as the high computational costs for high-resolution maps and loss of environmental information for low-resolution maps. This article proposes an environmental modeling method using a hybrid map that combines topology units and grids. The approach involves calibrating key nodes based on the watershed skeleton line, constructing a topology map using these nodes, decomposing the original map into unit maps, converting each unit map into a grid map, and creating a hybrid map environment model that comprises topology maps, unit map sets, and grid map sets. Then, the article introduces an improved genetic algorithm, called Genetic Algorithm with Feedback (FGA), to address path planning in hybrid maps. Experimental results demonstrate that FGA has better computational efficiency than other algorithms in similar experimental environments. In hybrid maps, path planning with FGA reduces the path lengths and time consumption, and the paths are more logical, smooth, and continuous. These findings contribute to enhancing the quality of path planning and the practical value of USVs.

Attentional decoder networks for chest X-ray image recognition on high-resolution features

Background and objective:This paper introduces an encoder–decoder-based attentional decoder network to recognize small-size lesions in chest X-ray images. In the encoder-only network, small-size lesions disappear during the down-sampling steps or are indistinguishable in the low-resolution feature maps. To address these issues, the proposed network processes images in the encoder–decoder architecture similar to U-Net families and classifies lesions by globally pooling high-resolution feature maps. However, two challenging obstacles prohibit U-Net families from being extended to classification: (1) the up-sampling procedure consumes considerable resources, and (2) there needs to be an effective pooling approach for the high-resolution feature maps. Methods:Therefore, the proposed network employs a lightweight attentional decoder and harmonic magnitude transform. The attentional decoder up-samples the given features with the low-resolution features as the key and value while the high-resolution features as the query. Since multi-scaled features interact, up-sampled features embody global context at a high resolution, maintaining pathological locality. In addition, harmonic magnitude transform is devised for pooling high-resolution feature maps in the frequency domain. We borrow the shift theorem of the Fourier transform to preserve the translation invariant property and further reduce the parameters of the pooling layer by an efficient embedding strategy. Results:The proposed network achieves state-of-the-art classification performance on the three public chest X-ray datasets, such as NIH, CheXpert, and MIMIC-CXR. Conclusions:In conclusion, the proposed efficient encoder–decoder network recognizes small-size lesions well in chest X-ray images by efficiently up-sampling feature maps through an attentional decoder and processing high-resolution feature maps with harmonic magnitude transform. We open-source our implementation at https://github.com/Lab-LVM/ADNet.

High-resolution 3T to 7T ADC map synthesis with a hybrid CNN-transformer model.

7 Tesla (7T) apparent diffusion coefficient (ADC) maps derived from diffusion-weighted imaging (DWI) demonstrate improved image quality and spatial resolution over 3 Tesla (3T) ADC maps. However, 7T magnetic resonance imaging (MRI) currently suffers from limited clinical unavailability, higher cost, and increased susceptibility to artifacts. To address these issues, we propose a hybrid CNN-transformer model to synthesize high-resolution 7T ADC maps from multimodal 3T MRI. The Vision CNN-Transformer (VCT), composed of both Vision Transformer (ViT) blocks and convolutional layers, is proposed to produce high-resolution synthetic 7T ADC maps from 3T ADC maps and 3T T1-weighted (T1w) MRI. ViT blocks enabled global image context while convolutional layers efficiently captured fine detail. The VCT model was validated on the publicly available Human Connectome Project Young Adult dataset, comprising 3T T1w, 3T DWI, and 7T DWI brain scans. The Diffusion Imaging in Python library was used to compute ADC maps from the DWI scans. A total of 171 patient cases were randomly divided into 130 training cases, 20 validation cases, and 21 test cases. The synthetic ADC maps were evaluated by comparing their similarity to the ground truth volumes with the following metrics: peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and mean squared error (MSE). In addition, RESULTS: The results are as follows: PSNR: 27.0±0.9dB, SSIM: 0.945±0.010, and MSE: 2.0E-3±0.4E-3. Both qualitative and quantitative results demonstrate that VCT performs favorably against other state-of-the-art methods. We have introduced various efficiency improvements, including the implementation of flash attention and training on 176×208 resolution images. These enhancements have resulted in the reduction of parameters and training time per epoch by 50% in comparison to ResViT. Specifically, the training time per epoch has been shortened from 7.67 min to 3.86 min. We propose a novel method to predict high-resolution 7T ADC maps from low-resolution 3T ADC maps and T1w MRI. Our predicted images demonstrate better spatial resolution and contrast compared to 3T MRI and prediction results made by ResViT and pix2pix. These high-quality synthetic 7T MR images could be beneficial for disease diagnosis and intervention, producing higher resolution and conformal contours, and as an intermediate step in generating synthetic CT for radiation therapy, especially when 7T MRI scanners are unavailable.

Structural basis for dimerization of a paramyxovirus polymerase complex

The transcription and replication processes of non-segmented, negative-strand RNA viruses (nsNSVs) are catalyzed by a multi-functional polymerase complex composed of the large protein (L) and a cofactor protein, such as phosphoprotein (P). Previous studies have shown that the nsNSV polymerase can adopt a dimeric form, however, the structure of the dimer and its function are poorly understood. Here we determine a 2.7 Å cryo-EM structure of human parainfluenza virus type 3 (hPIV3) L–P complex with the connector domain (CD′) of a second L built, while reconstruction of the rest of the second L–P obtains a low-resolution map of the ring-like L core region. This study reveals detailed atomic features of nsNSV polymerase active site and distinct conformation of hPIV3 L with a unique β-strand latch. Furthermore, we report the structural basis of L–L dimerization, with CD′ located at the putative template entry of the adjoining L. Disruption of the L–L interface causes a defect in RNA replication that can be overcome by complementation, demonstrating that L dimerization is necessary for hPIV3 genome replication. These findings provide further insight into how nsNSV polymerases perform their functions, and suggest a new avenue for rational drug design.

Optimal 3D angular sampling with applications to cryo-EM problems

The goal of cryo-EM experiments in the biological sciences is to determine the atomic structure of a molecule and deduce insights into its functions and mechanisms. Despite improvements in instrumentation for data collection and new software algorithms, in most cases, individual atoms are not resolved. Model building of proteins, nucleic acids, or molecules in general, is feasible from the experimentally determined density maps at resolutions up to the range of 3–4 Angstroms. For lower-resolution maps or parts of maps, fitting smaller structures obtained by modelling or experimental techniques with higher resolution is a way to resolve the issue. In practice, we have an atomic structure, generate its density map at a given resolution, and translate/rotate the map within a region of interest in the experimental map, computing a measure-of-fit score with the corresponding areas of the experimental map. This procedure is computationally intensive since we work in 6D space. An optimal ordered list of rotations will reduce the angular error and help to find the best-fitting positions faster for a coarse global search or a local refinement. It can be used for adaptive approaches to stop fitting algorithms earlier once the desired accuracy has been achieved. We demonstrate how the performance of some fitting algorithms can be improved by grouping sets of rotations. We present an approach to generate more efficient 3D angular sampling, and provide the computer code to generate lists of optimal orientations for single and grouped rotations and the lists themselves.

Efficient Small-Object Detection in Underwater Images Using the Enhanced YOLOv8 Network

Underwater object detection plays a significant role in marine ecosystem research and marine species conservation. The improvement of related technologies holds practical significance. Although existing object-detection algorithms have achieved an excellent performance on land, they are not satisfactory in underwater scenarios due to two limitations: the underwater objects are often small, densely distributed, and prone to occlusion characteristics, and underwater embedded devices have limited storage and computational capabilities. In this paper, we propose a high-precision, lightweight underwater detector specifically optimizing for underwater scenarios based on the You Only Look Once Version 8 (YOLOv8) model. Firstly, we replace the Darknet-53 backbone of YOLOv8s with FasterNet-T0, reducing model parameters by 22.52%, FLOPS by 23.59%, and model size by 22.73%, achieving model lightweighting. Secondly, we add a Prediction Head for Small Objects, increase the number of channels for high-resolution feature map detection heads, and decrease the number of channels for low-resolution feature map detection heads. This results in a 1.2% improvement in small-object detection accuracy, while the remaining model parameters and memory consumption are nearly unchanged. Thirdly, we use Deformable ConvNets and Coordinate Attention in the neck part to enhance the accuracy in the detection of irregularly shaped and densely occluded small targets. This is achieved by learning convolution offsets from feature maps and emphasizing the regions of interest (RoIs). Our method achieves 52.12% AP on the underwater dataset UTDAC2020, with only 8.5 M parameters, 25.5 B FLOPS, and 17 MB model size. It surpasses the performance of large model YOLOv8l, at 51.69% AP, with 43.6 M parameters, 164.8 B FLOPS, and 84 MB model size. Furthermore, by increasing the input image resolution to 1280 × 1280 pixels, our model achieves 53.18% AP, making it the state-of-the-art (SOTA) model for the UTDAC2020 underwater dataset. Additionally, we achieve 84.4% mAP on the Pascal VOC dataset, with a substantial reduction in model parameters compared to previous, well-established detectors. The experimental results demonstrate that our proposed lightweight method retains effectiveness on underwater datasets and can be generalized to common datasets.

Native myocardial T1 mapping: influence of spatial resolution on quantitative results and reproducibility.

Myocardial mapping techniques can be used to quantitatively assess alterations in myocardial tissue properties. This study aims to evaluate the influence of spatial resolution on quantitative results and reproducibility of native myocardial T1 mapping in cardiac magnetic resonance imaging (MRI). In this cross-sectional study with prospective data collection between October 2019 and February 2020, 50 healthy adults underwent two identical cardiac MRI examinations in the radiology department on the same day. T1 mapping was performed using a MOLLI 5(3)3 sequence with higher (1.4 mm × 1.4 mm) and lower (1.9 mm × 1.9 mm) in-plane spatial resolution. Global quantitative results of T1 mapping were compared between high-resolution and low-resolution acquisitions using paired t-test. Intra-class correlation coefficient (ICC) and Bland-Altman statistics (absolute and percentage differences as means ± SD) were used for assessing test-retest reproducibility. There was no significant difference between global quantitative results acquired with high vs. low-resolution T1 mapping. The reproducibility of global T1 values was good for high-resolution (ICC: 0.88) and excellent for low-resolution T1 mapping (ICC: 0.95, P=0.003). In subgroup analyses, inferior test-retest reproducibility was observed for high spatial resolution in women compared to low spatial resolution (ICC: 0.71 vs. 0.91, P=0.001) and heart rates >77 bpm (ICC: 0.53 vs. 0.88, P=0.004). Apical segments had higher T1 values and variability compared to other segments. Regional T1 values for basal (ICC: 0.81 vs. 0.89, P=0.023) and apical slices (ICC: 0.86 vs. 0.92, P=0.024) showed significantly higher reproducibility in low-resolution compared to high-resolution acquisitions but without differences for midventricular slice (ICC: 0.91 vs. 0.92, P=0.402). Based on our data, we recommend a spatial resolution on the order of 1.9 mm × 1.9 mm for native myocardial T1 mapping using a MOLLI 5(3)3 sequence at 1.5 T particularly in individuals with higher heart rates and women.