Search

Deep learning has shown great potential for automated medical image segmentation to improve the precision and speed of disease diagnostics. However, the task presents significant difficulties due to variations in the scale, shape, texture, and contrast of the pathologies. Traditional convolutional neural network (CNN) models have certain limitations when it comes to effectively modelling multiscale context information and facilitating information interaction between skip connections across levels. To overcome these limitations, a novel deep learning architecture is introduced for medical image segmentation, taking advantage of CNNs and vision transformers. Our proposed model, named TBConvL-Net, involves a hybrid network that combines the local features of a CNN encoder-decoder architecture with long-range and temporal dependencies using biconvolutional long-short-term memory (LSTM) networks and vision transformers (ViT). This enables the model to capture contextual channel relationships in the data and account for the uncertainty of segmentation over time. Additionally, we introduce a novel composite loss function that considers both the segmentation robustness and the boundary agreement of the predicted output with the gold standard. Our proposed model shows consistent improvement over the state of the art on ten publicly available datasets of seven different medical imaging modalities.

Retinal diseases can cause irreversible vision loss in both eyes if not diagnosed and treated early. Since retinal diseases are so complicated, retinal imaging is likely to show two or more abnormalities. Current deep learning techniques for segmenting retinal images with many labels and attributes have poor detection accuracy and generalisability. This paper presents a multipath convolutional neural network for multifeature segmentation. The proposed network is lightweight and spatially sensitive to information. A patch-based implementation is used to extract local image features, and focal modulation attention blocks are incorporated between the encoder and the decoder for improved segmentation. Filter optimisation is used to prevent filter overlaps and speed up model convergence. A combination of convolution operations and group convolution operations is used to reduce computational costs. This is the first robust and generalisable network capable of segmenting multiple features of fundus images (including retinal vessels, microaneurysms, optic discs, haemorrhages, hard exudates, and soft exudates). The results of our experimental evaluation on more than ten publicly available datasets with multiple features show that the proposed network outperforms recent networks despite having a small number of learnable parameters.

Quadrotor systems are common and beneficial for many fields, but their intricate behavior often makes it challenging to design effective and optimal control strategies. Some traditional approaches to nonlinear control often rely on local linearizations or complex nonlinear models, which can be inaccurate or computationally expensive. We present a data-driven approach to identify the dynamics of a given quadrotor system using Koopman operator theory. Koopman theory offers a framework for representing nonlinear dynamics as linear operators acting on observable functions of the state space. This allows to approximate nonlinear systems with globally linear models in a higher dimensional space, which can be analyzed and controlled using standard linear optimal control techniques. We leverage the method of extended dynamic mode decomposition (EDMD) to identify Koopman operator from data with total least squares. We demonstrate that the identified model can be stabilized and controllable by designing a controller using linear quadratic regulator (LQR).

Multi-robot systems must have the ability to accurately estimate relative states between robots in order to perform collaborative tasks, possibly with no external aiding. Three-dimensional relative pose estimation using range measurements oftentimes suffers from a finite number of non-unique solutions, or ambiguities. This paper: 1) identifies and accurately estimates all possible ambiguities in 2D; 2) treats them as components of a Gaussian mixture model; and 3) presents a computationally-efficient estimator, in the form of a Gaussian-sum filter (GSF), to realize range-based relative pose estimation in an infrastructure-free, 3D, setup. This estimator is evaluated in simulation and experiment and is shown to avoid divergence to local minima induced by the ambiguous poses. Furthermore, the proposed GSF outperforms an extended Kalman filter, demonstrates similar performance to the computationally-demanding particle filter, and is shown to be consistent., Comment: 7 pages, 9 figures, submitted to IEEE Conference on Control Technology and Applications 2024

Segmentation is an important task in a wide range of computer vision applications, including medical image analysis. Recent years have seen an increase in the complexity of medical image segmentation approaches based on sophisticated convolutional neural network architectures. This progress has led to incremental enhancements in performance on widely recognised benchmark datasets. However, most of the existing approaches are computationally demanding, which limits their practical applicability. This paper presents an expand-squeeze dual multiscale residual network (ESDMR-Net), which is a fully convolutional network that is particularly well-suited for resource-constrained computing hardware such as mobile devices. ESDMR-Net focuses on extracting multiscale features, enabling the learning of contextual dependencies among semantically distinct features. The ESDMR-Net architecture allows dual-stream information flow within encoder-decoder pairs. The expansion operation (depthwise separable convolution) makes all of the rich features with multiscale information available to the squeeze operation (bottleneck layer), which then extracts the necessary information for the segmentation task. The Expand-Squeeze (ES) block helps the network pay more attention to under-represented classes, which contributes to improved segmentation accuracy. To enhance the flow of information across multiple resolutions or scales, we integrated dual multiscale residual (DMR) blocks into the skip connection. This integration enables the decoder to access features from various levels of abstraction, ultimately resulting in more comprehensive feature representations. We present experiments on seven datasets from five distinct examples of applications. Our model achieved the best results despite having significantly fewer trainable parameters, with a reduction of two or even three orders of magnitude.

Blinding eye diseases are often correlated with altered retinal morphology, which can be clinically identified by segmenting retinal structures in fundus images. However, current methodologies often fall short in accurately segmenting delicate vessels. Although deep learning has shown promise in medical image segmentation, its reliance on repeated convolution and pooling operations can hinder the representation of edge information, ultimately limiting overall segmentation accuracy. In this paper, we propose a lightweight pixel-level CNN named LMBiS-Net for the segmentation of retinal vessels with an exceptionally low number of learnable parameters \textbf{(only 0.172 M)}. The network used multipath feature extraction blocks and incorporates bidirectional skip connections for the information flow between the encoder and decoder. Additionally, we have optimized the efficiency of the model by carefully selecting the number of filters to avoid filter overlap. This optimization significantly reduces training time and enhances computational efficiency. To assess the robustness and generalizability of LMBiS-Net, we performed comprehensive evaluations on various aspects of retinal images. Specifically, the model was subjected to rigorous tests to accurately segment retinal vessels, which play a vital role in ophthalmological diagnosis and treatment. By focusing on the retinal blood vessels, we were able to thoroughly analyze the performance and effectiveness of the LMBiS-Net model. The results of our tests demonstrate that LMBiS-Net is not only robust and generalizable but also capable of maintaining high levels of segmentation accuracy. These characteristics highlight the potential of LMBiS-Net as an efficient tool for high-speed and accurate segmentation of retinal images in various clinical applications.

Skin cancer holds the highest incidence rate among all cancers globally. The importance of early detection cannot be overstated, as late-stage cases can be lethal. Classifying skin lesions, however, presents several challenges due to the many variations they can exhibit, such as differences in colour, shape, and size, significant variation within the same class, and notable similarities between different classes. This paper introduces a novel class-wise attention technique that equally regards each class while unearthing more specific details about skin lesions. This attention mechanism is progressively used to amalgamate discriminative feature details from multiple scales. The introduced technique demonstrated impressive performance, surpassing more than 15 cutting-edge methods including the winners of HAM1000 and ISIC 2019 leaderboards. It achieved an impressive accuracy rate of 97.40% on the HAM10000 dataset and 94.9% on the ISIC 2019 dataset.

In this study, we propose LDMRes-Net, a lightweight dual-multiscale residual block-based computational neural network tailored for medical image segmentation on IoT and edge platforms. Conventional U-Net-based models face challenges in meeting the speed and efficiency demands of real-time clinical applications, such as disease monitoring, radiation therapy, and image-guided surgery. LDMRes-Net overcomes these limitations with its remarkably low number of learnable parameters (0.072M), making it highly suitable for resource-constrained devices. The model's key innovation lies in its dual multi-residual block architecture, which enables the extraction of refined features on multiple scales, enhancing overall segmentation performance. To further optimize efficiency, the number of filters is carefully selected to prevent overlap, reduce training time, and improve computational efficiency. The study includes comprehensive evaluations, focusing on segmentation of the retinal image of vessels and hard exudates crucial for the diagnosis and treatment of ophthalmology. The results demonstrate the robustness, generalizability, and high segmentation accuracy of LDMRes-Net, positioning it as an efficient tool for accurate and rapid medical image segmentation in diverse clinical applications, particularly on IoT and edge platforms. Such advances hold significant promise for improving healthcare outcomes and enabling real-time medical image analysis in resource-limited settings.

Timely and affordable computer-aided diagnosis of retinal diseases is pivotal in precluding blindness. Accurate retinal vessel segmentation plays an important role in disease progression and diagnosis of such vision-threatening diseases. To this end, we propose a Multi-resolution Contextual Network (MRC-Net) that addresses these issues by extracting multi-scale features to learn contextual dependencies between semantically different features and using bi-directional recurrent learning to model former-latter and latter-former dependencies. Another key idea is training in adversarial settings for foreground segmentation improvement through optimization of the region-based scores. This novel strategy boosts the performance of the segmentation network in terms of the Dice score (and correspondingly Jaccard index) while keeping the number of trainable parameters comparatively low. We have evaluated our method on three benchmark datasets, including DRIVE, STARE, and CHASE, demonstrating its superior performance as compared with competitive approaches elsewhere in the literature.

PHENIX presents a simultaneous measurement of the production of direct $\gamma$ and $\pi^0$ in $d$$+$Au collisions at $\sqrt{s_{_{NN}}}=200$ GeV over a $p_T$ range of 7.5 to 18 GeV/$c$ for different event samples selected by event activity, i.e. charged-particle multiplicity detected at forward rapidity. Direct-photon yields are used to empirically estimate the contribution of hard-scattering processes in the different event samples. Using this estimate, the average nuclear-modification factor $R_{d\rm Au,EXP}^{\gamma^{\rm dir}}$ is $0.925{\pm}0.023({\rm stat}){\pm}0.15^{\rm (scale)}$, consistent with unity for minimum-bias (MB) $d$$+$Au events. For event classes with moderate event activity, $R_{d\rm Au,EXP}^{\gamma^{\rm dir}}$ is consistent with the MB value within 5\% uncertainty. These results confirm that the previously observed enhancement of high-$p_T$ $\pi^0$ production found in small-system collisions with low event activity is a result of a bias in interpreting event activity within the Glauber framework. In contrast, for the top 5\% of events with the highest event activity, $R_{d\rm Au,EXP}^{\gamma^{\rm dir}}$ is suppressed by 20\% relative to the MB value with a significance of $4.5\sigma$, which may be due to final-state effects., Comment: 279 authors from 69 institutions, 8 pages, 3 figures, v1 is version submitted to Physical Review Letters. HEPdata tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html

Compression of convolutional neural network models has recently been dominated by pruning approaches. A class of previous works focuses solely on pruning the unimportant filters to achieve network compression. Another important direction is the design of sparsity-inducing constraints which has also been explored in isolation. This paper presents a novel training scheme based on composite constraints that prune redundant filters and minimize their effect on overall network learning via sparsity promotion. Also, as opposed to prior works that employ pseudo-norm-based sparsity-inducing constraints, we propose a sparse scheme based on gradient counting in our framework. Our tests on several pixel-wise segmentation benchmarks show that the number of neurons and the memory footprint of networks in the test phase are significantly reduced without affecting performance. MobileNetV3 and UNet, two well-known architectures, are used to test the proposed scheme. Our network compression method not only results in reduced parameters but also achieves improved performance compared to MobileNetv3, which is an already optimized architecture.

Recently, the PHENIX Collaboration has published second- and third-harmonic Fourier coefficients $v_2$ and $v_3$ for midrapidity ($|\eta|<0.35$) charged hadrons in 0\%--5\% central $p$$+$Au, $d$$+$Au, and $^3$He$+$Au collisions at $\sqrt{s_{_{NN}}}=200$ GeV utilizing three sets of two-particle correlations for two detector combinations with different pseudorapidity acceptance [Phys. Rev. C {\bf 105}, 024901 (2022)]. This paper extends these measurements of $v_2$ to all centralities in $p$$+$Au, $d$$+$Au, and $^3$He$+$Au collisions, as well as $p$$+$$p$ collisions, as a function of transverse momentum ($p_T$) and event multiplicity. The kinematic dependence of $v_2$ is quantified as the ratio $R$ of $v_2$ between the two detector combinations as a function of event multiplicity for $0.5$$<$$p_T$$<$$1$ and $2$$<$$p_T$$<$$2.5$ GeV/$c$. A multiphase-transport (AMPT) model can reproduce the observed $v_2$ in most-central to midcentral $d$$+$Au and $^3$He$+$Au collisions. However, the AMPT model systematically overestimates the measurements in $p$$+$$p$, $p$$+$Au, and peripheral $d$$+$Au and $^3$He$+$Au collisions, indicating a higher nonflow contribution in AMPT than in the experimental data. The AMPT model fails to describe the observed $R$ for $0.5$$<$$p_T$$<$$1$ GeV/$c$, but there is qualitative agreement with the measurements for $2$$<$$p_T$$<$$2.5$ GeV/$c$., Comment: 393 authors from 72 institutions, 14 pages, 10 figures, 2014, 2015, and 2016 data. v2 is version accepted for publication in Physical Review C. HEPdata tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html

Diabetic retinopathy refers to the pathology of the retina induced by diabetes and is one of the leading causes of preventable blindness in the world. Early detection of diabetic retinopathy is critical to avoid vision problem through continuous screening and treatment. In traditional clinical practice, the involved lesions are manually detected using photographs of the fundus. However, this task is cumbersome and time-consuming and requires intense effort due to the small size of lesion and low contrast of the images. Thus, computer-assisted diagnosis of diabetic retinopathy based on the detection of red lesions is actively being explored recently. In this paper, we present a convolutional neural network with residual skip connection for the segmentation of exudates in retinal images. To improve the performance of network architecture, a suitable image augmentation technique is used. The proposed network can robustly segment exudates with high accuracy, which makes it suitable for diabetic retinopathy screening. Comparative performance analysis of three benchmark databases: HEI-MED, E-ophtha, and DiaretDB1 is presented. It is shown that the proposed method achieves accuracy (0.98, 0.99, 0.98) and sensitivity (0.97, 0.92, and 0.95) on E-ophtha, HEI-MED, and DiaReTDB1, respectively.

Over recent years, increasingly complex approaches based on sophisticated convolutional neural network architectures have been slowly pushing performance on well-established benchmark datasets. In this paper, we take a step back to examine the real need for such complexity. We present RC-Net, a fully convolutional network, where the number of filters per layer is optimized to reduce feature overlapping and complexity. We also used skip connections to keep spatial information loss to a minimum by keeping the number of pooling operations in the network to a minimum. Two publicly available retinal vessel segmentation datasets were used in our experiments. In our experiments, RC-Net is quite competitive, outperforming alternatives vessels segmentation methods with two or even three orders of magnitude less trainable parameters.

Recent progress in encoder-decoder neural network architecture design has led to significant performance improvements in a wide range of medical image segmentation tasks. However, state-of-the-art networks for a given task may be too computationally demanding to run on affordable hardware, and thus users often resort to practical workarounds by modifying various macro-level design aspects. Two common examples are downsampling of the input images and reducing the network depth to meet computer memory constraints. In this paper we investigate the effects of these changes on segmentation performance and show that image complexity can be used as a guideline in choosing what is best for a given dataset. We consider four statistical measures to quantify image complexity and evaluate their suitability on ten different public datasets. For the purpose of our experiments we also propose two new encoder-decoder architectures representing shallow and deep networks that are more memory efficient than currently popular networks. Our results suggest that median frequency is the best complexity measure in deciding about an acceptable input downsampling factor and network depth. For high-complexity datasets, a shallow network running on the original images may yield better segmentation results than a deep network running on downsampled images, whereas the opposite may be the case for low-complexity images.

There is strong evidence for the formation of small droplets of quark-gluon plasma in $p/d/^{3}$He+Au collisions at the Relativistic Heavy Ion Collider (RHIC) and in $p$+$p$/Pb collisions at the Large Hadron Collider. In particular, the analysis of data at RHIC for different geometries obtained by varying the projectile size and shape has proven insightful. In the present analysis, we find excellent agreement with the previously published PHENIX at RHIC results on elliptical and triangular flow with an independent analysis via the two-particle correlation method, which has quite different systematic uncertainties and an independent code base. In addition, the results are extended to other detector combinations with different kinematic (pseudorapidity) coverage. These results provide additional constraints on contributions from nonflow and longitudinal decorrelations., Comment: 383 authors from 70 institutions, 13 pages, 12 figures, 4 tables, 2014,15,16 data (and supplemental material of 9 pages with 6 figures and 16 tables). v2 is version accepted for publication in Physical Review C. Plain text data tables for the points plotted in figures for this and previous PHENIX publications are (or will be) publicly available at http://www.phenix.bnl.gov/papers.html