arXiv subjects

Shuo Li

Publications and source records attributed to Shuo Li.

At least 217 records · Page 12Linked to original sources

arXiv2022

Dual Multi-scale Mean Teacher Network for Semi-supervised Infection Segmentation in Chest CT Volume for COVID-19

Automated detecting lung infections from computed tomography (CT) data plays an important role for combating COVID-19. However, there are still some challenges for developing AI system. 1) Most current COVID-19 infection segmentation methods mainly relied on 2D CT images, which lack 3D sequential constraint. 2) Existing 3D CT segmentation methods focus on single-scale representations, which do not achieve the multiple level receptive field sizes on 3D volume. 3) The emergent breaking out of COVID-19 makes it hard to annotate sufficient CT volumes for training deep model. To address these issues, we first build a multiple dimensional-attention convolutional neural network (MDA-CNN) to aggregate multi-scale information along different dimension of input feature maps and impose supervision on multiple predictions from different CNN layers. Second, we assign this MDA-CNN as a basic network into a novel dual multi-scale mean teacher network (DM${^2}$T-Net) for semi-supervised COVID-19 lung infection segmentation on CT volumes by leveraging unlabeled data and exploring the multi-scale information. Our DM${^2}$T-Net encourages multiple predictions at different CNN layers from the student and teacher networks to be consistent for computing a multi-scale consistency loss on unlabeled data, which is then added to the supervised loss on the labeled data from multiple predictions of MDA-CNN. Third, we collect two COVID-19 segmentation datasets to evaluate our method. The experimental results show that our network consistently outperforms the compared state-of-the-art methods.

eess.IV↗

arXiv2022

Synthetic Data Supervised Salient Object Detection

Although deep salient object detection (SOD) has achieved remarkable progress, deep SOD models are extremely data-hungry, requiring large-scale pixel-wise annotations to deliver such promising results. In this paper, we propose a novel yet effective method for SOD, coined SODGAN, which can generate infinite high-quality image-mask pairs requiring only a few labeled data, and these synthesized pairs can replace the human-labeled DUTS-TR to train any off-the-shelf SOD model. Its contribution is three-fold. 1) Our proposed diffusion embedding network can address the manifold mismatch and is tractable for the latent code generation, better matching with the ImageNet latent space. 2) For the first time, our proposed few-shot saliency mask generator can synthesize infinite accurate image synchronized saliency masks with a few labeled data. 3) Our proposed quality-aware discriminator can select highquality synthesized image-mask pairs from noisy synthetic data pool, improving the quality of synthetic data. For the first time, our SODGAN tackles SOD with synthetic data directly generated from the generative model, which opens up a new research paradigm for SOD. Extensive experimental results show that the saliency model trained on synthetic data can achieve $98.4\%$ F-measure of the saliency model trained on the DUTS-TR. Moreover, our approach achieves a new SOTA performance in semi/weakly-supervised methods, and even outperforms several fully-supervised SOTA methods. Code is available at https://github.com/wuzhenyubuaa/SODGAN

cs.CV↗

arXiv2022

Efficient View Path Planning for Autonomous Implicit Reconstruction

Implicit neural representations have shown promising potential for the 3D scene reconstruction. Recent work applies it to autonomous 3D reconstruction by learning information gain for view path planning. Effective as it is, the computation of the information gain is expensive, and compared with that using volumetric representations, collision checking using the implicit representation for a 3D point is much slower. In the paper, we propose to 1) leverage a neural network as an implicit function approximator for the information gain field and 2) combine the implicit fine-grained representation with coarse volumetric representations to improve efficiency. Further with the improved efficiency, we propose a novel informative path planning based on a graph-based planner. Our method demonstrates significant improvements in the reconstruction quality and planning efficiency compared with autonomous reconstructions with implicit and explicit representations. We deploy the method on a real UAV and the results show that our method can plan informative views and reconstruct a scene with high quality.

cs.RO↗

arXiv2022

An upper bound of the number of distinct powers in binary words

A power is a word of the form $\underbrace{uu...u}_{k \; \text{times}}$, where $u$ is a word and $k$ is a positive integer and a square is a word of the form $uu$. Fraenkel and Simpson conjectured in 1998 that the number of distinct squares in a word is bounded by the length of the word. This conjecture was proven recently by Brlek and Li. Besides, there exists a stronger upper bound for binary words conjectured by Jonoska, Manea and Seki stating that for a word of length $n$ over the alphabet $\left\{a, b\right\}$, if we let $k$ be the least of the number of a's and the number of b's and $k \geq 2$, then the number of distinct squares is upper bounded by $\frac{2k-1}{2k+2}n$. In this article, we prove this conjecture by giving a stronger statement on the number of distinct powers in a binary word.

math.CO↗

arXiv2022

Extreme Continuous Treatment Effects: Measures, Estimation and Inference

This paper concerns estimation and inference for treatment effects in deep tails of the counterfactual distribution of unobservable potential outcomes corresponding to a continuously valued treatment. We consider two measures for the deep tail characteristics: the extreme quantile function and the tail mean function defined as the conditional mean beyond a quantile level. Then we define the extreme quantile treatment effect (EQTE) and the extreme average treatment effect (EATE), which can be identified through the commonly adopted unconfoundedness condition and estimated with the aid of extreme value theory. Our limiting theory is for the EQTE and EATE processes indexed by a set of quantile levels and hence facilitates uniform inference. Simulations suggest that our method works well in finite samples and an empirical application illustrates its practical merit.

math.ST↗

arXiv2022

Fluorescence profile of a nitrogen-vacancy center in a nanodiamond

Nanodiamonds containing luminescent point defects are widely explored for applications in quantum bio-sensing such as nanoscale magnetometry, thermometry, and electrometry. A key challenge in the development of such applications is a large variation in fluorescence properties observed between particles, even when obtained from the same batch or nominally identical fabrication processes. By theoretically modelling the emission of nitrogen-vacancy colour centres in spherical nanoparticles, we are able to show that the fluorescence spectrum varies with the exact position of the emitter within the nanoparticle, with noticeable effects seen when the diamond radius, $a$, is larger than around 100 nm, and significantly modified fluorescence profiles found for larger particles when $a=200$ nm and $a=300$ nm, while negligible effects below $a=100$ nm. These results show that the reproducible geometry of point defect position within narrowly sized batch of diamond crystals is necessary for controlling the emission properties. Our results are useful for understanding the extent to which nanodiamonds can be optimised for bio-sensing applications.

physics.optics↗

arXiv2022

Identification and parameter determination of F-type Herbig stars from LAMOST DR8

We identify 20 F-type Herbig stars and provide a list of 22 pre-main-sequence candidates from LAMOST DR8. The effective temperature, distance, extinction, stellar luminosity, mass, and radius are derived for each Herbig star based on optical spectra, photometry, Gaia EDR3 parallaxes, and pre-main-sequence evolutionary tracks. According to spectral energy distributions, 19 F-type Herbig stars belong to Class II YSOs, and one belongs to the flat-spectrum class. Four have Spitzer IRS spectra, of which three show extremely weak polycyclic aromatic hydrocarbons emissions, and three with both amorphous and crystalline silicate emissions share the similar parameters and are at the same evolutionary stage. We detect a solar-nearby outbursting EXor Herbig star J034344.48+314309.3, possible precursor of a Herbig Ae star. Intense emission lines of HI, HeI, OI, NaI, and CaII originated from the rapid accretion during the outbursts are detected in its optical spectra, and silicate emission features are detected in its infrared spectrum. We also make a statistic analysis on the disk properties of all known Herbig stars using the defined infrared spectral indices. The proportion of Herbig stars with moderate infrared excesses decreases as effective temperature increases. The majority of the precursors (F-, G-, or K- type) have moderate infrared excesses. Hotter Herbig stars tend to have a larger proportion with large infrared excesses. The trends may be due to the fact that hotter stars have larger areas of re-emitting dust, although there is some scatter due to the particularities of each disk.

astro-ph.SR↗

arXiv2022

A note on the Lie complexity and beyond

In a recent paper, Jason P. Bell and Jeffrey Shallit introduced the notion of {\em Lie complexity} and proved that the Lie complexity function of an automatic sequence is automatic. In this note, we give more facts concerning Lie complexity and define the extended Lie complexity and the prefix Lie complexity. Further, we prove that some proprieties of Lie complexity also hold for the extended Lie complexity. Particularly, we prove that the extended Lie complexity function and the first-order difference sequence of the prefix Lie complexity function of an automatic sequence are both automatic.

math.CO↗

arXiv2022

FFCNet: Fourier Transform-Based Frequency Learning and Complex Convolutional Network for Colon Disease Classification

Reliable automatic classification of colonoscopy images is of great significance in assessing the stage of colonic lesions and formulating appropriate treatment plans. However, it is challenging due to uneven brightness, location variability, inter-class similarity, and intra-class dissimilarity, affecting the classification accuracy. To address the above issues, we propose a Fourier-based Frequency Complex Network (FFCNet) for colon disease classification in this study. Specifically, FFCNet is a novel complex network that enables the combination of complex convolutional networks with frequency learning to overcome the loss of phase information caused by real convolution operations. Also, our Fourier transform transfers the average brightness of an image to a point in the spectrum (the DC component), alleviating the effects of uneven brightness by decoupling image content and brightness. Moreover, the image patch scrambling module in FFCNet generates random local spectral blocks, empowering the network to learn long-range and local diseasespecific features and improving the discriminative ability of hard samples. We evaluated the proposed FFCNet on an in-house dataset with 2568 colonoscopy images, showing our method achieves high performance outperforming previous state-of-the art methods with an accuracy of 86:35% and an accuracy of 4.46% higher than the backbone. The project page with code is available at https://github.com/soleilssss/FFCNet.

eess.IV↗

arXiv2022

Position-prior Clustering-based Self-attention Module for Knee Cartilage Segmentation

The morphological changes in knee cartilage (especially femoral and tibial cartilages) are closely related to the progression of knee osteoarthritis, which is expressed by magnetic resonance (MR) images and assessed on the cartilage segmentation results. Thus, it is necessary to propose an effective automatic cartilage segmentation model for longitudinal research on osteoarthritis. In this research, to relieve the problem of inaccurate discontinuous segmentation caused by the limited receptive field in convolutional neural networks, we proposed a novel position-prior clustering-based self-attention module (PCAM). In PCAM, long-range dependency between each class center and feature point is captured by self-attention allowing contextual information re-allocated to strengthen the relative features and ensure the continuity of segmentation result. The clutsering-based method is used to estimate class centers, which fosters intra-class consistency and further improves the accuracy of segmentation results. The position-prior excludes the false positives from side-output and makes center estimation more precise. Sufficient experiments are conducted on OAI-ZIB dataset. The experimental results show that the segmentation performance of combination of segmentation network and PCAM obtains an evident improvement compared to original model, which proves the potential application of PCAM in medical segmentation tasks. The source code is publicly available from link: https://github.com/LeongDong/PCAMNet

eess.IV↗

arXiv2022

PAC-Wrap: Semi-Supervised PAC Anomaly Detection

Anomaly detection is essential for preventing hazardous outcomes for safety-critical applications like autonomous driving. Given their safety-criticality, these applications benefit from provable bounds on various errors in anomaly detection. To achieve this goal in the semi-supervised setting, we propose to provide Probably Approximately Correct (PAC) guarantees on the false negative and false positive detection rates for anomaly detection algorithms. Our method (PAC-Wrap) can wrap around virtually any existing semi-supervised and unsupervised anomaly detection method, endowing it with rigorous guarantees. Our experiments with various anomaly detectors and datasets indicate that PAC-Wrap is broadly effective.

cs.LG↗

arXiv2022

XMorpher: Full Transformer for Deformable Medical Image Registration via Cross Attention

An effective backbone network is important to deep learning-based Deformable Medical Image Registration (DMIR), because it extracts and matches the features between two images to discover the mutual correspondence for fine registration. However, the existing deep networks focus on single image situation and are limited in registration task which is performed on paired images. Therefore, we advance a novel backbone network, XMorpher, for the effective corresponding feature representation in DMIR. 1) It proposes a novel full transformer architecture including dual parallel feature extraction networks which exchange information through cross attention, thus discovering multi-level semantic correspondence while extracting respective features gradually for final effective registration. 2) It advances the Cross Attention Transformer (CAT) blocks to establish the attention mechanism between images which is able to find the correspondence automatically and prompts the features to fuse efficiently in the network. 3) It constrains the attention computation between base windows and searching windows with different sizes, and thus focuses on the local transformation of deformable registration and enhances the computing efficiency at the same time. Without any bells and whistles, our XMorpher gives Voxelmorph 2.8% improvement on DSC , demonstrating its effective representation of the features from the paired images in DMIR. We believe that our XMorpher has great application potential in more paired medical images. Our XMorpher is open on https://github.com/Solemoon/XMorpher

cs.CV↗

arXiv2022

ULTRA: Uncertainty-aware Label Distribution Learning for Breast Tumor Cellularity Assessment

Neoadjuvant therapy (NAT) for breast cancer is a common treatment option in clinical practice. Tumor cellularity (TC), which represents the percentage of invasive tumors in the tumor bed, has been widely used to quantify the response of breast cancer to NAT. Therefore, automatic TC estimation is significant in clinical practice. However, existing state-of-the-art methods usually take it as a TC score regression problem, which ignores the ambiguity of TC labels caused by subjective assessment or multiple raters. In this paper, to efficiently leverage the label ambiguities, we proposed an Uncertainty-aware Label disTRibution leArning (ULTRA) framework for automatic TC estimation. The proposed ULTRA first converted the single-value TC labels to discrete label distributions, which effectively models the ambiguity among all possible TC labels. Furthermore, the network learned TC label distributions by minimizing the Kullback-Leibler (KL) divergence between the predicted and ground-truth TC label distributions, which better supervised the model to leverage the ambiguity of TC labels. Moreover, the ULTRA mimicked the multi-rater fusion process in clinical practice with a multi-branch feature fusion module to further explore the uncertainties of TC labels. We evaluated the ULTRA on the public BreastPathQ dataset. The experimental results demonstrate that the ULTRA outperformed the regression-based methods for a large margin and achieved state-of-the-art results. The code will be available from https://github.com/PerceptionComputingLab/ULTRA

eess.IV↗

arXiv2022

Pressure-induced superconductivity in flat-band Kagome compounds Pd$_3$P$_2$(S$_{1-x}$Se$_x$)$_8$

We performed high-pressure transport studies on the flat-band Kagome compounds, Pd$_3$P$_2$(S$_{1-x}$Se$_x$)$_8$ ($x$ = 0, 0.25), with a diamond anvil cell. For both compounds, the resistivity exhibits an insulating behavior with pressure up to 17 GPa. With pressure above 20 GPa, a metallic behavior is observed at high temperatures in Pd$_3$P$_2$S$_8$, and superconductivity emerges at low temperatures. The onset temperature of superconducting transition $T_{\rm C}$ rises monotonically from 2 K to 4.8 K and does not saturate with pressure up to 43 GPa. For the Se-doped compound Pd$_3$P$_2$(S$_{0.75}$Se$_{0.25}$)$_8$, the $T_{\rm C}$ is about 1.5 K higher than that of the undoped one over the whole pressure range, and reaches 6.4 K at 43 GPa. The upper critical field with field applied along the $c$ axis at typical pressures is about 50$\%$ of the Pauli limit, suggesting a 3D superconductivity. The Hall coefficient in the metallic phase is low and exhibits a peaked behavior at about 30 K, which suggests either a multi-band electronic structure or an electron correlation effect in the system.

cond-mat.supr-con↗

arXiv2022

A note on the maximum number of $k$-powers in a finite word

A \emph{power} is a word of the form $\underbrace{uu...u}_{k \; \text{times}}$, where $u$ is a word and $k$ is a positive integer; the power is also called a {\em $k$-power} and $k$ is its {\em exponent}. We prove that for any $k \ge 2$, the maximum number of different non-empty $k$-power factors in a word of length $n$ is between $\frac{n}{k-1}-Θ(\sqrt{n})$ and $\frac{n-1}{k-1}$. We also show that the maximum number of different non-empty power factors of exponent at least 2 in a length-$n$ word is at most $n-1$. Both upper bounds generalize the recent upper bound of $n-1$ on the maximum number of different square factors in a length-$n$ word by Brlek and Li (2022).

math.CO↗

arXiv2022

Revisit the rate of tidal disruption events: the role of the partial tidal disruption event

Tidal disruption of stars in dense nuclear star clusters containing supermassive central black holes (SMBH) is modeled by high-accuracy direct N-body simulation. Stars getting too close to the SMBH are tidally disrupted and a tidal disruption event (TDE) happens. TDEs probe properties of SMBH, their accretion disks, and the surrounding nuclear stellar cluster. In this paper we compare rates of full tidal disruption events (FTDE) with partial tidal disruption events (PTDE). Since a PTDE does not destroy the star, a leftover object emerges; we use the term 'leftover star' for it; two novel effects occur in the simulation: (1) variation of the leftover star's mass and radius, (2) variation of the leftover star's orbital energy. After switching on these two effects in our simulation, the number of FTDEs is reduced by roughly 28%, and the reduction is mostly due to the ejection of the leftover stars from PTDEs coming originally from relatively large distance. The number of PTDEs is about 75% higher than the simple estimation given by Stone et al. (2020), and the enhancement is mainly due to the multiple PTDEs produced by the leftover stars residing in the diffusive regime. We compute the peak mass fallback rate for the PTDEs and FTDEs recorded in the simulation, and find 58% of the PTDEs have peak mass fallback rate exceeding the Eddington limit, and the number of super-Eddington PTDEs is 2.3 times the number of super-Eddington FTDEs.

astro-ph.GA↗

arXiv2022

MNet: Rethinking 2D/3D Networks for Anisotropic Medical Image Segmentation

The nature of thick-slice scanning causes severe inter-slice discontinuities of 3D medical images, and the vanilla 2D/3D convolutional neural networks (CNNs) fail to represent sparse inter-slice information and dense intra-slice information in a balanced way, leading to severe underfitting to inter-slice features (for vanilla 2D CNNs) and overfitting to noise from long-range slices (for vanilla 3D CNNs). In this work, a novel mesh network (MNet) is proposed to balance the spatial representation inter axes via learning. 1) Our MNet latently fuses plenty of representation processes by embedding multi-dimensional convolutions deeply into basic modules, making the selections of representation processes flexible, thus balancing representation for sparse inter-slice information and dense intra-slice information adaptively. 2) Our MNet latently fuses multi-dimensional features inside each basic module, simultaneously taking the advantages of 2D (high segmentation accuracy of the easily recognized regions in 2D view) and 3D (high smoothness of 3D organ contour) representations, thus obtaining more accurate modeling for target regions. Comprehensive experiments are performed on four public datasets (CT\&MR), the results consistently demonstrate the proposed MNet outperforms the other methods. The code and datasets are available at: https://github.com/zfdong-code/MNet

eess.IV↗

arXiv2022

On the number of squares in a finite word

A {\em square} is a word of the form $uu$. In this paper we prove that for a given finite word $w$, the number of distinct square factors of $w$ is bounded by $|w|-|\Alphabet(w)|+1$, where $|w|$ denotes the length of $w$ and $|\Alphabet(w)|$ denotes the number of distinct letters in $w$. This result answers a conjecture of Fraenkel and Simpson stated in 1998.

math.CO↗