Boosting Cardiac Color Doppler Frame Rates with Deep Learning.

Color Doppler echocardiography enables visualization of blood flow within the heart. However, the limited frame rate impedes the quantitative assessment of blood velocity throughout the cardiac cycle, thereby compromising a comprehensive analysis of ventricular filling. Concurrently, deep learning is demonstrating promising outcomes in post-processing of echocardiographic data for various applications. This work explores the use of deep learning models for intracardiac Doppler velocity estimation from a reduced number of filtered I/Q signals. We used a supervised learning approach by simulating patient-based cardiac color Doppler acquisitions and proposed data augmentation strategies to enlarge the training dataset. We implemented architectures based on convolutional neural networks. In particular, we focused on comparing the U-Net model and the recent ConvNeXt models, alongside assessing real-valued versus complex-valued representations. We found that both models outperformed the state-of-the-art autocorrelator method, effectively mitigating aliasing and noise. We did not observe significant differences between the use of real and complex data. Finally, we validated the models on in vitro and in vivo experiments. All models produced quantitatively comparable results to the baseline and were more robust to noise. ConvNeXt emerged as the sole model to achieve high-quality results on in vivo aliased samples. These results demonstrate the interest of supervised deep learning methods for Doppler velocity estimation from a reduced number of acquisitions.

Physics-Guided Neural Networks for Intraventricular Vector Flow Mapping.

Intraventricular vector flow mapping (iVFM) seeks to enhance and quantify color Doppler in cardiac imaging. In this study, we propose novel alternatives to the traditional iVFM optimization scheme by utilizing physics-informed neural networks (PINNs) and a physics-guided nnU-Net-based supervised approach. When evaluated on simulated color Doppler images derived from a patient-specific computational fluid dynamics model and in vivo Doppler acquisitions, both approaches demonstrate comparable reconstruction performance to the original iVFM algorithm. The efficiency of PINNs is boosted through dual-stage optimization and pre-optimized weights. On the other hand, the nnU-Net method excels in generalizability and real-time capabilities. Notably, nnU-Net shows superior robustness on sparse and truncated Doppler data while maintaining independence from explicit boundary conditions. Overall, our results highlight the effectiveness of these methods in reconstructing intraventricular vector blood flow. The study also suggests potential applications of PINNs in ultrafast color Doppler imaging and the incorporation of fluid dynamics equations to derive biomarkers for cardiovascular diseases based on blood flow.

Multi-transcriptomics identifies targets of the endoribonuclease DNE1 and highlights its coordination with decapping.

Decapping is a crucial step in mRNA degradation in eucaryotes and requires the formation of a holoenzyme complex between the decapping enzyme DECAPPING 2 (DCP2) and the decapping enhancer DCP1. In Arabidopsis (Arabidopsis thaliana), DCP1-ASSOCIATED NYN ENDORIBONUCLEASE 1 (DNE1) is a direct protein partner of DCP1. The function of both DNE1 and decapping are necessary to maintain phyllotaxis, the regularity of organ emergence in the apex. In this study, we combined in vivo mRNA editing, RNA degradome sequencing, transcriptomics and small RNA-omics to identify targets of DNE1 and study how DNE1 and DCP2 cooperate in controlling mRNA fate. Our data reveal that DNE1 mainly contacts and cleaves mRNAs in the coding sequence and has sequence cleavage preferences. DNE1 targets are also degraded through decapping, and both RNA degradation pathways influence the production of mRNA-derived small interfering RNAs. Finally, we detected mRNA features enriched in DNE1 targets including RNA G-quadruplexes and translated upstream open reading frames. Combining these four complementary high-throughput sequencing strategies greatly expands the range of DNE1 targets and allowed us to build a conceptual framework describing the influence of DNE1 and decapping on mRNA fate. These data will be crucial to unveil the specificity of DNE1 action and understand its importance for developmental patterning.

SIMUS3: An open-source simulator for 3-D ultrasound imaging.

BACKGROUND AND OBJECTIVE: Computational Ultrasound Imaging (CUI) has become increasingly popular in the medical ultrasound community, facilitated by free simulation software. These tools enable the design and exploration of transmit sequences, transducer arrays, and signal processing. We recently introduced SIMUS, a frequency-based ultrasound simulator within the open-source MUST toolbox, which offers numerical advantages and allows easy consideration of frequency-dependent factors. In response to the growing interest in simulating ultrasound imaging with 2-D matrix arrays, we present 3-D versions, PFIELD3 and SIMUS3. METHOD: The linear acoustic equations driving these functions are described, with theoretical assumptions reviewed for user guidance. RESULTS: Comparative analyses with Field II, using a 32×32 element 3-MHz matrix array, highlight the performance of PFIELD3 and SIMUS3 under various transmission conditions. CONCLUSIONS: This work extends the capabilities of existing CUI tools and provides researchers with valuable resources for advanced ultrasound simulations.

Heat stress promotes Arabidopsis AGO1 phase separation and association with stress granule components.

In Arabidopsis thaliana, ARGONAUTE1 (AGO1) plays a central role in microRNA (miRNA) and small interfering RNA (siRNA)-mediated silencing. AGO1 associates to the rough endoplasmic reticulum to conduct miRNA-mediated translational repression, mRNA cleavage, and biogenesis of phased siRNAs. Here, we show that a 37°C heat stress (HS) promotes AGO1 protein accumulation in cytosolic condensates where it colocalizes with components of siRNA bodies and of stress granules. AGO1 contains a prion-like domain in its poorly characterized N-terminal Poly-Q domain, which is sufficient to undergo phase separation independently of the presence of SGS3. HS only moderately affects the small RNA repertoire, the loading of AGO1 by miRNAs, and the signatures of target cleavage, suggesting that its localization in condensates protects AGO1 rather than promoting or impairing its activity in reprogramming gene expression during stress. Collectively, our work sheds new light on the impact of high temperature on a main effector of RNA silencing in plants.

Reply to "Comment on "So you think you can DAS? A viewpoint on delay-and-sum beamforming" - Think twice before f-numbering", published by M.F. Schiffner and G. Schmitz.

In a 2021 paper, we delved into the details of delay-sum beamforming (DAS) in high-frame-rate ultrasound for medical imaging [1]. We also proposed a simple and fast method of determining an f-number, which is based on the directivity of the transducer elements. In their comment, Martin F. Schiffner and Georg Schmitz argue that we mistakenly link image quality enhancement to the reduction of measurement noise. They disapprove our proposed f-number, claiming it deteriorates the signal-to-noise ratio (SNR). Based on their previous work [2], they also highlight that the f-number should be derived from the grating lobe angles. In this reply, we explain their error in the SNR argument. We also illustrate the potential drawbacks of exclusively relying on grating lobes to establish an f-number with a DAS, suggesting that alternative approaches might be worthy of consideration.

Efficient Stratified 3-D Scatterer Sampling for Freehand Ultrasound Simulation.

Ultrasound image simulation is a well-explored field with the main objective of generating realistic synthetic images, further used as ground truth for computational imaging algorithms or for radiologists' training. Several ultrasound simulators are already available, most of them consisting in similar steps: 1) generate a collection of tissue mimicking individual scatterers with random spatial positions and random amplitudes; 2) model the ultrasound probe and the emission and reception schemes; and 3) generate the radio frequency (RF) signals resulting from the interaction between the scatterers and the propagating ultrasound waves. This article is focused on the first step. To ensure fully developed speckle, a few tens of scatterers by resolution cell are needed, demanding to handle high amounts of data (especially in 3-D) and resulting into important computational time. The objective of this work is to explore new scatterer spatial distributions, with application to multiple coherent 2-D slice simulations from 3-D volumes. More precisely, lazy evaluation of pseudorandom schemes proves them to be highly computationally efficient compared with uniform random distribution commonly used. We also propose an end-to-end method from the 3-D tissue volume to resulting ultrasound images using coherent and 3-D-aware scatterer generation and usage in a real-time context.

Ultrafast Cardiac Imaging Using Deep Learning for Speckle-Tracking Echocardiography.

High-quality ultrafast ultrasound imaging is based on coherent compounding from multiple transmissions of plane waves (PW) or diverging waves (DW). However, compounding results in reduced frame rate, as well as destructive interferences from high-velocity tissue motion if motion compensation (MoCo) is not considered. While many studies have recently shown the interest of deep learning for the reconstruction of high-quality static images from PW or DW, its ability to achieve such performance while maintaining the capability of tracking cardiac motion has yet to be assessed. In this article, we addressed such issue by deploying a complex-weighted convolutional neural network (CNN) for image reconstruction and a state-of-the-art speckle-tracking method. The evaluation of this approach was first performed by designing an adapted simulation framework, which provides specific reference data, i.e., high-quality, motion artifact-free cardiac images. The obtained results showed that, while using only three DWs as input, the CNN-based approach yielded an image quality and a motion accuracy equivalent to those obtained by compounding 31 DWs free of motion artifacts. The performance was then further evaluated on nonsimulated, experimental in vitro data, using a spinning disk phantom. This experiment demonstrated that our approach yielded high-quality image reconstruction and motion estimation, under a large range of velocities and outperforms a state-of-the-art MoCo-based approach at high velocities. Our method was finally assessed on in vivo datasets and showed consistent improvement in image quality and motion estimation compared to standard compounding. This demonstrates the feasibility and effectiveness of deep learning reconstruction for ultrafast speckle-tracking echocardiography.

Phase Unwrapping of Color Doppler Echocardiography Using Deep Learning.

Color Doppler echocardiography is a widely used noninvasive imaging modality that provides real-time information about intracardiac blood flow. In an apical long-axis view of the left ventricle, color Doppler is subject to phase wrapping, or aliasing, especially during cardiac filling and ejection. When setting up quantitative methods based on color Doppler, it is necessary to correct this wrapping artifact. We developed an unfolded primal-dual network (PDNet) to unwrap (dealias) color Doppler echocardiographic images and compared its effectiveness against two state-of-the-art segmentation approaches based on nnU-Net and transformer models. We trained and evaluated the performance of each method on an in-house dataset and found that the nnU-Net-based method provided the best dealiased results, followed by the primal-dual approach and the transformer-based technique. Noteworthy, the PDNet, which had significantly fewer trainable parameters, performed competitively with respect to the other two methods, demonstrating the high potential of deep unfolding methods. Our results suggest that deep learning (DL)-based methods can effectively remove aliasing artifacts in color Doppler echocardiographic images, outperforming DeAN, a state-of-the-art semiautomatic technique. Overall, our results show that DL-based methods have the potential to effectively preprocess color Doppler images for downstream quantitative analysis.

An extensive survey of phytoviral RNA 3' uridylation identifies extreme variations and virus-specific patterns.

Viral RNAs can be uridylated in eukaryotic hosts. However, our knowledge of uridylation patterns and roles remains rudimentary for phytoviruses. Here, we report global 3' terminal RNA uridylation profiles for representatives of the main families of positive single-stranded RNA phytoviruses. We detected uridylation in all 47 viral RNAs investigated here, revealing its prevalence. Yet, uridylation levels of viral RNAs varied from 0.2% to 90%. Unexpectedly, most poly(A) tails of grapevine fanleaf virus (GFLV) RNAs, including encapsidated tails, were strictly monouridylated, which corresponds to an unidentified type of viral genomic RNA extremity. This monouridylation appears beneficial for GFLV because it became dominant when plants were infected with nonuridylated GFLV transcripts. We found that GFLV RNA monouridylation is independent of the known terminal uridylyltransferases (TUTases) HEN1 SUPPRESSOR 1 (HESO1) and UTP:RNA URIDYLYLTRANSFERASE 1 (URT1) in Arabidopsis (Arabidopsis thaliana). By contrast, both TUTases can uridylate other viral RNAs like turnip crinkle virus (TCV) and turnip mosaic virus (TuMV) RNAs. Interestingly, TCV and TuMV degradation intermediates were differentially uridylated by HESO1 and URT1. Although the lack of both TUTases did not prevent viral infection, we detected degradation intermediates of TCV RNA at higher levels in an Arabidopsis heso1 urt1 mutant, suggesting that uridylation participates in clearing viral RNA. Collectively, our work unveils an extreme diversity of uridylation patterns across phytoviruses and constitutes a valuable resource to further decipher pro- and antiviral roles of uridylation.

Arabidopsis RNA processing body components LSM1 and DCP5 aid in the evasion of translational repression during Cauliflower mosaic virus infection.

Viral infections impose extraordinary RNA stress, triggering cellular RNA surveillance pathways such as RNA decapping, nonsense-mediated decay, and RNA silencing. Viruses need to maneuver among these pathways to establish infection and succeed in producing high amounts of viral proteins. Processing bodies (PBs) are integral to RNA triage in eukaryotic cells, with several distinct RNA quality control pathways converging for selective RNA regulation. In this study, we investigated the role of Arabidopsis thaliana PBs during Cauliflower mosaic virus (CaMV) infection. We found that several PB components are co-opted into viral factories that support virus multiplication. This pro-viral role was not associated with RNA decay pathways but instead, we established that PB components are helpers in viral RNA translation. While CaMV is normally resilient to RNA silencing, dysfunctions in PB components expose the virus to this pathway, which is similar to previous observations for transgenes. Transgenes, however, undergo RNA quality control-dependent RNA degradation and transcriptional silencing, whereas CaMV RNA remains stable but becomes translationally repressed through decreased ribosome association, revealing a unique dependence among PBs, RNA silencing, and translational repression. Together, our study shows that PB components are co-opted by the virus to maintain efficient translation, a mechanism not associated with canonical PB functions.

SIMUS: An open-source simulator for medical ultrasound imaging. Part II: Comparison with four simulators.

BACKGROUND AND OBJECTIVE: Computational ultrasound imaging has become a well-established methodology in the ultrasound community. In the accompanying paper (part I), we described a new ultrasound simulator (SIMUS) for MATLAB, which belongs to the Matlab UltraSound Toolbox (MUST). SIMUS can generate pressure fields and radiofrequency RF signals for simulations in medical ultrasound imaging. It works in a harmonic domain and uses far-field and paraxial linear equations. METHODS: In this article (part II), we illustrate how SIMUS compares with other ultrasound simulators (Field II, k-Wave, FOCUS, and Verasonics) for a homogeneous medium. We designed different transmit sequences (focused, planar, and diverging wavefronts) and calculated the corresponding 2-D and 3-D (with elevation focusing) RMS pressure fields. RESULTS: SIMUS produced pressure fields similar to those of Field II, FOCUS, and k-Wave. The acoustic fields provided by the Verasonics simulator were significantly different from those of SIMUS and k-Wave, although the overall appearance remained consistent. CONCLUSION: Our simulations tend to demonstrate that SIMUS is reliable and can be used for realistic medical ultrasound simulations.

Full-volume three-component intraventricular vector flow mapping by triplane color Doppler.

Objective. Intraventricular vector flow mapping (iVFM) is a velocimetric technique for retrieving two-dimensional velocity vector fields of blood flow in the left ventricular cavity. This method is based on conventional color Doppler imaging, which makesiVFM compatible with the clinical setting. We have generalized theiVFM for a three-dimensional reconstruction (3D-iVFM).Approach.3D-iVFM is able to recover three-component velocity vector fields in a full intraventricular volume by using a clinical echocardiographic triplane mode. The 3D-iVFM problem was written in the spherical (radial, polar, azimuthal) coordinate system associated to the six half-planes produced by the triplane mode. As with the 2D version, the method is based on the mass conservation, and free-slip boundary conditions on the endocardial wall. These mechanical constraints were imposed in a least-squares minimization problem that was solved through the method of Lagrange multipliers. We validated 3D-iVFMin silicoin a patient-specific CFD (computational fluid dynamics) model of cardiac flow and tested its clinical feasibilityin vivoin patients and in one volunteer.Main results.The radial and polar components of the velocity were recovered satisfactorily in the CFD setup (correlation coefficients,r = 0.99 and 0.78). The azimuthal components were estimated with larger errors (r = 0.57) as only six samples were available in this direction. In bothin silicoandin vivoinvestigations, the dynamics of the intraventricular vortex that forms during diastole was deciphered by 3D-iVFM. In particular, the CFD results showed that the mean vorticity can be estimated accurately by 3D-iVFM.Significance. Our results tend to indicate that 3D-iVFM could provide full-volume echocardiographic information on left intraventricular hemodynamics from the clinical modality of triplane color Doppler.

SIMUS: An open-source simulator for medical ultrasound imaging. Part I: Theory & examples.

BACKGROUND AND OBJECTIVE: Computational ultrasound imaging has become a well-established methodology in the ultrasound community. Simulations of ultrasound sequences and images allow the study of innovative techniques in terms of emission strategy, beamforming, and probe design. There is a wide spectrum of software dedicated to ultrasound imaging, each having its specificities in its applications and the numerical method. METHODS: We describe in this two-part paper a new ultrasound simulator (SIMUS) for MATLAB, which belongs to the MATLAB UltraSound Toolbox (MUST). The SIMUS software simulates acoustic pressure fields and radiofrequency RF signals for uniform linear or convex probes. SIMUS is an open-source software whose features are 1) ease of use, 2) time-harmonic analysis, 3) pedagogy. The main goal was to offer a comprehensive turnkey tool, along with a detailed theory for pedagogical and research purposes. RESULTS: This article describes in detail the underlying linear theory of SIMUS and provides examples of simulated acoustic fields and ultrasound images. The accompanying article (part II) is devoted to the comparison of SIMUS with several software packages: Field II, k-Wave, FOCUS, and the Verasonics simulator. The MATLAB open codes for the simulator SIMUS are distributed under the terms of the GNU Lesser General Public License, and can be downloaded from https://www.biomecardio.com/MUST. CONCLUSIONS: The simulations described in this part and in the accompanying paper (Part II) show that SIMUS can be used for realistic simulations in medical ultrasound imaging.

Motion Estimation by Deep Learning in 2D Echocardiography: Synthetic Dataset and Validation.

Motion estimation in echocardiography plays an important role in the characterization of cardiac function, allowing the computation of myocardial deformation indices. However, there exist limitations in clinical practice, particularly with regard to the accuracy and robustness of measurements extracted from images. We therefore propose a novel deep learning solution for motion estimation in echocardiography. Our network corresponds to a modified version of PWC-Net which achieves high performance on ultrasound sequences. In parallel, we designed a novel simulation pipeline allowing the generation of a large amount of realistic B-mode sequences. These synthetic data, together with strategies during training and inference, were used to improve the performance of our deep learning solution, which achieved an average endpoint error of 0.07 ± 0.06 mm per frame and 1.20 ± 0.67 mm between ED and ES on our simulated dataset. The performance of our method was further investigated on 30 patients from a publicly available clinical dataset acquired from a GE system. The method showed promise by achieving a mean absolute error of the global longitudinal strain of 2.5 ± 2.1% and a correlation of 0.77 compared to GLS derived from manual segmentation, much better than one of the most efficient methods in the state-of-the-art (namely the FFT-Xcorr block-matching method). We finally evaluated our method on an auxiliary dataset including 30 patients from another center and acquired with a different system. Comparable results were achieved, illustrating the ability of our method to maintain high performance regardless of the echocardiographic data processed.

A Pipeline for the Generation of Synthetic Cardiac Color Doppler.

Color Doppler imaging (CDI) is the modality of choice for simultaneous visualization of myocardium and intracavitary flow over a wide scan area. This visualization modality is subject to several sources of error, the main ones being aliasing and clutter. Mitigation of these artifacts is a major concern for better analysis of intracardiac flow. One option to address these issues is through simulations. In this article, we present a numerical framework for generating clinical-like CDI. Synthetic blood vector fields were obtained from a patient-specific computational fluid dynamics CFD model. Realistic texture and clutter artifacts were simulated from real clinical ultrasound cineloops. We simulated several scenarios highlighting the effects of 1) flow acceleration; 2) wall clutter; and 3) transmit wavefronts, on Doppler velocities. As a comparison, an "ideal" color Doppler was also simulated, without these harmful effects. This synthetic dataset is made publicly available and can be used to evaluate the quality of Doppler estimation techniques. Besides, this approach can be seen as a first step toward the generation of comprehensive datasets for training neural networks to improve the quality of Doppler imaging.

SCORE underestimates cardiovascular mortality in hypertension: insight from the OLD-HTA and NEW-HTA Lyon cohorts.

AIMS: Current European guidelines recommend the SCORE to estimate 10-year cardiovascular mortality in patients with moderate/low cardiovascular risk. SCORE was derived from the general population. The objective of this study was to investigate the estimated 10-year cardiovascular mortality according to the SCORE in a historic and a contemporary cohort of hypertensive patients. METHODS AND RESULTS: After exclusion of secondary prevention and diabetes, 3086 patients were analysed in the OLD-HTA (1969-90) and 1081 in the NEW-HTA (1997-2014) Lyon cohorts. SCORE was calculated using the low and high cardiovascular risk equations and charts, and patients classified as being at low (0%), moderate (1-4%), high (5-9%), and very high (≥10%) risk. In the OLD-HTA cohort, 10-year cardiovascular mortality was higher (1.2%, 5.5%, 17.7%, and 27.0%) than that predicted by the low-risk equation (0%, 1.7%, 6.4%, and 14.8%). In the NEW-HTA cohort, similar results were observed (1.1%, 4.7%, 15.1%, and 15.2% vs. 0%, 1.9%, 6.2%, and 11.7%, respectively). Using the high-risk equation, mortality was underestimated in both cohorts, but the difference was smaller. The diagnostic performance of the high-risk equation was lower than the low-risk equation in both cohorts, considering the SCORE as a continuous or a categorical variable (Likelihood ratio test P < 0.05 for all comparisons in OLD-HTA). Similar results were obtained using SCORE charts. CONCLUSION: SCORE underestimates the 10-year cardiovascular mortality risk in hypertensive patients in a historic cohort and in a contemporary one. The algorithm to predict cardiovascular mortality in hypertensive patients needs an update given new information since its creation.

Complex Convolutional Neural Networks for Ultrafast Ultrasound Imaging Reconstruction From In-Phase/Quadrature Signal.

Ultrafast ultrasound imaging remains an active area of interest in the ultrasound community due to its ultrahigh frame rates. Recently, a wide variety of studies based on deep learning have sought to improve ultrafast ultrasound imaging. Most of these approaches have been performed on radio frequency (RF) signals. However, in- phase/quadrature (I/Q) digital beamformers are now widely used as low-cost strategies. In this work, we used complex convolutional neural networks for reconstruction of ultrasound images from I/Q signals. We recently described a convolutional neural network architecture called ID-Net, which exploited an inception layer designed for reconstruction of RF diverging-wave ultrasound images. In the present study, we derive the complex equivalent of this network, i.e., complex-valued inception for diverging-wave network (CID-Net) that operates on I/Q data. We provide experimental evidence that CID-Net provides the same image quality as that obtained from RF-trained convolutional neural networks, i.e., using only three I/Q images, CID-Net produces high-quality images that can compete with those obtained by coherently compounding 31 RF images. Moreover, we show that CID-Net outperforms the straightforward architecture that consists of processing real and imaginary parts of the I/Q signal separately, which thereby indicates the importance of consistently processing the I/Q signals using a network that exploits the complex nature of such signals.

A NYN domain protein directly interacts with DECAPPING1 and is required for phyllotactic pattern.

In eukaryotes, general mRNA decay requires the decapping complex. The activity of this complex depends on its catalytic subunit, DECAPPING2 (DCP2), and its interaction with decapping enhancers, including its main partner DECAPPING1 (DCP1). Here, we report that in Arabidopsis thaliana, DCP1 also interacts with a NYN domain endoribonuclease, hence named DCP1-ASSOCIATED NYN ENDORIBONUCLEASE 1 (DNE1). Interestingly, we found DNE1 predominantly associated with DCP1, but not with DCP2, and reciprocally, suggesting the existence of two distinct protein complexes. We also showed that the catalytic residues of DNE1 are required to repress the expression of mRNAs in planta upon transient expression. The overexpression of DNE1 in transgenic lines led to growth defects and a similar gene deregulation signature than inactivation of the decapping complex. Finally, the combination of dne1 and dcp2 mutations revealed a functional redundancy between DNE1 and DCP2 in controlling phyllotactic pattern formation. Our work identifies DNE1, a hitherto unknown DCP1 protein partner highly conserved in the plant kingdom and identifies its importance for developmental robustness.