Dense Field Research Articles

Bacterial Diversity in Deep-Sea Sediment of West Pacific Nodule Province

Dense polymetallic nodule fields are found in different areas of the Pacific and Indian Oceans. However, limited knowledge exists about microbial diversity, processes and functions in deep-sea polymetallic nodule sediments. This study investigated microbial diversity, composition and function in sediments from various locations and depths in a western Pacific polymetallic nodule province. Sediment cores were collected, DNA extracted, and the V3–V4 regions of the 16S rRNA gene were sequenced using Illumina MiSeq. The test results show that the abundance and diversity of microbial communities in sediments from different sites vary significantly. The dominant microbial communities at the family level in the three sediment samples were all mainly Marinobacteraceae and Alcanivoracaceae. Sediment samples from core 1 had similar microbial structures and microbial community functions. Surface sediment had higher species richness, diversity and evenness than the middle layer. The dominant phylum at different depths was consistent. There was significant spatial heterogeneity in the microbial community within sediments from polymetallic nodule regions. This study expands on our knowledge of the spatial and vertical distribution of microbial community diversity beneath polymetallic nodules in deep-sea settings.

Quantifying Uncertainty in Atmospheric Winds Retrieved from Optical Flow: Dependence on Weather Regime

Abstract This study explores the performance of a dense optical flow method in comparison to pattern-matching techniques for retrieving atmospheric motion vectors (AMVs) from water vapor images. Using high-resolution simulated datasets that represent various weather phenomena, we assess the performance of these methods across different weather regimes, time intervals, and pressure levels and quantify the uncertainties associated with retrieved winds. The optical flow algorithm consistently outperforms the feature matching approach. Notably, it produces wind speeds and AMVs that closely resemble the wind fields from the simulations, and unlike the feature matching algorithm, the optical flow algorithm exhibits consistent performance across different time intervals. In contrast, the feature matching approach yields vector fields that exhibit oversmoothing in certain areas and erratic behavior in others, while also producing less detailed, regionally constant speed maps. Furthermore, unlike feature matching, the optical flow method effectively calculates AMV near regions with missing data, generating a dense AMV field for every pixel in a pair of images. This superior performance and flexibility significantly influence the planning for future satellite missions aimed at retrieving atmospheric winds. As such, our work plays a critical role in determining the mission architecture and projected instrument performance for future atmospheric wind retrieval satellite missions. The study underscores the potential of the optical flow algorithm as a robust and efficient approach for atmospheric motion retrieval, thus contributing to advances in climate research and weather prediction. Significance Statement This research investigates the efficacy of two methods, optical flow and feature matching, for detecting atmospheric winds, referred to as atmospheric motion vectors, from satellite images of water vapor. Employing detailed simulated datasets that replicate real-world weather patterns, we found that optical flow consistently outperforms feature matching in various aspects. Notably, the optical flow method is not only more precise but also maintains its accuracy across different scenarios. These insights are critical for the design of future satellite missions focused on advancing our understanding of the atmosphere and enhancing weather predictions. This study contributes to advancements in climate research and supports improved weather forecasting, benefiting both scientific and societal needs.

Holographic image denoising for dense droplet field using conditional diffusion model.

The Letter delves into an approach to holographic image denoising, drawing inspiration from the generative paradigm. It introduces a conditional diffusion model framework that effectively suppresses twin-image noises and speckle noises in dense particle fields with a large depth of field (DOF). Specific training and inference configurations are meticulously outlined. For evaluation, the method is tested using calibration dot board data and droplet field data, encompassing gel atomization captured via inline holography and aviation kerosene swirl spray through off-axis holography. The performance is assessed using three distinct metrics. The metric outcomes, along with representative examples, robustly demonstrate its superior noise reduction, detail preservation, and generalization capabilities when compared to two other methods. The proposed method not only pioneers the field of generative holographic image denoising but also highlights its potential for industrial applications, given its reduced dependency on high-quality training labels.

Transsphenoidal Approach for Sellar Region Tumor : A Case Series

Introduction: Transsphenoidal approaches to the sellar region, with endoscopic or microscopic techniques, are regularly used to treat a diverse collection of pathologies. Case Description: Three patients with presenting symptoms visual disturbance and headache were evaluated with MRI that showes sellar region masses. All patients were undergo transphenoid tumor removal. Post operation evaluation show improvement in visual disturbance and headache with no adverse event after procedure. Discussion : Transsphenoidal endoscopic surgery (TSS) for functional pituitary adenomas yields better endocrinologic results for noninvasive macroadenomas. However, the rate of postoperative CSF leakage was greater with the endoscopic method. In patients with pituitary adenomas, younger age, dense visual field defect, and the preoperative absence of central or bilateral visual field abnormalities were predictive factors of visual field recovery following transsphenoidal approach-tumor excision. Overall survival rates are generally high, ranging from 91% to 98%. Conclusion: Transphenoid approach was a good surgical method for sellar region masses and show good outcomes after operation. Keyword : Sellar region masses; Surgical; Transsphenoid approach

Laboratory Abnormal Behavior Detection Based on Multimodal Information Fusion

The traditional laboratory anomaly detection methods mainly focus on the hidden dangers caused by chemical leaks and other items, ignoring the impact of abnormal behaviors such as incorrect operations and improper behavior on safety in the laboratory. This paper proposes a laboratory abnormal behavior detection method based on multimodal information fusion. The method generates a dense optical flow field of RGB image sequences based on optical flow theory and global smoothing constraints, and mines motion mode information. Meanwhile, the contour modal information of behavior is captured through convolution and adjacency matrix operations. Using decision level and proximity functions to integrate student behavior motion mode information and contour mode information, and using the maximum value as the behavior detection result. The experimental results show that the method can effectively detect abnormal behavior in the laboratory environment, with small detection errors and a specificity close to 1.00, effectively ensuring the safety of the laboratory environment.

Weakly supervised semantic segmentation of leukocyte images based on class activation maps.

Leukocytes are an essential component of the human defense system, accurate segmentation of leukocyte images is a crucial step towards automating detection. Most existing methods for leukocyte images segmentation relied on fully supervised semantic segmentation (FSSS) with extensive pixel-level annotations, which are time-consuming and labor-intensive. To address this issue, this paper proposes a weakly supervised semantic segmentation (WSSS) approach for leukocyte images utilizing improved class activation maps (CAMs). Firstly, to alleviate ambiguous boundary problem between leukocytes and background, preprocessing technique is employed to enhance the image quality. Secondly, attention mechanism is added to refine the CAMs generated by improving the matching of local and global features. Random walks, dense conditional random fields and hole filling were leveraged to obtain final pseudo-segmentation labels. Finally, a fully supervised segmentation network is trained with pseudo-segmentation labels. The method is evaluated on BCCD and TMAMD datasets. Experimental results demonstrate that by employing the pseudo segmentation annotations generated through this method can be utilized to train UNet as close as possible to FSSS. This method effectively reduces manual annotation cost while achieving WSSS of leukocyte images.

Insights from a topo-bathymetric and oceanographic dataset for coastal flooding studies: the French Flooding Prevention Action Program of Saint-Malo

Abstract. The French Flooding Prevention Action Program of Saint-Malo, France, requires the assessment of coastal flooding risks and the development of a local flood warning system. The first prerequisite is knowledge of the topography and bathymetry of the bay of Saint-Malo; the acquisition of new multibeam bathymetric data was performed in 2018 and 2019 to increase the resolution of the existing topo-bathymetric datasets and to produce two high-resolution (20 and 5 m) topo-bathymetric digital terrain models. Second, the hydrodynamics associated with coastal flooding were investigated through a dense and extensive oceanographic field experiment conducted during winter 2018–2019 using a network of 22 moorings with 37 sensors: the network included 2 directional buoys, 2 pressure tide gauges, 18 wave pressure gauges, 4 single-point current meters, 7 current profilers, and 4 acoustic wave current profilers from mid-depth (25 m) up to the upper beach and the dike system. The oceanographic dataset thus provides an extended overview of the hydrodynamics and wave processes in the bay of Saint-Malo from the coast up to over-flooding and over-topping areas. This dataset helps to identify the physical drivers of the coastal flooding and provides a quantification of their respective contributions. In particular, the wave processes at the foot of the protection structures can be observed: in this macro-tidal environment, during high spring tides, short and infragravity waves propagate up to the protection structures, while the wave set-up remains negligible, and over-topping by sea packs can occur. The combination of high-resolution topo-bathymetric and oceanographic datasets allows the construction, calibration and validation of a wave and hydrodynamic coupled model that is used to investigate flooding processes more deeply and might be integrated into a future local warning system by means of Saint-Malo inter-communality. The topo-bathymetric and oceanographic datasets are available freely at https://doi.org/10.17183/MNT_COTIER_GNB_PAPI_SM_20m_WGS84, https://doi.org/10.17183/MNT_COTIER_PORT_SM_PAPI_SM_5m_WGS84 and https://doi.org/10.17183/CAMPAGNE_OCEANO_STMALO (Shom, 2020a, b, 2021).

Variational Feature Extraction in Scientific Visualization

Across many scientific disciplines, the pursuit of even higher grid resolutions leads to a severe scalability problem in scientific computing. Feature extraction is a commonly chosen approach to reduce the amount of information from dense fields down to geometric primitives that further enable a quantitative analysis. Examples of common features are isolines, extremal lines, or vortex corelines. Due to the rising complexity of the observed phenomena, or in the event of discretization issues with the data, a straightforward application of textbook feature definitions is unfortunately insufficient. Thus, feature extraction from spatial data often requires substantial pre- or post-processing to either clean up the results or to include additional domain knowledge about the feature in question. Such a separate pre- or post-processing of features not only leads to suboptimal and incomparable solutions, it also results in many specialized feature extraction algorithms arising in the different application domains. In this paper, we establish a mathematical language that not only encompasses commonly used feature definitions, it also provides a set of regularizers that can be applied across the bounds of individual application domains. By using the language of variational calculus, we treat features as variational minimizers, which can be combined and regularized as needed. Our formulation not only encompasses existing feature definitions as special case, it also opens the path to novel feature definitions. This work lays the foundations for many new research directions regarding formal definitions, data representations, and numerical extraction algorithms.

A Bayesian Approach To Locally Varying Regularization In Optical Flow Velocimetry

A recent method for determining accurate, dense velocity fields from a pair of particle images is to solve the optical flow problem, but the ill-posed inverse problem of optical flow velocimetry (OFV) generally entails minimizing a weighted sum of two terms--fidelity and regularization--and the weights in the sum are parameters that require manual tuning based on the flow properties and particle images. This tuning of the weighting parameters makes the experimental applicability of the method much more challenging as the calculated velocity field has demonstrated sensitivity to the value of the weights. This work proposes a hierarchical model for the weighting parameters in the framework of maximum a posteriori (MAP)-based Bayesian optimization approaches. The resulting method is tested on three different synthetic datasets and on experimental particle images. The method is found to be capable of self-adjusting the local weights of the optimization process in real-time while simultaneously determining the velocity field, leading to a more accurate estimate of the velocity field without requiring any dataset-specific manual tuning of parameters.

Towards High-Speed And High-Resolution Real-Time Optical Flow Particle Image Velocimetry

One of the major advantages of Optical Flow PIV (Particle Image Velocimetry) algorithms over Cross-Correlation PIV is their scalability leading to potentially very high computational speeds. This is confirmed in this study using different GPUs (Graphics Processor Unit) and different image sizes. The other advantage is the possibility of obtaining dense velocity fields of up to one vector per pixel. It is well known that particle seeding plays a crucial role in the results of standard particle image velocimetry based on cross-correlation algorithms. Its influence on the quality of the optical flow algorithm is not as well established. In this article the influence of particle concentration is quantified by introducing a criterion considering the proportion of “active” pixels in a snapshot. It is shown that it is possible to optimize particle concentration to maximize the percentage of active pixels, leading to better spatial resolution, down to one vector per pixel. The principle is validated on a vortex-free flow and applied to the complex 3D flow downstream a backward-facing step.

Benchmarking Data Assimilation Algorithms For 3D Lagrangian Particle Tracking

This paper reports a comparison of three state-of-the-art data assimilation (DA) algorithms for 3D Lagrangian particle tracking (LPT): Vortex-In-Cell sharp (VIC#), FlowFit3, and neural-implicit particle advection (NIPA). Particle trajectories, termed “tracks,” are spatially sparse, and a reconstruction algorithm is commonly employed to estimate dense Eulerian fields using position, velocity, and/or acceleration data from the tracks. DA algorithms for LPT combine the tracks with a set of governing equations (or constraints) to enhance the accuracy of the velocity field, relative to interpolation methods, and infer additional quantities like pressure. We compare the performance of VIC#, FlowFit3, and NIPA with respect to their accuracy and computational cost. Accuracy is assessed in terms of the mean inter-particle distance, frame rate, and magnitude of tracking errors, and costs are reported in wall time. Synthetic and experimental data sets for homogeneous isotropic turbulence and turbulent boundary layer flows are evaluated; direct numerical simulations are used for the synthetic tests, and realistic localization errors are added to the simulated tracks. All three DA methods perform well in cases with the highest spatio-temporal sampling of the flow, and all three methods are relatively robust to the frame rate and magnitude of localization errors. NIPA is more resilient to sparse seeding than FlowFit or VIC#, which exhibit similar performance, but NIPA is also the most expensive technique by some margin. DA reconstructions are superior to interpolation across the board, with a reduction of error up to 80% in the experimental demonstration.

Scanning Lagrangian Particle Tracking To Measure 3D Large Scale Aerodynamics Of Quadcopter Flight

We present large-scale time-resolved and volumetric measurements of the flow surrounding a quadcopter drone. Cases with hovering flight, forward flight - with and without ground-effect - as well as lift-off scenarios have been realized. Five high-speed cameras were used to capture images of hundreds of thousands heliumfilled soap bubbles being illuminated by pulsed LED arrays. The arrays were operated in a dual-scanning mode, where the front and the back part of the volume are illuminated intermittently, each with a frequency of 3 kHz. The cameras were recording at 6 kHz, thereby capturing both sub-time-series in a single measurement run. Compared to a full illumination of the volume, we can either reduce the number of particles imaged on the cameras given a fixed tracer concentration - facilitating the reconstruction - or operate at a higher tracer concentration while keeping the number of imaged particles fixed. The dual time-series were evaluated by Lagrangian Particle Tracking, using the Shake-The-Box method, yielding dense fields of particle tracks. For the current results, the time series for both sub-volumes were evaluated independently. For the final contribution, an integrated evaluation scheme, alternatingly operating on both time-series is foreseen. After the tracking step, flow structures were identified using the data assimilation method FlowFit. The temporally and spatially highly resolved results document the strong interaction of the wakes induced by several rotors, as well as the interaction of wing tip vortices stemming from a single rotor. Brown-out effects are documented in forward flight. Averaged results of hovering flight enable an evaluation of the azimuthal distribution of the downwash- and outwash-velocities resulting from the geometric layout of the four rotors. By having highly resolved data encompassing the whole flight vehicle, the forces conveyed by the drone to the liquid can be determined via surface- or volume-integration.

An improved 3D-UNet-based brain hippocampus segmentation model based on MR images

ObjectiveAccurate delineation of the hippocampal region via magnetic resonance imaging (MRI) is crucial for the prevention and early diagnosis of neurosystemic diseases. Determining how to accurately and quickly delineate the hippocampus from MRI results has become a serious issue. In this study, a pixel-level semantic segmentation method using 3D-UNet is proposed to realize the automatic segmentation of the brain hippocampus from MRI results. Methods: Two hundred three-dimensional T1-weighted (3D-T1) nongadolinium contrast-enhanced magnetic resonance (MR) images were acquired at Hangzhou Cancer Hospital from June 2020 to December 2022. These samples were divided into two groups, containing 175 and 25 samples. In the first group, 145 cases were used to train the hippocampus segmentation model, and the remaining 30 cases were used to fine-tune the hyperparameters of the model. Images for twenty-five patients in the second group were used as the test set to evaluate the performance of the model. The training set of images was processed via rotation, scaling, grey value augmentation and transformation with a smooth dense deformation field for both image data and ground truth labels. A filling technique was introduced into the segmentation network to establish the hippocampus segmentation model. In addition, the performance of models established with the original network, such as VNet, SegResNet, UNetR and 3D-UNet, was compared with that of models constructed by combining the filling technique with the original segmentation network. Results: The results showed that the performance of the segmentation model improved after the filling technique was introduced. Specifically, when the filling technique was introduced into VNet, SegResNet, 3D-UNet and UNetR, the segmentation performance of the models trained with an input image size of 48 × 48 × 48 improved. Among them, the 3D-UNet-based model with the filling technique achieved the best performance, with a Dice score (Dice score) of 0.7989 ± 0.0398 and a mean intersection over union (mIoU) of 0.6669 ± 0.0540, which were greater than those of the original 3D-UNet-based model. In addition, the oversegmentation ratio (OSR), average surface distance (ASD) and Hausdorff distance (HD) were 0.0666 ± 0.0351, 0.5733 ± 0.1018 and 5.1235 ± 1.4397, respectively, which were better than those of the other models. In addition, when the size of the input image was set to 48 × 48 × 48, 64 × 64 × 64 and 96 × 96 × 96, the model performance gradually improved, and the Dice scores of the proposed model reached 0.7989 ± 0.0398, 0.8371 ± 0.0254 and 0.8674 ± 0.0257, respectively. In addition, the mIoUs reached 0.6669 ± 0.0540, 0.7207 ± 0.0370 and 0.7668 ± 0.0392, respectively. Conclusion: The proposed hippocampus segmentation model constructed by introducing the filling technique into a segmentation network performed better than models built solely on the original network and can improve the efficiency of diagnostic analysis.

CSST Dense Star Field Preparation: A Framework for Astrometry and Photometry for Dense Star Field Images Obtained by the China Space Station Telescope (CSST)

The China Space Station Telescope (CSST) is a telescope with 2 m diameter, obtaining images with high quality through wide-field observations. In its first observation cycle, to capture time-domain observation data, the CSST is proposed to observe the Galactic halo across different epochs. These data have significant potential for the study of properties of stars and exoplanets. However, the density of stars in the Galactic center is high, and it is a well-known challenge to perform astrometry and photometry in such a dense star field. This paper presents a deep learning-based framework designed to process dense star field images obtained by the CSST, which includes photometry, astrometry, and classifications of targets according to their light curve periods. With simulated CSST observation data, we demonstrate that this deep learning framework achieves photometry accuracy of 2% and astrometry accuracy of 0.03 pixel for stars with moderate brightness mag = 24 (i band), surpassing results obtained by traditional methods. Additionally, the deep learning based light curve classification algorithm could pick up celestial targets whose magnitude variations are 1.7 times larger than magnitude variations brought by Poisson photon noise. We anticipate that our framework could be effectively used to process dense star field images obtained by the CSST.

Iterative modal reconstruction for sparse particle tracking data

A method to reconstruct the dense velocity field from relatively sparse particle tracks is introduced. The approach leverages the properties of proper orthogonal decomposition (POD), and it iteratively reconstructs the detailed spatial modes from a first, coarse estimation thereof. The initially coarse Cartesian representation of the velocity field is obtained by local data averaging, where POD is applied. The spatial resolution of the POD modes is enhanced by reprojecting them onto the sparse particle velocity to iteratively improve the reconstruction of the temporal coefficients. Finally, the enhanced velocity field is represented at high-resolution with a reduced order model using the dominant POD modes. The method is referred to as iterative modal reconstruction (IMR), as an extension of the recently proposed data-enhanced particle tracking velocimetry algorithm, introduced for cross correlation-based velocity data. Experiments in the wake of a cylinder at ReD = 27 000 are used to assess the suitability of the method to resolve the turbulent Kármán–Benard wake. The approach is benchmarked against traditional as well as state-of-the-art reconstruction methods, illustrating the capability of IMR of enhancing the spatial resolution of sparse velocity data.

A fused LASSO operator for fast 3D magnetic particle imaging reconstruction

Objective. Magnetic particle imaging (MPI) is a promising imaging modality that leverages the nonlinear magnetization behavior of superparamagnetic iron oxide nanoparticles to determine their concentration distribution. Previous optimization models with multiple regularization terms have been proposed to achieve high-quality MPI reconstruction, but these models often result in increased computational burden, particularly for dense gridding 3D fields of view. In order to achieve faster reconstruction speeds without compromising reconstruction quality, we have developed a novel fused LASSO operator, total sum-difference (TSD), which effectively captures the sparse and smooth priors of MPI images. Methods. Through an analysis-synthesis equivalence strategy and a constraint smoothing strategy, the TSD regularized model was solved using the fast iterative soft-thresholding algorithm (FISTA). The resulting reconstruction method, TSD-FISTA, boasts low computational complexity and quadratic convergence rate over iterations. Results. Experimental results demonstrated that TSD-FISTA required only 10% and 37% of the time to achieve comparable or superior reconstruction quality compared to commonly used fused LASSO-based alternating direction method of multipliers and Tikhonov-based algebraic reconstruction techniques, respectively. Significance. TSD-FISTA shows promise for enabling real-time 3D MPI reconstruction at high frame rates for large fields of view.

Maize smart-canopy architecture enhances yield at high densities.

Increasing planting density is a key strategy for enhancing maize yields1-3. An ideotype for dense planting requires a 'smart canopy' with leaf angles at different canopy layers differentially optimized to maximize light interception and photosynthesis4-6, among other features. Here we identified leaf angle architecture of smart canopy 1 (lac1), a natural mutant with upright upper leaves, less erect middle leaves and relatively flat lower leaves. lac1 has improved photosynthetic capacity and attenuated responses to shadeunder dense planting. lac1 encodes a brassinosteroid C-22 hydroxylase that predominantly regulates upper leaf angle. Phytochrome A photoreceptors accumulate in shade and interact with the transcription factor RAVL1 to promote its degradation via the 26S proteasome, thereby inhibiting activation of lac1 by RAVL1 and decreasing brassinosteroid levels. This ultimately decreases upper leaf angle in dense fields. Large-scale field trials demonstrate that lac1 boosts maize yields under high planting densities. To quickly introduce lac1 into breeding germplasm, we transformed a haploid inducer and recovered homozygous lac1 edits from 20 diverse inbred lines. The tested doubled haploids uniformly acquired smart-canopy-like plant architecture. We provide an important target and an accelerated strategy for developing high-density-tolerant cultivars, with lac1 serving as a genetic chassis for further engineering of a smart canopy in maize.

Comment on Ádám et al. (2023) : Large fraction of already known systems reported

In this work, I report that large fraction of stars detected by A A, 674, A170 2023A&A...674A.170A and noted in that work as new discoveries are in fact known systems. This is especially true for the dense bulge fields with large blending of nearby sources. Among the published 245 stars determined to be doubly eclipsing (i.e. containing two eclipsing signals), I identified 53 blends. In other words, about a quarter of the systems noted by 2023A&A...674A.170A are not actually doubly eclipsing; rather, these are contaminations of known nearby sources that have already been detected by OGLE. Such a high proportion of reported false positives should not be readily ignored and ought to be addressed in future studies.

RegFSC-Net: Medical Image Registration via Fourier Transform With Spatial Reorganization and Channel Refinement Network.

Medical image registration is crucial in medical image analysis applications. Recently, U-Net-style networks have been commonly used for unsupervised image registration, predicting dense displacement fields in full-resolution space. However, this process is resource-intensive and time-consuming for high-resolution volumetric image data. To address this challenge, this paper proposes a novel model named RegFSC-Net, which utilizes Fourier transform with spatial reorganization (SR) and channel refinement (CR) network for registration. We embed efficient feature extraction modules SR and CR modules into the encoder, and adopt a parameter-free model to drive the decoder to improve the U-shaped network. Precisely, RegFSC-Net does not directly predict the full-resolution displacement field in space but learns the low-dimensional representation of the displacement field in the bandlimited Fourier domain, which is beneficial in reducing network parameters, memory usage, and computational costs. Experimental results show that RegFSC-Net outperforms various state-of-the-art methods. Specifically, in comparison to the widely recognized Transformer-based method TransMorph, RegFSC-Net utilizes only around 8.2% of its parameters, resulting in a 1.95% higher Dice score and significantly faster inference speeds of 126.67% and 419.99% on GPU and CPU, respectively. Furthermore, we also designed three variants of RegFSC-Net and demonstrated their potential applications in computer-aided diagnosis.

Uncertainty estimation and evaluation of deformation image registration based convolutional neural networks

Objective. Fast and accurate deformable image registration (DIR), including DIR uncertainty estimation, is essential for safe and reliable clinical deployment. While recent deep learning models have shown promise in predicting DIR with its uncertainty, challenges persist in proper uncertainty evaluation and hyperparameter optimization for these methods. This work aims to develop and evaluate a model that can perform fast DIR and predict its uncertainty in seconds. Approach. This study introduces a novel probabilistic multi-resolution image registration model utilizing convolutional neural networks to estimate a multivariate normal distributed dense displacement field (DDF) in a multimodal image registration problem. To assess the quality of the DDF distribution predicted by the model, we propose a new metric based on the Kullback–Leibler divergence. The performance of our approach was evaluated against three other DIR algorithms (VoxelMorph, Monte Carlo dropout, and Monte Carlo B-spline) capable of predicting uncertainty. The evaluation of the models included not only the quality of the deformation but also the reliability of the estimated uncertainty. Our application investigated the registration of a treatment planning computed tomography (CT) to follow-up cone beam CT for daily adaptive radiotherapy. Main results. The hyperparameter tuning of the models showed a trade-off between the estimated uncertainty’s reliability and the deformation’s accuracy. In the optimal trade-off, our model excelled in contour propagation and uncertainty estimation (p <0.05) compared to existing uncertainty estimation models. We obtained an average dice similarity coefficient of 0.89 and a KL-divergence of 0.15. Significance. By addressing challenges in DIR uncertainty estimation and evaluation, our work showed that both the DIR and its uncertainty can be reliably predicted, paving the way for safe deployment in a clinical environment.