Search

We analyze, considering the contribution of $P$-wave charmonia, lattice data for the $D \overline{D}$-$D_s \overline{D}_s$ coupled-channel and $D \overline{D}{}^\ast$ systems and search for hidden-charm states with $J^{PC}=0^{++}$, $1^{++}$, and $2^{++}$ quantum numbers. For $0^{++}$, the analysis reveals three poles in the $D\overline{D}$-$D_s \overline{D}_s$ coupled-channel amplitude, corresponding to three states. Two of these poles, located near the $D\overline{D}$ and $D_s \overline{D}_s$ thresholds, can be interpreted as mostly molecular states. A third pole above the $D_s \overline{D}_s$ threshold is originated from the $P$-wave $\chi_{c0}(2P)$ charmonium state. In the $1^{++}$ sector, we find two poles in the complex energy plane. The first one is related to the molecular $X(3872)$ state, with a compositeness exceeding $90\%$, while the second one, stemming from the $\chi_{c1}(2P)$ charmonium, appears above the $D\overline{D}{}^\ast$ threshold and it likely corresponds to the recently discovered $\chi_{c1}(4010)$ state. In the $2^{++}$ sector, we also report two poles and find that the dressed $\chi_{c2}(2P)$ is lighter than the $D^\ast\overline{D}{}^\ast$ molecular state, with the dynamics of the latter closely related to that of the heavy-quark spin-symmetry partner of the $X(3872)$., Comment: 23 pages, 15 figures

By solving the Lippmann-Schwinger equation, possible hadronic molecules in the $D^*\bar K$-$D\bar K^*$ coupled-channel system are investigated with the one-meson exchange potentials, where both vector and pseudoscalar mesons are considered as exchange particles. We find an S-wave virtual state with mass $M=2487$ MeV, and a resonance with $M=2759$ and width $\Gamma=18$ MeV. In the $D^* \bar K$ invariant mass distribution, the virtual state appears as a cusp at the $D^*\bar K$ threshold, while the resonance potentially manifests as a dip. In particular, we take into account the $D\bar K \pi$ three-body dynamics due to the on-shell pion exchange and the finite decay width for $\bar K^*$. Additionally, the SU(4) breaking effect in the coupling between charmed and light mesons is investigated in our work. Our results also indicate that the accurate measurement for the decay width of the $D\bar K^*$ resonance can help us to evaluate this effect in the future., Comment: 22 pages, 3 figures

With a clean environment and high collision energy, the Circular Electron Positron Collider (CEPC) would be an excellent facility for heavy flavor physics. Using the Monte Carlo event generator Pythia, we simulate the production of the charmed (bottom) hadron pairs in the electron-positron collisions at the $Z$ factory of CEPC, and the inclusive production rates for typical candidates of the hidden/double-charm and hidden/double-bottom $S$-wave hadronic molecules are estimated at an order-of-magnitude level with the final state interactions after the hadron pair production. The predicted cross sections for the hidden-charm meson-meson molecules $X(3872)$ and $Z_c(3900)$ are at $\rm{pb}$ level, which are about two to three orders of magnitude larger than the production cross sections for the double-charm meson-meson molecules $T_{cc}$ and $T_{cc}^{*}$, as the double-charmed ones require the production of two pairs of $c\bar{c}$ from the $Z$ boson decay. The production cross sections for the hidden-charm pentaquark states $P_{c}$ and $P_{cs}$ as meson-baryon molecules are a few to tens of fb, which are about one magnitude larger than those of the possible hidden-charm baryon-antibaryon and double-charm meson-baryon molecules. In the bottom sector, the production cross sections for the $Z_b$ states as $B^{(*)}\bar{B}^{*}$ molecules are about tens to hundreds of fb, indicating $10^6$ - $10^7$ events from a two-year operation of CEPC, and the expected events from the double-bottom molecules are about 2 - 5 orders of magnitude smaller than the $Z_b$ states. Our results shows great prospects of probing heavy exotic hadrons at CEPC., Comment: 24 pages, 10 figures, published version

A novel thermodynamic phenomenon has been observed in the quantum Ba\~{n}ados-Teitelboim-Zanelli (qBTZ) black hole, utilizing generalized free energy and Kramer escape rate. This phenomenon also reveals the unique property of the quantum black hole. The stochastic thermal motion of various thermodynamic states within the black hole system induces phase transitions, under the influence of generalized free energy which obtained by extending Maxwell's construction. Through the analysis of Kramer escape rate, it is discovered that the qBTZ black hole thermodynamic system exhibits a bounce effect. Furthermore, the overall thermodynamic picture of the qBTZ black hole has been obtained under different quantum backreactions., Comment: 8 pages, 3 figures

In this study, we obtain specific picture of the phase transitions for the 6-dimensional Gauss-Bonnet Anti-de Sitter (AdS) black hole with triple phases, using the generalized free energy we constructed and Kramers escape rate in stochastic motion. There are six possible phase transition processes between the three different stable states (small, medium, and large black hole states). During these phase transitions, there are two key temperatures. One is the temperature at which the medium black hole state emerges, and the other is the temperature at which the small black hole state annihilates. Meanwhile, two dynamic equilibrium processes are formed. One is a dynamic equilibrium of the transition from the medium black hole state to the large black hole state and the transition from the small black hole state to the medium black hole state. The other is a dynamic equilibrium of the transition from the small black hole state to the medium black hole state and the transition from the medium black hole state to the small black hole state., Comment: 10 pages, 4 figures

Xiu-Li Xiao,1,* Qian-Xue Yang,2,* Hui-Zhong Niu,1 Li-Jing Li,1 Jian-Lei Geng,1 Dan Zhao,1 Hui Ren,1 Pan-Pan Yang1 1Department of General Surgery I, Hebei Children’s Hospital, Shijiazhaung, 050000, People’s Republic of China; 2Department of Operating Room, Hebei Children’s Hospital, Shijiazhaung, 050000, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xiu-Li Xiao, Department of General Surgery I, Hebei Children’s Hospital, No. 133 Jianhuanan Street, Shijiazhuang, Yuhua District, 050000, People’s Republic of China, Tel +86 0311 8591 1321, Email xxlwwvt7@email.cnPurpose: This study aimed to investigate the maintenance effect of two puncture methods using non-coring needles in children with totally implantable venous access device (TIVAD).Methods: The 110 children who received TIVAD implantation for short bowel syndrome and solid tumors in our department from 2021.12 to 2022.12 were selected as the study subjects. Blinded method was used and divided into experimental group and control group according to random number table The experimental group underwent painless surround puncture method to place the needles and compound lidocaine ointment for topical anesthesia, while the control group underwent traditional puncture method to complete this operation. The effects of the two puncture methods on pain, catheter seal fluid volume, and catheter occlusion rate were evaluated using the Facial Pain Scale Revised, Behavioral Assessment Scale, and in vitro digital subtraction angiography test.Results: In the control group, the degree of puncture pain was mild in 5 patients, moderate in 19 patients, and severe in 28 patients; the amount of catheter sealing solution was 9.32 ± 1.32 mL, and the catheter occlusion rate was 25.00%. In the experimental group, the degree of puncture pain was mild in 16 patients, moderate in 22 patients, and severe in 16 patients; the amount of sealing solution was 7.66 ± 1.0

Understanding mechanistic details of the nickel-catalyzed coupling reactions of Csp3 alcohol derivatives is key to developing selective reactions of this widely prevalent functional group. In this manuscript, we utilize a combination of experimental data and DFT studies to define the key intermediates, stereochemical outcome, and competing pathways of a nickel-catalyzed cross-electrophile coupling reaction of 1,3-dimesylates. Stereospecific formation of a 1,3-diiodide intermediate is achieved in situ by the Grignard reagent. The overall stereoablative stereochemical outcome is due to a nickel-catalyzed halogen atom abstraction with a radical rebound that is slower than epimerization of the alkyl radical. Finally, lifetimes of this alkyl radical intermediate are compared to radical clocks to enhance the understanding of the lifetime of the secondary alkyl radical.

In 2021, the Belle collaboration reported the first observation of a new structure in the $\psi(2S) \gamma$ final state produced in the two-photon fusion process. In the hadronic molecule picture, this new structure can be associated with the shallow isoscalar $D^*\bar{D}^*$ bound state and as such is an excellent candidate for the spin-2 partner of the $X(3872)$ with the quantum numbers $J^{PC}=2^{++}$ conventionally named $X_2$. In this work we evaluate the electronic width of this new state and argue that its nature is sensitive to its total width, the experimental measurement currently available being unable to distinguish between different options. Our estimates demonstrate that the planned Super $\tau$-Charm Facility offers a promising opportunity to search for and study this new state in the invariant mass distributions for the final states $J/\psi\gamma$ and $\psi(2S)\gamma$., Comment: 9 pages, 1 figures. Version to appear in CPL

Pan Pan, Tao Cheng, Tianyong Han, Yu Cao Department of Emergency Medicine, West China Hospital, Sichuan University, Chengdu, People’s Republic of ChinaCorrespondence: Yu Cao, Email yuyuer@126.comBackground: Post-intubation hypotension (PIH) frequently occurs in the management of critically ill patients and is associated with prognosis. The study aimed to construct a prediction model for PIH events by analyzing risk factors in patients with severe pneumonia in the emergency department.Methods: We retrospectively enrolled 572 patients with severe pneumonia diagnosed in the emergency department of West China Hospital of Sichuan University. Five hundred patients with severe pneumonia who underwent endotracheal intubation were included in the study. All patients were randomized according to 7:3 and divided into a training cohort (n=351) and a validation cohort (n=149). Risk factors for PIH were analyzed using Least Absolute Shrinkage and Selection Operator (LASSO) and multivariable logistic regression. Calibration curves, receiver operating characteristic (ROC) curve, and decision curve analysis were applied to assess the predictive model’s fitness, discrimination, and clinical utility.Results: A total of 500 patients with severe pneumonia who underwent endotracheal intubation were enrolled in this study, and PIH occurred in 234 (46.8%) of these patients. Age, heart rate, systolic blood pressure, chronic obstructive pulmonary disease, acute physiology and chronic health evaluation II score, and induction agent use were identified as significant risk factors for the occurrence of PIH. Additionally, the body mass index was the opposite of the above. The area under the ROC curve (AUC) for the model was 0.856 (95% CI, 0.818– 0.894) in the training cohort and 0.849 (95% CI, 0.788– 0.910) in the validation cohort. The nomogram model was validated and demonstrated good calibration and high net clinical benefit. Finally, to facilitate application by clinicians, an online

In 2021, the Belle collaboration reported the first observation of a new structure in the $\psi(2S) \gamma$ final state produced in the two-photon fusion process. In the hadronic molecule picture, this new structure can be associated with the shallow isoscalar $D^*\bar{D}^*$ bound state and as such is an excellent candidate for the spin-2 partner of the $X(3872)$ with the quantum numbers $J^{PC}=2^{++}$ conventionally named $X_2$. In this work we evaluate the electronic width of this new state and argue that its nature is sensitive to its total width, the experimental measurement currently available being unable to distinguish between different options. Our estimates demonstrate that the planned Super $\tau$-Charm Facility offers a promising opportunity to search for and study this new state in the invariant mass distributions for the final states $J/\psi\gamma$ and $\psi(2S)\gamma$., Comment: 9 pages, 1 figures. Version to appear in CPL

We developed dysarthric speech intelligibility classifiers on 551,176 disordered speech samples contributed by a diverse set of 468 speakers, with a range of self-reported speaking disorders and rated for their overall intelligibility on a five-point scale. We trained three models following different deep learning approaches and evaluated them on ~94K utterances from 100 speakers. We further found the models to generalize well (without further training) on the TORGO database (100% accuracy), UASpeech (0.93 correlation), ALS-TDI PMP (0.81 AUC) datasets as well as on a dataset of realistic unprompted speech we gathered (106 dysarthric and 76 control speakers,~2300 samples)., Comment: ICASSP 2023 camera-ready

It has been generally expected that the $X(3872)$ has a spin-2 partner, $X_2$, with quantum numbers $J^{PC}=2^{++}$. In the hadronic molecular model, its mass was predicted to be below the $D^*\bar D^*$ threshold, and the new structure reported in the $\gamma \psi(2S)$ invariant mass distribution by the Belle Collaboration with mass $M= (4014.3 \pm 4.0 \pm 1.5)$ MeV and decay width $\Gamma= (4 \pm 11 \pm 6)$ MeV, with a global significance of 2.8 $\sigma$, is a nice candidate for it. We consider the radiative decay widths for the $X_{2}\to \gamma\psi$ with $\psi=J/\psi, \psi(2S)$ treating the $X_2$ as a $D^*\bar{D}^*$ shallow bound state, and estimate the events of $X_2$ in two-photon collisions that can be collected in the $\gamma J/\psi\to\gamma\ell^+\ell^-$ ($\ell=e,\mu$) final states at Belle. Based on the upper limit for the ratio of decay widths of $X(3872)\to \gamma \psi(2S)$ and $X(3872)\to \gamma J/\psi$ measured by BESIII, we predict the similar ratio $\Gamma(X_2\to \gamma \psi(2S))/\Gamma(X_2\to \gamma J/\psi)$ to be smaller than $1.0$. We suggest searching for the $X_2$ signal in the $\gamma J/\psi$ invariant mass distribution via two-photon fusions. The results will lead to insights into both the $X(3872)$ and the new structure observed by Belle., Comment: 21 pages, 3 figures. Version to appear in PLB

In this work, a series of butterfly-like isomers named oxacalix[2]naphthalene[2]pyrazine (ONP) were conveniently synthesized by a one-step catalyst-free reaction in a facile manner, and they exhibit typical characteristics of aggregation-induced emission (AIE). The mechanism study shows that restriction of intramolecular vibration (RIV) is the reason for their AIE properties. The pyrazine groups endow ONP molecules with good coordination ability, which makes them ideal ligands for constructing metal-organic frameworks (MOFs). Thus, three ONP-based luminescent MOFs were constructed, and they exhibit intense emission with lifetimes in the order of microseconds. More importantly, different ONP isomers have different binding capacities, and thus only one kind of MOF can be obtained even when using an isomer mixture of ONP ligands. This result suggested that the conformation of ONPs is an important determining factor for their application as bridging ligands. This work not only reports a series of new RIV-type AIEgens, but also offers a new platform for the construction of luminescent MOFs. © 2023 The Royal Society of Chemistry.

Starting from the hypothesis that the $T_{cc}^+$ discovered at LHCb is a $D^{\ast+} D^0/D^{\ast 0}D^+$ hadronic molecule, we consider the partial width of its heavy quark spin partner, the $T_{cc}^{\ast +}$ as a $D^{\ast +} D^{\ast 0}$ shallow bound state, decaying into the $D^{\ast}D\pi$ final states including the contributions of the $D^{\ast} D$ and $D^{\ast} \pi$ final state interaction by using a nonrelativistic effective field theory. Because of the existence of the $T_{cc}^+$ pole, the $I=0$ $D^{\ast} D$ rescattering can give a sizeable correction up to about $40\%$ to the decay widths considering only the tree diagrams, and the $D^{\ast} \pi$ rescattering correction is about $10\%$. The four-body partial widths of the $T_{cc}^{*+}$ into $D D\pi\pi$ are also explicitly calculated, and we find that the interference effect between different intermediate $D^*D\pi$ states is small. The total width of the $T_{cc}^{*+}$ is predicted to be about 41 keV., Comment: 31 pages, 10 figues. Results updated and extended, version to be published in Phys. Rev. D

The semi-inclusive electroproduction of exotic hadrons, including the $T_{cc}$, $P_{cs}$, and hidden-charm baryon-antibaryon states, is explored under the assumption that they are $S$-wave hadronic molecules of a pair of charmed hadrons. We employ the Monte Carlo event generator Pythia to produce the hadron pairs and then bind them together to form hadronic molecules. With the use of such a production mechanism, the semi-inclusive electroproduction rates are estimated at the order-of-magnitude level. Our results indicate that a larger number of $P_{cs}$ states and $\Lambda_c\bar{\Lambda}_c$ molecules can be produced at the proposed electron-ion colliders in China (EicC) and in the US (EIC). The results also suggest that the $T_{cc}$ states and other hidden-charm baryon-antibaryon states can be searched for at EIC. Besides, the potential 24-GeV upgrade of the Continuous Beam Accelerator Facility at the Thomas Jefferson National Accelerator Facility can play an important role in the search for the hidden-charm tetraquark and pentaquark states due to its high luminosity., Comment: 20 pages, 4 figures. Version to appear in PRD

Tianyong Han,1,* Tao Cheng,1,* Ye Liao,2 Yarong He,1 Bofu Liu,1 Qiang Lai,1 Pan Pan,1 Junzhao Liu,1 Chenxi Lei,1 Yu Cao1 1Emergency Department, West China Hospital of Sichuan University, Chengdu, Sichuan, People’s Republic of China; 2Medical Intensive Care Unit, West China Hospital of Sichuan University, Chengdu, Sichuan, People’s Republic of China*These authors contributed equally to this workCorrespondence: Yu Cao, Emergency Department, West China Hospital of Sichuan University, 37 Guoxue Lane, Wuhou District, Chengdu, 610041, Sichuan, People’s Republic of China, Tel +86 28-85422288, Email caoyu@wchscu.cnBackground: Acute pancreatitis (AP) is a multifactorial disease that is associated with substantial morbidity and mortality. Thrombosis and inflammation are involved in the development and progression of AP.Aim: To develop and validate a novel and simple scoring system for predicting 28-day adverse outcomes in AP patients based on a thrombotic and an inflammatory biomarker.Methods: A single-center, retrospective cohort study was used to establish the new scoring system (thrombo-inflammatory prognostic score; TIPS), and another study was used to verify it. The study end points were 28-day mortality, requirement for mechanical ventilation (MV), persistent organ failure (POF), and admission to the intensive care unit (AICU). Receiver operating characteristic (ROC) curves was drawn to validate the predictive value of the TIPS. The performance of the TIPS was compared with that of conventional predictive scoring systems. Logistic regression models were used to investigate the relationship between the TIPS and the different end points.Results: Among 440 patients with AP in the derivation group, 27 patients died within the 28-day follow-up period. Prothrombin time (PT) and interleukin-6 (IL-6) were used to calculate the TIPS. The TIPS (AUC=0.843) showed a performance comparable to that of the more established APACHE II (AUC=0.841)

Although personalized automatic speech recognition (ASR) models have recently been designed to recognize even severely impaired speech, model performance may degrade over time for persons with degenerating speech. The aims of this study were to (1) analyze the change of performance of ASR over time in individuals with degrading speech, and (2) explore mitigation strategies to optimize recognition throughout disease progression. Speech was recorded by four individuals with degrading speech due to amyotrophic lateral sclerosis (ALS). Word error rates (WER) across recording sessions were computed for three ASR models: Unadapted Speaker Independent (U-SI), Adapted Speaker Independent (A-SI), and Adapted Speaker Dependent (A-SD or personalized). The performance of all three models degraded significantly over time as speech became more impaired, but the performance of the A-SD model improved markedly when it was updated with recordings from the severe stages of speech progression. Recording additional utterances early in the disease before speech degraded significantly did not improve the performance of A-SD models. Overall, our findings emphasize the importance of continuous recording (and model retraining) when providing personalized models for individuals with progressive speech impairments., Comment: Submitted to ICASSP 2023

Decentralized optimization is effective to save communication in large-scale machine learning. Although numerous algorithms have been proposed with theoretical guarantees and empirical successes, the performance limits in decentralized optimization, especially the influence of network topology and its associated weight matrix on the optimal convergence rate, have not been fully understood. While (Lu and Sa, 2021) have recently provided an optimal rate for non-convex stochastic decentralized optimization with weight matrices defined over linear graphs, the optimal rate with general weight matrices remains unclear. This paper revisits non-convex stochastic decentralized optimization and establishes an optimal convergence rate with general weight matrices. In addition, we also establish the optimal rate when non-convex loss functions further satisfy the Polyak-Lojasiewicz (PL) condition. Following existing lines of analysis in literature cannot achieve these results. Instead, we leverage the Ring-Lattice graph to admit general weight matrices while maintaining the optimal relation between the graph diameter and weight matrix connectivity. Lastly, we develop a new decentralized algorithm to nearly attain the above two optimal rates under additional mild conditions.

The semi-inclusive electroproduction of exotic hadrons, including the $T_{cc}$, $P_{cs}$, and hidden-charm baryon-antibaryon states, is explored under the assumption that they are $S$-wave hadronic molecules of a pair of charmed hadrons. We employ the Monte Carlo event generator Pythia to produce the hadron pairs and then bind them together to form hadronic molecules. With the use of such a production mechanism, the semi-inclusive electroproduction rates are estimated at the order-of-magnitude level. Our results indicate that a larger number of $P_{cs}$ states and $\Lambda_c\bar{\Lambda}_c$ molecules can be produced at the proposed electron-ion colliders in China (EicC) and in the US (EIC). The results also suggest that the $T_{cc}$ states and other hidden-charm baryon-antibaryon states can be searched for at EIC. Besides, the potential 24-GeV upgrade of the Continuous Beam Accelerator Facility at the Thomas Jefferson National Accelerator Facility can play an important role in the search for the hidden-charm tetraquark and pentaquark states due to its high luminosity., Comment: 20 pages, 4 figures. Version to appear in PRD

Current methods for 3D hand pose estimation fail to generalize well to in-the-wild new scenarios due to varying camera viewpoints, self-occlusions, and complex environments. To address this problem, we propose CycleHand to improve the generalization ability of the model in a self-supervised manner. Our motivation is based on an observation: if one globally rotates the whole hand and reversely rotates it back, the estimated 3D poses of fingers should keep consistent before and after the rotation because the wrist-relative hand poses stay unchanged during global 3D rotation. Hence, we propose arbitrary-rotation self-supervised consistency learning to improve the model's robustness for varying viewpoints. Another innovation of CycleHand is that we propose a high-fidelity texture map to render the photorealistic rotated hand with different lighting conditions, backgrounds, and skin tones to further enhance the effectiveness of our self-supervised task. To reduce the potential negative effects brought by the domain shift of synthetic images, we use the idea of contrastive learning to learn a synthetic-real consistent feature extractor in extracting domain-irrelevant hand representations. Experiments show that CycleHand can largely improve the hand pose estimation performance in both canonical datasets and real-world applications.

Xiao-Dan Wu,1 Fan-Fang Zeng,1 Xiao-Xuan Yu,1 Pan-Pan Yang,1 Jun-Peng Wu,1 Ping Xv,2 Hai-Tang Wang,1 You-Ming Pei1 1Department of Anesthesiology, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, 510515, People’s Republic of China; 2Department of Anesthesiology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, 510280, People’s Republic of ChinaCorrespondence: Hai-Tang Wang; You-Ming Pei, Department of Anesthesiology, Nanfang Hospital, Southern medical university, 1838 Guangzhou North Road, Guangzhou, Guangdong, 510515, People’s Republic of China, Tel +86-18718911653 ; +86-15889942610, Fax +86 20-62787271 ; +86 20-62787271, Email wanghaitang8888@163.com; 71119099@qq.comPurpose: Chronic post-surgical pain (CPSP) is one of the adverse outcomes after surgery, especially in thoracotomy. However, the prevalence of CPSP in elderly adults (≥ 65 years), is still limited. Therefore, the present study was undertaken to establish and validate the prediction model of CPSP in those patients after thoracic surgery, including thoracotomy and video-assisted thoracoscopic surgery.Patients and Methods: This retrospective, observational single-center cohort study was conducted in Nanfang Hospital, Southern Medical University, which randomly and consecutively collected 577 elderly patients who underwent thoracic surgery between January 1, 2017, and December 31, 2020. According to the Akaike information criterion, the prediction model was built based on all the data and was validated by calibration with 500 bootstrap samples.Results: The mean age of participants was 69.09± 3.80 years old, and 63.1% were male. The prevalence of CPSP was 26.9%. Age more than 75 years, BMI, blood loss, longer length of hospital stays, and higher pre-operative neutrophil count were associated with CPSP. Except for these factors, we incorporated history of drinking to build up the prediction model. The areas under the curve (AUCs) of the prediction models were 0

Space-time memory (STM) based video object segmentation (VOS) networks usually keep increasing memory bank every several frames, which shows excellent performance. However, 1) the hardware cannot withstand the ever-increasing memory requirements as the video length increases. 2) Storing lots of information inevitably introduces lots of noise, which is not conducive to reading the most important information from the memory bank. In this paper, we propose a Recurrent Dynamic Embedding (RDE) to build a memory bank of constant size. Specifically, we explicitly generate and update RDE by the proposed Spatio-temporal Aggregation Module (SAM), which exploits the cue of historical information. To avoid error accumulation owing to the recurrent usage of SAM, we propose an unbiased guidance loss during the training stage, which makes SAM more robust in long videos. Moreover, the predicted masks in the memory bank are inaccurate due to the inaccurate network inference, which affects the segmentation of the query frame. To address this problem, we design a novel self-correction strategy so that the network can repair the embeddings of masks with different qualities in the memory bank. Extensive experiments show our method achieves the best tradeoff between performance and speed. Code is available at https://github.com/Limingxing00/RDE-VOS-CVPR2022., Comment: Accepted by CVPR 2022

3D-aware image synthesis aims to generate images of objects from multiple views by learning a 3D representation. However, one key challenge remains: existing approaches lack geometry constraints, hence usually fail to generate multi-view consistent images. To address this challenge, we propose Multi-View Consistent Generative Adversarial Networks (MVCGAN) for high-quality 3D-aware image synthesis with geometry constraints. By leveraging the underlying 3D geometry information of generated images, i.e., depth and camera transformation matrix, we explicitly establish stereo correspondence between views to perform multi-view joint optimization. In particular, we enforce the photometric consistency between pairs of views and integrate a stereo mixup mechanism into the training process, encouraging the model to reason about the correct 3D shape. Besides, we design a two-stage training strategy with feature-level multi-view joint optimization to improve the image quality. Extensive experiments on three datasets demonstrate that MVCGAN achieves the state-of-the-art performance for 3D-aware image synthesis., Comment: Accepted by CVPR 2022

Temporal representation is the cornerstone of modern action detection techniques. State-of-the-art methods mostly rely on a dense anchoring scheme, where anchors are sampled uniformly over the temporal domain with a discretized grid, and then regress the accurate boundaries. In this paper, we revisit this foundational stage and introduce Recurrent Continuous Localization (RCL), which learns a fully continuous anchoring representation. Specifically, the proposed representation builds upon an explicit model conditioned with video embeddings and temporal coordinates, which ensure the capability of detecting segments with arbitrary length. To optimize the continuous representation, we develop an effective scale-invariant sampling strategy and recurrently refine the prediction in subsequent iterations. Our continuous anchoring scheme is fully differentiable, allowing to be seamlessly integrated into existing detectors, e.g., BMN and G-TAD. Extensive experiments on two benchmarks demonstrate that our continuous representation steadily surpasses other discretized counterparts by ~2% mAP. As a result, RCL achieves 52.92% mAP@0.5 on THUMOS14 and 37.65% mAP on ActivtiyNet v1.3, outperforming all existing single-model detectors., Comment: 9 pages, 7 figures, CVPR2022

Cross-modality interaction is a critical component in Text-Video Retrieval (TVR), yet there has been little examination of how different influencing factors for computing interaction affect performance. This paper first studies the interaction paradigm in depth, where we find that its computation can be split into two terms, the interaction contents at different granularity and the matching function to distinguish pairs with the same semantics. We also observe that the single-vector representation and implicit intensive function substantially hinder the optimization. Based on these findings, we propose a disentangled framework to capture a sequential and hierarchical representation. Firstly, considering the natural sequential structure in both text and video inputs, a Weighted Token-wise Interaction (WTI) module is performed to decouple the content and adaptively exploit the pair-wise correlations. This interaction can form a better disentangled manifold for sequential inputs. Secondly, we introduce a Channel DeCorrelation Regularization (CDCR) to minimize the redundancy between the components of the compared vectors, which facilitate learning a hierarchical representation. We demonstrate the effectiveness of the disentangled representation on various benchmarks, e.g., surpassing CLIP4Clip largely by +2.9%, +3.1%, +7.9%, +2.3%, +2.8% and +6.5% R@1 on the MSR-VTT, MSVD, VATEX, LSMDC, AcitivityNet, and DiDeMo, respectively., Comment: 22 pages, 11 figures, Tech report

Inspired by the combinatorial algebraic approach proposed by Dhurandhar {\it et al.}, we propose two novel classes of second-generation time-delay interferometry (TDI) solutions and their further generalization. The primary strategy of the algorithm is to enumerate specific types of residual laser frequency noise associated with second-order commutators in products of time-displacement operators. The derivations are based on analyzing the delay time residual when expanded in time derivatives of the armlengths order by order. It is observed that the solutions obtained by such a scheme are primarily captured by the geometric TDI approach and therefore possess an intuitive interpretation. Nonetheless, the fully-symmetric Sagnac and Sagnac-inspired combinations inherit the properties from the original algebraic approach, and subsequently lie outside of the scope of geometric TDI. We explicitly show that novel solutions, distinct from existing ones in terms of both algebraic structure and sensitivity curve, are encountered. Moreover, at its lowest order, the solution is furnished by commutators of relatively compact form. Besides the original Michelson-type solution, we elaborate on other types of solutions such as the Monitor, Beacon, Relay, Sagnac, fully-symmetric Sagnac, and Sagnac-inspired ones. The average response functions, residual noise power spectral density, and sensitivity curves are evaluated for the obtained solutions. Also, the relations between the present scheme and other existing algorithms are discussed., Comment: 22 pages, 4 figures

We generalize the combinatorial algebraic approach first proposed by Dhurandhar et al. to construct various classes of modified second-generation time-delay interferometry (TDI) solutions. The main idea behind the algorithm is to enumerate, in a given order, a specific type of commutator between two monomials defined by the products of particular time-displacement operators. On the one hand, the above commutators can be systematically rewritten as the elements of a left ideal, defined by the l.h.s. of the relevant equation for the TDI solution. On the other hand, these commutators are shown to vanish if we only keep up the first-order contributions regarding the rate of change of armlengths. In other words, each commutator furnishes a valid TDI solution pertaining to the given type of modified second-generation combinations. In this work, the original algorithm, which only involves time-delay operators, is extended by introducing the time-advance ones and then utilized to seek solutions of the Beacon, Relay, Monitor, Sagnac, and fully symmetric Sagnac types. We discuss the relation between the present scheme's solutions and those obtained by the geometric TDI approach, a well-known method of exhaustion of virtual optical paths. In particular, we report the results on novel Sagnac-inspired solutions that cannot be straightforwardly obtained using the geometric TDI algorithm. The average response functions, floor noise power spectral densities, and sensitivity functions are evaluated for the obtained solutions.

Tao Cheng,1,2,&ast; Tian-Yong Han,1,2,&ast; Bo-Fu Liu,1,2 Pan Pan,1,2 Qiang Lai,1,2 Haifang Yu,1,2 Yu Cao1,2 1Emergency Department, West China Hospital of Sichuan University, Chengdu, Sichuan, People’s Republic of China; 2Laboratory of Emergency Medicine, West China Hospital of Sichuan University, Chengdu, Sichuan, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Yu CaoEmergency Department of West China Hospital and Laboratory of Emergency Medicine of West China Hospital, Sichuan University, 37 Guoxue Road, Chengdu, 610041, Sichuan, People’s Republic of China, Tel +86-28-85422288, Email caoyu@wchscu.cnObjective: To develop a new approach based on Balthazar grades of acute pancreatitis (AP) and to assess this modified method as a tool for the early prediction of AP severity in the emergency department (ED).Methods: Data pertaining to AP patients ≥ 18 years old that had undergone computed tomography (CT) scanning within 24 h following ED admission between January 1, 2017 and September 30, 2017 were retrospectively analyzed. Patients were separated into two groups based on the length of time between the onset of their AP symptoms and the completion of CT scanning (Group 1: < 72 h; Group 2: ≥ 72 h). Modified Balthazar grades for these patients were then assessed, with the concordance between these modified grades and the 2012 revised Atlanta classification being assessed using the Kappa (κ) statistic. The modified grade with the largest κ value was evaluated based on performance traits including Harrell’s concordance index (C-index), area under the receiver operating characteristic curve (AUC) analyses, calibration curves, and decision curve analyses (DCA) in comparison with bedside index for severity in AP (BISAP) scores.Results: In total, 372 patients were included in the present analysis. These patients were regraded according to six methods, with the method yielding the largest κ value consist

The mass spectrum of hidden charm pentaquark states composed of two diquarks and an antiquark are calculated by use of an effective Hamiltonian which includes explicitly the spin, color, and flavor dependent interactions. The results show that the $P_c(4312)^+$ and $P_c(4440)^+$ states could be explained as hidden charm pentaquark states with isospin and spin-parity $IJ^P=1/2\left(3/2^-\right)$, the $P_c(4457)^+$ state could be explained as a hidden charm pentaquark state with $IJ^P=1/2\left(5/2^-\right)$, and the $P_{cs}(4459)^+$ state could be explained as a hidden charm pentaquark state with $IJ^P=0\left(1/2^-\right)$ or $0\left(3/2^-\right)$. Predications for the masses of other possible pentaquark states are also given, and the possible decay channels of these hidden charm pentaquark states are discussed., Comment: 9 pages

The purpose of the present study is to explore the mass spectrum of the hidden charm tetraquark states within a diquark model. Proposing that a tetraquark state is composed of a diquark and an antidiquark, the masses of all possible $[qc][\bar{q}\bar{c}]$, $[sc][\bar{s}\bar{c}]$, and $[qc][\bar{s}\bar{c}]$ $\left([sc][\bar{q}\bar{c}]\right)$ hidden charm tetraquark states are systematically calculated by use of an effective Hamiltonian, which contains color, spin, and flavor dependent interactions. Apart from the $X(3872)$, $Z(3900)$, $\chi_{c2}(3930)$, and $X(4350)$ which are taken as input to fix the model parameters, the calculated results support that the $\chi_{c0}(3860)$, $X(4020)$, $X(4050)$ are $[qc][\bar{q}\bar{c}]$ states with $I^GJ^{PC}=0^+0^{++}$, $1^+1^{+-}$, and $1^-2^{++}$, respectively, the $\chi_{c1}(4274)$ is an $[sc][\bar{s}\bar{c}]$ state with $I^GJ^{PC}=0^+1^{++}$, the $X(3940)$ is a $[qc][\bar{q}\bar{c}]$ state with $I^GJ^{PC}=1^-0^{++}$ or $1^-1^{++}$, the $Z_{cs}(3985)^-$ is an $[sc][\bar{q}\bar{c}]$ state with $J^{P}=0^{+}$ or $1^+$, and the $Z_{cs}(4000)^+$ and $Z_{cs}(4220)^+$ are $[qc][\bar{s}\bar{c}]$ states with $J^{P}=1^{+}$. Predictions for other possible tetraquark states are also given., Comment: 9 pages. Version published in Phys. Rev. D

Contains fulltext : 237259.pdf (Publisher’s version ) (Closed access)

There must be Coulomb bound states of a pair of hadrons, which are stable against the strong interaction, with opposite electric charges. Such bound states are hadronic atoms. We study the properties and the production of the ground-state $D^+D^-$ hadronic atom $A_{D^+D^-}$, called dionium, with quantum numbers $J^{PC}=0^{++}$. Using a nonrelativistic effective field theory for the $D^0\bar{D}^0$-$D^+D^-$ coupled-channel system, the mass of the ground-state dionium is predicted to be $(3739.3 \pm 0.1)~\text{MeV}$, with the binding energy reduced by about 10% compared to the Coulomb binding energy due to the strong interaction. Its width for the decay into the neutral $D^0\bar D^0$ channel is predicted to be $1.8^{+1.4}_{-0.6}$ keV using lattice inputs for the $D\bar D$ strong interaction. The cross section for the inclusive prompt production of the dionium at CMS and LHCb and that for the direct production $p\bar p\to A_{D^+D^-}$ at PANDA are estimated at an order-of-magnitude level. In particular, we expect that $\mathcal{O}(10^3\sim 10^5)$ events of the reaction chain $p\bar p\to A_{D^+D^-}\to D^0\bar D^0 \to K^-\pi^+K^+\pi^-$ can be collected at PANDA, and valuable information on the charmed meson interaction and on understanding charmoniumlike states will be obtained., Comment: 25 pages, 5 figures. Version to appear in Phys. Rev. D

The role of focal amplifications and extrachromosomal DNA (ecDNA) is unknown in gastric cardia adenocarcinoma (GCA). Here, we identify frequent focal amplifications and ecDNAs in Chinese GCA patient samples, and find focal amplifications in the GCA cohort are associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits. We observe diverse correlations between the presence of oncogene focal amplifications and prognosis, where ERBB2 focal amplifications positively correlate with prognosis and EGFR focal amplifications negatively correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients show survival probability of ERBB2 positive patients is lower than that of ERBB2 negative patients when their surviving time is under 2 years, however, the tendency is opposite when their surviving time is longer than 2 years. Our observations indicate that the ERBB2 focal amplifications may represent a good prognostic marker in GCA patients.

The role of focal amplifications and extrachromosomal DNA (ecDNA) is unknown in gastric cardia adenocarcinoma (GCA). Here, we identify frequent focal amplifications and ecDNAs in Chinese GCA patient samples, and find focal amplifications in the GCA cohort are associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits. We observe diverse correlations between the presence of oncogene focal amplifications and prognosis, where ERBB2 focal amplifications positively correlate with prognosis and EGFR focal amplifications negatively correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients show survival probability of ERBB2 positive patients is lower than that of ERBB2 negative patients when their surviving time is under 2 years, however, the tendency is opposite when their surviving time is longer than 2 years. Our observations indicate that the ERBB2 focal amplifications may represent a good prognostic marker in GCA patients.

The role of focal amplifications and extrachromosomal DNA (ecDNA) is unknown in gastric cardia adenocarcinoma (GCA). Here, we identify frequent focal amplifications and ecDNAs in Chinese GCA patient samples, and find focal amplifications in the GCA cohort are associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits. We observe diverse correlations between the presence of oncogene focal amplifications and prognosis, where ERBB2 focal amplifications positively correlate with prognosis and EGFR focal amplifications negatively correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients show survival probability of ERBB2 positive patients is lower than that of ERBB2 negative patients when their surviving time is under 2 years, however, the tendency is opposite when their surviving time is longer than 2 years. Our observations indicate that the ERBB2 focal amplifications may represent a good prognostic marker in GCA patients.

Contains fulltext : 237259.pdf (Publisher’s version ) (Closed access)

There must be Coulomb bound states of a pair of hadrons, which are stable against the strong interaction, with opposite electric charges. Such bound states are hadronic atoms. We study the properties and the production of the ground-state $D^+D^-$ hadronic atom $A_{D^+D^-}$, called dionium, with quantum numbers $J^{PC}=0^{++}$. Using a nonrelativistic effective field theory for the $D^0\bar{D}^0$-$D^+D^-$ coupled-channel system, the mass of the ground-state dionium is predicted to be $(3739.3 \pm 0.1)~\text{MeV}$, with the binding energy reduced by about 10% compared to the Coulomb binding energy due to the strong interaction. Its width for the decay into the neutral $D^0\bar D^0$ channel is predicted to be $1.8^{+1.4}_{-0.6}$ keV using lattice inputs for the $D\bar D$ strong interaction. The cross section for the inclusive prompt production of the dionium at CMS and LHCb and that for the direct production $p\bar p\to A_{D^+D^-}$ at PANDA are estimated at an order-of-magnitude level. In particular, we expect that $\mathcal{O}(10^3\sim 10^5)$ events of the reaction chain $p\bar p\to A_{D^+D^-}\to D^0\bar D^0 \to K^-\pi^+K^+\pi^-$ can be collected at PANDA, and valuable information on the charmed meson interaction and on understanding charmoniumlike states will be obtained., Comment: 25 pages, 5 figures. Version to appear in Phys. Rev. D

Contains fulltext : 237259.pdf (Publisher’s version ) (Closed access)

The purpose of the present study is to explore the mass spectrum of the hidden charm tetraquark states within a diquark model. Proposing that a tetraquark state is composed of a diquark and an antidiquark, the masses of all possible $[qc][\bar{q}\bar{c}]$, $[sc][\bar{s}\bar{c}]$, and $[qc][\bar{s}\bar{c}]$ $\left([sc][\bar{q}\bar{c}]\right)$ hidden charm tetraquark states are systematically calculated by use of an effective Hamiltonian, which contains color, spin, and flavor dependent interactions. Apart from the $X(3872)$, $Z(3900)$, $\chi_{c2}(3930)$, and $X(4350)$ which are taken as input to fix the model parameters, the calculated results support that the $\chi_{c0}(3860)$, $X(4020)$, $X(4050)$ are $[qc][\bar{q}\bar{c}]$ states with $I^GJ^{PC}=0^+0^{++}$, $1^+1^{+-}$, and $1^-2^{++}$, respectively, the $\chi_{c1}(4274)$ is an $[sc][\bar{s}\bar{c}]$ state with $I^GJ^{PC}=0^+1^{++}$, the $X(3940)$ is a $[qc][\bar{q}\bar{c}]$ state with $I^GJ^{PC}=1^-0^{++}$ or $1^-1^{++}$, the $Z_{cs}(3985)^-$ is an $[sc][\bar{q}\bar{c}]$ state with $J^{P}=0^{+}$ or $1^+$, and the $Z_{cs}(4000)^+$ and $Z_{cs}(4220)^+$ are $[qc][\bar{s}\bar{c}]$ states with $J^{P}=1^{+}$. Predictions for other possible tetraquark states are also given., Comment: 9 pages. Version published in Phys. Rev. D

The mass spectrum of hidden charm pentaquark states composed of two diquarks and an antiquark are calculated by use of an effective Hamiltonian which includes explicitly the spin, color, and flavor dependent interactions. The results show that the $P_c(4312)^+$ and $P_c(4440)^+$ states could be explained as hidden charm pentaquark states with isospin and spin-parity $IJ^P=1/2\left(3/2^-\right)$, the $P_c(4457)^+$ state could be explained as a hidden charm pentaquark state with $IJ^P=1/2\left(5/2^-\right)$, and the $P_{cs}(4459)^+$ state could be explained as a hidden charm pentaquark state with $IJ^P=0\left(1/2^-\right)$ or $0\left(3/2^-\right)$. Predications for the masses of other possible pentaquark states are also given, and the possible decay channels of these hidden charm pentaquark states are discussed., Comment: 9 pages

The inhibitory effect of three degraded sesquiterpene lactones, iso-seco-tanapartholide, arteludooicinolide A and millifolide A isolated from Achillea millefolium L., on anti-human lung cancer cells was examined using MTT and reporter gene assays. Millifolide A has significant inhibitory effects on the proliferation of human lung cancer cells probably through inducing cell apoptosis.

The role of focal amplifications and extrachromosomal DNA (ecDNA) is unknown in gastric cardia adenocarcinoma (GCA). Here, we identify frequent focal amplifications and ecDNAs in Chinese GCA patient samples, and find focal amplifications in the GCA cohort are associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits. We observe diverse correlations between the presence of oncogene focal amplifications and prognosis, where ERBB2 focal amplifications positively correlate with prognosis and EGFR focal amplifications negatively correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients show survival probability of ERBB2 positive patients is lower than that of ERBB2 negative patients when their surviving time is under 2 years, however, the tendency is opposite when their surviving time is longer than 2 years. Our observations indicate that the ERBB2 focal amplifications may represent a good prognostic marker in GCA patients.

Yi Zheng,1 Pan-Pan Ye,2 Yue Zhou,1 Su-Ying Wu,3 Xi-Ting Liu,1 Bin Du,1 Bo-Hao Tang,1 Min Kan,1 Ai-Qing Nie,1 Rui Yin,1 Meng Wang,1 Guo-Xiang Hao,1 Lin-Lin Song,2 Xin-Mei Yang,2 Xin Huang,2 Le-Qun Su,2 Wen-Qi Wang,2 John van den Anker,4– 6 Wei Zhao1,2 1Department of Clinical Pharmacy, School of Pharmaceutical Sciences, Cheeloo College of Medicine, Shandong University, Jinan, 250012, People’s Republic of China; 2Department of Clinical Pharmacy, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Engineering and Technology Research Center for Pediatric Drug Development, Shandong Medicine and Health Key Laboratory of Clinical Pharmacy, Jinan, 250014, People’s Republic of China; 3Shandong University of Traditional Chinese Medicine, Jinan, People’s Republic of China; 4Division of Clinical Pharmacology, Children’s National Hospital, Washington, DC, USA; 5Departments of Pediatrics, Pharmacology & Physiology, Genomics & Precision Medicine, George Washington University, School of Medicine and Health Sciences, Washington, DC, USA; 6Department of Paediatric Pharmacology and Pharmacometrics, University Children’s Hospital Basel, University of Basel, Basel, SwitzerlandCorrespondence: Wei ZhaoDepartment of Clinical Pharmacy, School of Pharmaceutical Sciences, Cheeloo College of Medicine, Shandong University, Jinan, 250012, People’s Republic of ChinaEmail zhao4wei2@hotmail.comPurpose: Inflammation has a significant impact on CYP3A activity. We hypothesized that this effect might be age dependent. Our objective was to conduct a population pharmacokinetic study of midazolam in mice at different developmental stages with varying degrees of inflammation to verify our hypothesis.Methods: Different doses (2 and 5 mg/kg) of lipopolysaccharide (LPS) were used to induce different degrees of systemic inflammation in Swiss mice (postnatal age 9– 42 days, n = 220). The CYP3A substrate midazolam was s

There must be Coulomb bound states of a pair of hadrons, which are stable against the strong interaction, with opposite electric charges. Such bound states are hadronic atoms. We study the properties and the production of the ground-state $D^+D^-$ hadronic atom $A_{D^+D^-}$, called dionium, with quantum numbers $J^{PC}=0^{++}$. Using a nonrelativistic effective field theory for the $D^0\bar{D}^0$-$D^+D^-$ coupled-channel system, the mass of the ground-state dionium is predicted to be $(3739.3 \pm 0.1)~\text{MeV}$, with the binding energy reduced by about 10% compared to the Coulomb binding energy due to the strong interaction. Its width for the decay into the neutral $D^0\bar D^0$ channel is predicted to be $1.8^{+1.4}_{-0.6}$ keV using lattice inputs for the $D\bar D$ strong interaction. The cross section for the inclusive prompt production of the dionium at CMS and LHCb and that for the direct production $p\bar p\to A_{D^+D^-}$ at PANDA are estimated at an order-of-magnitude level. In particular, we expect that $\mathcal{O}(10^3\sim 10^5)$ events of the reaction chain $p\bar p\to A_{D^+D^-}\to D^0\bar D^0 \to K^-\pi^+K^+\pi^-$ can be collected at PANDA, and valuable information on the charmed meson interaction and on understanding charmoniumlike states will be obtained., Comment: 25 pages, 5 figures. Version to appear in Phys. Rev. D

Decentralized SGD is an emerging training method for deep learning known for its much less (thus faster) communication per iteration, which relaxes the averaging step in parallel SGD to inexact averaging. The less exact the averaging is, however, the more the total iterations the training needs to take. Therefore, the key to making decentralized SGD efficient is to realize nearly-exact averaging using little communication. This requires a skillful choice of communication topology, which is an under-studied topic in decentralized optimization. In this paper, we study so-called exponential graphs where every node is connected to $O(\log(n))$ neighbors and $n$ is the total number of nodes. This work proves such graphs can lead to both fast communication and effective averaging simultaneously. We also discover that a sequence of $\log(n)$ one-peer exponential graphs, in which each node communicates to one single neighbor per iteration, can together achieve exact averaging. This favorable property enables one-peer exponential graph to average as effective as its static counterpart but communicates more efficiently. We apply these exponential graphs in decentralized (momentum) SGD to obtain the state-of-the-art balance between per-iteration communication and iteration complexity among all commonly-used topologies. Experimental results on a variety of tasks and models demonstrate that decentralized (momentum) SGD over exponential graphs promises both fast and high-quality training. Our code is implemented through BlueFog and available at https://github.com/Bluefog-Lib/NeurIPS2021-Exponential-Graph.

The role of focal amplifications and extrachromosomal DNA (ecDNA) is unknown in gastric cardia adenocarcinoma (GCA). Here, we identify frequent focal amplifications and ecDNAs in Chinese GCA patient samples, and find focal amplifications in the GCA cohort are associated with the chromothripsis process and may be induced by accumulated DNA damage due to local dietary habits. We observe diverse correlations between the presence of oncogene focal amplifications and prognosis, where ERBB2 focal amplifications positively correlate with prognosis and EGFR focal amplifications negatively correlate with prognosis. Large-scale ERBB2 immunohistochemistry results from 1668 GCA patients show survival probability of ERBB2 positive patients is lower than that of ERBB2 negative patients when their surviving time is under 2 years, however, the tendency is opposite when their surviving time is longer than 2 years. Our observations indicate that the ERBB2 focal amplifications may represent a good prognostic marker in GCA patients.

Communication overhead hinders the scalability of large-scale distributed training. Gossip SGD, where each node averages only with its neighbors, is more communication-efficient than the prevalent parallel SGD. However, its convergence rate is reversely proportional to quantity $1-\beta$ which measures the network connectivity. On large and sparse networks where $1-\beta \to 0$, Gossip SGD requires more iterations to converge, which offsets against its communication benefit. This paper introduces Gossip-PGA, which adds Periodic Global Averaging into Gossip SGD. Its transient stage, i.e., the iterations required to reach asymptotic linear speedup stage, improves from $\Omega(\beta^4 n^3/(1-\beta)^4)$ to $\Omega(\beta^4 n^3 H^4)$ for non-convex problems. The influence of network topology in Gossip-PGA can be controlled by the averaging period $H$. Its transient-stage complexity is also superior to Local SGD which has order $\Omega(n^3 H^4)$. Empirical results of large-scale training on image classification (ResNet50) and language modeling (BERT) validate our theoretical findings., Comment: Accepted to ICML 2021