Search

Pancreatic ductal adenocarcinoma (PDAC) presents a critical global health challenge, and early detection is crucial for improving the 5-year survival rate. Recent medical imaging and computational algorithm advances offer potential solutions for early diagnosis. Deep learning, particularly in the form of convolutional neural networks (CNNs), has demonstrated success in medical image analysis tasks, including classification and segmentation. However, the limited availability of clinical data for training purposes continues to provide a significant obstacle. Data augmentation, generative adversarial networks (GANs), and cross-validation are potential techniques to address this limitation and improve model performance, but effective solutions are still rare for 3D PDAC, where contrast is especially poor owing to the high heterogeneity in both tumor and background tissues. In this study, we developed a new GAN-based model, named 3DGAUnet, for generating realistic 3D CT images of PDAC tumors and pancreatic tissue, which can generate the interslice connection data that the existing 2D CT image synthesis models lack. Our innovation is to develop a 3D U-Net architecture for the generator to improve shape and texture learning for PDAC tumors and pancreatic tissue. Our approach offers a promising path to tackle the urgent requirement for creative and synergistic methods to combat PDAC. The development of this GAN-based model has the potential to alleviate data scarcity issues, elevate the quality of synthesized data, and thereby facilitate the progression of deep learning models to enhance the accuracy and early detection of PDAC tumors, which could profoundly impact patient outcomes. Furthermore, this model has the potential to be adapted to other types of solid tumors, hence making significant contributions to the field of medical imaging in terms of image processing models.

We investigate the $K_1(1270)-K_1(1400)$ mixing caused by the flavor $SU(3)$ symmetry breaking. The mixing angle is expressed by a $K_{1A}\to K_{1B}$ matrix element induced by the operators that breaks flavor $SU(3)$ symmetry. The QCD contribution to this matrix element is assumed to be dominated and calculated by QCD sum rules. A three-point correlation function is defined and handled both at the hadron and quark-gluon levels. The quark-gluon level calculation is based on operator product expansion up to dimension-5 condensates. A detailed numerical analysis is performed to determine the Borel parameters, and the obtained mixing angle is $\theta_{K_1}=22^{\circ}\pm 7^{\circ}$ or $\theta_{K_1}=68^{\circ}\pm 7^{\circ}$., Comment: 14 pages, 7 figures and 1 table

We explore the QED corrections to the $\Xi_c-\Xi_c^{\prime}$ mixing within the framework of light-front quark model (LFQM) in the three-quark picture. After explicitly investigating the relation between the $\Xi_c-\Xi_c^{\prime}$ mixing and the flavor $\rm {SU(3)}$ and heavy quark symmetry breaking, we derive the QED contributions to the mixing angle. Numerical results indicate the QED contribution is smaller than the one from the mass difference between the strange and up/down quark provided by a recent Lattice QCD analysis. Adding these contributions together we find that at this stage the $\Xi_c-\Xi_c^{\prime}$ mixing is small and still incapable to account for the large $\rm {SU(3)}$ symmetry breaking in the semi-leptonic $\Xi_c$ decays., Comment: 7 pages, 4figures

The actomyosin cortex is an active material that provides animal cells with a strong but flexible exterior, whose mechanics, including non-Gaussian fluctuations and occasional large displacements or cytoquakes, have defied explanation. We study the active fluctuations of the cortex using nanoscale tracking of arrays of flexible microposts adhered to multiple cultured cell types. When the confounding effects of static heterogeneity and tracking error are removed, the fluctuations are found to be heavy-tailed and well-described by a truncated L\'evy alpha-stable distribution over a wide range of timescales, in multiple cell types. The largest random displacements closely resemble the earlier-reported cytoquakes, but notably, we find these cytoquakes are not due to earthquake-like cooperative rearrangement of many cytoskeletal elements. Rather, they are indistinguishable from chance large excursions of a super-diffusive random process driven by heavy-tailed noise. The non-cooperative microscopic events driving these fluctuations need not be larger than the expected elastic energy of single tensed cortical actin filaments, and the implied distribution of microscopic event energies will need to be accounted for by future models of the cytoskeleton., Comment: 13 pages, 11 figures

Partial Label Learning (PLL) is a type of weakly supervised learning where each training instance is assigned a set of candidate labels, but only one label is the ground-truth. However, this idealistic assumption may not always hold due to potential annotation inaccuracies, meaning the ground-truth may not be present in the candidate label set. This is known as Unreliable Partial Label Learning (UPLL) that introduces an additional complexity due to the inherent unreliability and ambiguity of partial labels, often resulting in a sub-optimal performance with existing methods. To address this challenge, we propose the Unreliability-Robust Representation Learning framework (URRL) that leverages unreliability-robust contrastive learning to help the model fortify against unreliable partial labels effectively. Concurrently, we propose a dual strategy that combines KNN-based candidate label set correction and consistency-regularization-based label disambiguation to refine label quality and enhance the ability of representation learning within the URRL framework. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art PLL methods on various datasets with diverse degrees of unreliability and ambiguity. Furthermore, we provide a theoretical analysis of our approach from the perspective of the expectation maximization (EM) algorithm. Upon acceptance, we pledge to make the code publicly accessible.

Medical image analysis using deep learning is often challenged by limited labeled data and high annotation costs. Fine-tuning the entire network in label-limited scenarios can lead to overfitting and suboptimal performance. Recently, prompt tuning has emerged as a more promising technique that introduces a few additional tunable parameters as prompts to a task-agnostic pre-trained model, and updates only these parameters using supervision from limited labeled data while keeping the pre-trained model unchanged. However, previous work has overlooked the importance of selective labeling in downstream tasks, which aims to select the most valuable downstream samples for annotation to achieve the best performance with minimum annotation cost. To address this, we propose a framework that combines selective labeling with prompt tuning (SLPT) to boost performance in limited labels. Specifically, we introduce a feature-aware prompt updater to guide prompt tuning and a TandEm Selective LAbeling (TESLA) strategy. TESLA includes unsupervised diversity selection and supervised selection using prompt-based uncertainty. In addition, we propose a diversified visual prompt tuning strategy to provide multi-prompt-based discrepant predictions for TESLA. We evaluate our method on liver tumor segmentation and achieve state-of-the-art performance, outperforming traditional fine-tuning with only 6% of tunable parameters, also achieving 94% of full-data performance by labeling only 5% of the data., Comment: accepted by MICCAI 2023

Quantum phase estimation (QPE) of the eigenvalues of a unitary operator on a target quantum system is a crucial subroutine in various quantum algorithms. Conventional QPE is often expensive to implement as it requires a large number of ancilla qubits and the ability to perform quantum Fourier transform. Recent developments in iterative QPE reduce the implementation cost by repetitive uses of a single ancilla and classical post-processing. However, both conventional and iterative schemes often require preparation of the target system in an eigenstate of the unitary operator, while it remains ambiguous to achieve QPE of multiple eigenvalues with no need of initial state preparation. Here we clarify this issue by developing a theoretical framework based on sequential quantum channels for iterative QPE. We find that QPE of multiple eigenvalues can be efficiently realized for arbitrary initial target system state by actively utilizing the measurement backaction of iterative QPE on the target system with a long coherence time. Specifically, we investigate two iterative QPE schemes based on sequential Ramsey interferometry measurements (RIMs) of an ancilla qubit: (a) the repetitive scheme, which conducts repetitive RIMs to achieve the standard quantum limit in estimating the eigenvalues; (b) the adaptive scheme, which adjusts the parameters of each RIM based on prior measurement outcomes to attain the Heisenberg limit. In both schemes, sequential ancilla measurements generate sequential quantum channels on the target system, gradually steering it to the eigenstates of the estimated unitary operator, while the measurement statistics of the ancilla can reveal the embedded information about its eigenvalues with proper post-processing. We demonstrate the analysis by simulating a central spin model, and evaluate the performance and noise resilience of both schemes., Comment: 19 pages, 8 figures

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal cancer in which the tumor-vascular involvement greatly affects the resectability and, thus, overall survival of patients. However, current prognostic prediction methods fail to explicitly and accurately investigate relationships between the tumor and nearby important vessels. This paper proposes a novel learnable neural distance that describes the precise relationship between the tumor and vessels in CT images of different patients, adopting it as a major feature for prognosis prediction. Besides, different from existing models that used CNNs or LSTMs to exploit tumor enhancement patterns on dynamic contrast-enhanced CT imaging, we improved the extraction of dynamic tumor-related texture features in multi-phase contrast-enhanced CT by fusing local and global features using CNN and transformer modules, further enhancing the features extracted across multi-phase CT images. We extensively evaluated and compared the proposed method with existing methods in the multi-center (n=4) dataset with 1,070 patients with PDAC, and statistical analysis confirmed its clinical effectiveness in the external test set consisting of three centers. The developed risk marker was the strongest predictor of overall survival among preoperative factors and it has the potential to be combined with established clinical factors to select patients at higher risk who might benefit from neoadjuvant therapy., Comment: MICCAI 2023

Liver tumor segmentation and classification are important tasks in computer aided diagnosis. We aim to address three problems: liver tumor screening and preliminary diagnosis in non-contrast computed tomography (CT), and differential diagnosis in dynamic contrast-enhanced CT. A novel framework named Pixel-Lesion-pAtient Network (PLAN) is proposed. It uses a mask transformer to jointly segment and classify each lesion with improved anchor queries and a foreground-enhanced sampling loss. It also has an image-wise classifier to effectively aggregate global information and predict patient-level diagnosis. A large-scale multi-phase dataset is collected containing 939 tumor patients and 810 normal subjects. 4010 tumor instances of eight types are extensively annotated. On the non-contrast tumor screening task, PLAN achieves 95% and 96% in patient-level sensitivity and specificity. On contrast-enhanced CT, our lesion-level detection precision, recall, and classification accuracy are 92%, 89%, and 86%, outperforming widely used CNN and transformers for lesion segmentation. We also conduct a reader study on a holdout set of 250 cases. PLAN is on par with a senior human radiologist, showing the clinical significance of our results., Comment: MICCAI 2023, code: https://github.com/alibaba-damo-academy/pixel-lesion-patient-network

We extend the previously developed small $x$ parton shower algorithm to include the kinematic constraint effect and $k_t$ resummation effect. This work enables the Monte Carlo generator to simultaneously resum large $k_t$ and small $x$ logarithms in the saturation regime for the first time. It is an important step towards simulating processes involving multiple well separated hard scales, such as di-jet production in eA collisions at EIC., Comment: 12 pages, 4 figures. arXiv admin note: text overlap with arXiv:2211.07174

We study the heavy flavor conserved semi-leptonic decay $B_s\to B\ell\nu$ in the covariant light front approach. The covariant light front quark model is used to calculate the transition form factors of $B_s\to B^{(*)}$ as well as $D_s\to D$, which are consistent with the leading power predictions from the heavy quark symmetry. The angular distribution analysis on the $B_s\to B^{(*)}l\bar\nu$ decay is performed by investigating the forward-backward asymmetry of the lepton. We also study the angular distribution of $B_{s}\to B^{*}(\to B \gamma)l\bar\nu$ decay both through the lepton forward-backward asymmetry and the azimuth angle. The branching fractions of $B_s\to Bl\bar\nu$ and $B_s\to B^{*}l\bar\nu$ are at the order $10^{-8}$ and $10^{-9}$, respectively. The number of $B_s\to B l\bar\nu$ events is estimated to be $1.76$. The branching fraction of $B_{s}\to B^{*}(\to B \gamma)l\bar\nu$ is at the order $10^{-11}$, which is calculated by introducing Breit-Wigner distribution for the intermediate $B^*$., Comment: 12 pages, 6 figures, 2 tables

This study aims to develop a new computational pathology approach that automates the identification and quantification of myocardial inflammatory infiltration in digital HE-stained images to provide a quantitative histological diagnosis of myocarditis.898 HE-stained whole slide images (WSIs) of myocardium from 154 heart transplant patients diagnosed with myocarditis or dilated cardiomyopathy (DCM) were included in this study. An automated DL-based computational pathology approach was developed to identify nuclei and detect myocardial inflammatory infiltration, enabling the quantification of the lymphocyte nuclear density (LND) on myocardial WSIs. A cutoff value based on the quantification of LND was proposed to determine if the myocardial inflammatory infiltration was present. The performance of our approach was evaluated with a five-fold cross-validation experiment, tested with an internal test set from the myocarditis group, and confirmed by an external test from a double-blind trial group. An LND of 1.02/mm2 could distinguish WSIs with myocarditis from those without. The accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) in the five-fold cross-validation experiment were 0.899 plus or minus 0.035, 0.971 plus or minus 0.017, 0.728 plus or minus 0.073 and 0.849 plus or minus 0.044, respectively. For the internal test set, the accuracy, sensitivity, specificity, and AUC were 0.887, 0.971, 0.737, and 0.854, respectively. The accuracy, sensitivity, specificity, and AUC for the external test set reached 0.853, 0.846, 0.858, and 0.852, respectively. Our new approach provides accurate and reliable quantification of the LND of myocardial WSIs, facilitating automated quantitative diagnosis of myocarditis with HE-stained images., Comment: 21 pages,5 figures,6 Tables, 25 references

In-time particle trajectory reconstruction in the Large Hadron Collider is challenging due to the high collision rate and numerous particle hits. Using GNN (Graph Neural Network) on FPGA has enabled superior accuracy with flexible trajectory classification. However, existing GNN architectures have inefficient resource usage and insufficient parallelism for edge classification. This paper introduces a resource-efficient GNN architecture on FPGAs for low latency particle tracking. The modular architecture facilitates design scalability to support large graphs. Leveraging the geometric properties of hit detectors further reduces graph complexity and resource usage. Our results on Xilinx UltraScale+ VU9P demonstrate 1625x and 1574x performance improvement over CPU and GPU respectively.

Advances in deep learning have greatly improved structure prediction of molecules. However, many macroscopic observations that are important for real-world applications are not functions of a single molecular structure, but rather determined from the equilibrium distribution of structures. Traditional methods for obtaining these distributions, such as molecular dynamics simulation, are computationally expensive and often intractable. In this paper, we introduce a novel deep learning framework, called Distributional Graphormer (DiG), in an attempt to predict the equilibrium distribution of molecular systems. Inspired by the annealing process in thermodynamics, DiG employs deep neural networks to transform a simple distribution towards the equilibrium distribution, conditioned on a descriptor of a molecular system, such as a chemical graph or a protein sequence. This framework enables efficient generation of diverse conformations and provides estimations of state densities. We demonstrate the performance of DiG on several molecular tasks, including protein conformation sampling, ligand structure sampling, catalyst-adsorbate sampling, and property-guided structure generation. DiG presents a significant advancement in methodology for statistically understanding molecular systems, opening up new research opportunities in molecular science., Comment: 80 pages, 11 figures