arXiv subjects

Guang Yang

Publications and source records attributed to Guang Yang.

At least 253 records · Page 14Linked to original sources

arXiv2024

Toward a More Complete OMR Solution

Optical music recognition (OMR) aims to convert music notation into digital formats. One approach to tackle OMR is through a multi-stage pipeline, where the system first detects visual music notation elements in the image (object detection) and then assembles them into a music notation (notation assembly). Most previous work on notation assembly unrealistically assumes perfect object detection. In this study, we focus on the MUSCIMA++ v2.0 dataset, which represents musical notation as a graph with pairwise relationships among detected music objects, and we consider both stages together. First, we introduce a music object detector based on YOLOv8, which improves detection performance. Second, we introduce a supervised training pipeline that completes the notation assembly stage based on detection output. We find that this model is able to outperform existing models trained on perfect detection output, showing the benefit of considering the detection and assembly stages in a more holistic way. These findings, together with our novel evaluation metric, are important steps toward a more complete OMR solution.

cs.CV↗

arXiv2024

SeLoRA: Self-Expanding Low-Rank Adaptation of Latent Diffusion Model for Medical Image Synthesis

The persistent challenge of medical image synthesis posed by the scarcity of annotated data and the need to synthesize `missing modalities' for multi-modal analysis, underscored the imperative development of effective synthesis methods. Recently, the combination of Low-Rank Adaptation (LoRA) with latent diffusion models (LDMs) has emerged as a viable approach for efficiently adapting pre-trained large language models, in the medical field. However, the direct application of LoRA assumes uniform ranking across all linear layers, overlooking the significance of different weight matrices, and leading to sub-optimal outcomes. Prior works on LoRA prioritize the reduction of trainable parameters, and there exists an opportunity to further tailor this adaptation process to the intricate demands of medical image synthesis. In response, we present SeLoRA, a Self-Expanding Low-Rank Adaptation Module, that dynamically expands its ranking across layers during training, strategically placing additional ranks on crucial layers, to allow the model to elevate synthesis quality where it matters most. The proposed method not only enables LDMs to fine-tune on medical data efficiently but also empowers the model to achieve improved image quality with minimal ranking. The code of our SeLoRA method is publicly available on https://anonymous.4open.science/r/SeLoRA-980D .

cs.CV↗

arXiv2024

CIResDiff: A Clinically-Informed Residual Diffusion Model for Predicting Idiopathic Pulmonary Fibrosis Progression

The progression of Idiopathic Pulmonary Fibrosis (IPF) significantly correlates with higher patient mortality rates. Early detection of IPF progression is critical for initiating timely treatment, which can effectively slow down the advancement of the disease. However, the current clinical criteria define disease progression requiring two CT scans with a one-year interval, presenting a dilemma: a disease progression is identified only after the disease has already progressed. To this end, in this paper, we develop a novel diffusion model to accurately predict the progression of IPF by generating patient's follow-up CT scan from the initial CT scan. Specifically, from the clinical prior knowledge, we tailor improvements to the traditional diffusion model and propose a Clinically-Informed Residual Diffusion model, called CIResDiff. The key innovations of CIResDiff include 1) performing the target region pre-registration to align the lung regions of two CT scans at different time points for reducing the generation difficulty, 2) adopting the residual diffusion instead of traditional diffusion to enable the model focus more on differences (i.e., lesions) between the two CT scans rather than the largely identical anatomical content, and 3) designing the clinically-informed process based on CLIP technology to integrate lung function information which is highly relevant to diagnosis into the reverse process for assisting generation. Extensive experiments on clinical data demonstrate that our approach can outperform state-of-the-art methods and effectively predict the progression of IPF.

eess.IV↗

arXiv2024

Chain-of-Thought in Neural Code Generation: From and For Lightweight Language Models

Large Language Models (LLMs) have demonstrated remarkable potential in code generation. The integration of Chain of Thought (CoT) reasoning can further boost their performance. However, current CoT methods often require manual writing or LLMs with over 100 billion parameters to generate, impeding their applicability in resource-constrained scenarios. In this study, we investigate lightweight Language Models (lLMs), which are defined to have fewer than 10 billion parameters. Empirically, we find that most lLMs cannot generate high-quality CoTs when prompted by the few-shot method, but can take advantage of high-quality CoTs generated elsewhere to improve their performance in code generation. Based on these findings, we design a novel approach COTTON which can leverage lLMs to automatically generate CoTs for code generation. We synthesize new datasets and conduct extensive experiments on various benchmarks. The results show that the CoTs generated by COTTON outperform the baselines in terms of automated and human evaluation metrics. In particular, the CoTs generated by COTTON boost various lLMs to achieve higher performance gains than those generated by LLMs such as ChatGLM (130B), and are competitive with those generated by gpt-3.5-turbo (175B). Our study also showcases the potential of lLMs in software engineering applications.

cs.SE↗

arXiv2024

A dual-task mutual learning framework for predicting post-thrombectomy cerebral hemorrhage

Ischemic stroke is a severe condition caused by the blockage of brain blood vessels, and can lead to the death of brain tissue due to oxygen deprivation. Thrombectomy has become a common treatment choice for ischemic stroke due to its immediate effectiveness. But, it carries the risk of postoperative cerebral hemorrhage. Clinically, multiple CT scans within 0-72 hours post-surgery are used to monitor for hemorrhage. However, this approach exposes radiation dose to patients, and may delay the detection of cerebral hemorrhage. To address this dilemma, we propose a novel prediction framework for measuring postoperative cerebral hemorrhage using only the patient's initial CT scan. Specifically, we introduce a dual-task mutual learning framework to takes the initial CT scan as input and simultaneously estimates both the follow-up CT scan and prognostic label to predict the occurrence of postoperative cerebral hemorrhage. Our proposed framework incorporates two attention mechanisms, i.e., self-attention and interactive attention. Specifically, the self-attention mechanism allows the model to focus more on high-density areas in the image, which are critical for diagnosis (i.e., potential hemorrhage areas). The interactive attention mechanism further models the dependencies between the interrelated generation and classification tasks, enabling both tasks to perform better than the case when conducted individually. Validated on clinical data, our method can generate follow-up CT scans better than state-of-the-art methods, and achieves an accuracy of 86.37% in predicting follow-up prognostic labels. Thus, our work thus contributes to the timely screening of post-thrombectomy cerebral hemorrhage, and could significantly reform the clinical process of thrombectomy and other similar operations related to stroke.

eess.IV↗

arXiv2024

Probabilistic Method to Fundamental gap problems on the sphere

We provide a probabilistic proof of the fundamental gap estimate for Schrödinger operators in convex domains on the sphere, which extends the probabilistic proof of F. Gong, H. Li, and D. Luo for the Euclidean case. Our results further generalize the results achieved for the Laplacian by S. Seto, L. Wang, and G. Wei, as well as by C. He, G. Wei, and Qi S. Zhang. The essential ingredient in our analysis is the reflection coupling method on Riemannian manifolds.

math.PR↗

arXiv2024

CEERS Key Paper. IX. Identifying Galaxy Mergers in CEERS NIRCam Images Using Random Forests and Convolutional Neural Networks

A crucial yet challenging task in galaxy evolution studies is the identification of distant merging galaxies, a task which suffers from a variety of issues ranging from telescope sensitivities and limitations to the inherently chaotic morphologies of young galaxies. In this paper, we use random forests and convolutional neural networks to identify high-redshift JWST CEERS galaxy mergers. We train these algorithms on simulated $3<z<5$ CEERS galaxies created from the IllustrisTNG subhalo morphologies and the Santa Cruz SAM lightcone. We apply our models to observed CEERS galaxies at $3<z<5$. We find that our models correctly classify $\sim60-70\%$ of simulated merging and non-merging galaxies; better performance on the merger class comes at the expense of misclassifying more non-mergers. We could achieve more accurate classifications, as well as test for the dependency on physical parameters such as gas fraction, mass ratio, and relative orbits, by curating larger training sets. When applied to real CEERS galaxies using visual classifications as ground truth, the random forests correctly classified $40-60\%$ of mergers and non-mergers at $3<z<4$, but tended to classify most objects as non-mergers at $4<z<5$ (misclassifying $\sim70\%$ of visually-classified mergers). On the other hand, the CNNs tended to classify most objects as mergers across all redshifts (misclassifying $80-90\%$ of visually-classified non-mergers). We investigate what features the models find most useful, as well as characteristics of false positives and false negatives, and also calculate merger rates derived from the identifications made by the models.

astro-ph.GA↗

arXiv2024

Differentiable Voxelization and Mesh Morphing

In this paper, we propose the differentiable voxelization of 3D meshes via the winding number and solid angles. The proposed approach achieves fast, flexible, and accurate voxelization of 3D meshes, admitting the computation of gradients with respect to the input mesh and GPU acceleration. We further demonstrate the application of the proposed voxelization in mesh morphing, where the voxelized mesh is deformed by a neural network. The proposed method is evaluated on the ShapeNet dataset and achieves state-of-the-art performance in terms of both accuracy and efficiency.

cs.CV↗

arXiv2024

Artificial Immunofluorescence in a Flash: Rapid Synthetic Imaging from Brightfield Through Residual Diffusion

Immunofluorescent (IF) imaging is crucial for visualizing biomarker expressions, cell morphology and assessing the effects of drug treatments on sub-cellular components. IF imaging needs extra staining process and often requiring cell fixation, therefore it may also introduce artefects and alter endogenouous cell morphology. Some IF stains are expensive or not readily available hence hindering experiments. Recent diffusion models, which synthesise high-fidelity IF images from easy-to-acquire brightfield (BF) images, offer a promising solution but are hindered by training instability and slow inference times due to the noise diffusion process. This paper presents a novel method for the conditional synthesis of IF images directly from BF images along with cell segmentation masks. Our approach employs a Residual Diffusion process that enhances stability and significantly reduces inference time. We performed a critical evaluation against other image-to-image synthesis models, including UNets, GANs, and advanced diffusion models. Our model demonstrates significant improvements in image quality (p<0.05 in MSE, PSNR, and SSIM), inference speed (26 times faster than competing diffusion models), and accurate segmentation results for both nuclei and cell bodies (0.77 and 0.63 mean IOU for nuclei and cell true positives, respectively). This paper is a substantial advancement in the field, providing robust and efficient tools for cell image analysis.

eess.IV↗

arXiv2024

Versatile Method of Engineering the Band Alignment and the Electron Wavefunction Hybridization of Hybrid Quantum Devices

With the development of quantum technology, hybrid devices that combine superconductors (S) and semiconductors (Sm) have attracted great attention due to the possibility of engineering structures that benefit from the integration of the properties of both materials. However, until now, none of the experiments have reported good control of band alignment at the interface, which determines the strength of S-Sm coupling and the proximitized superconducting gap. Here, we fabricate hybrid devices in a generic way with argon milling to modify the interface while maintaining its high quality. First, after the milling the atomically connected S-Sm interfaces appear, resulting in a large induced gap, as well as the ballistic transport revealed by the multiple Andreev reflections and quantized above-gap conductance plateaus. Second, by comparing transport measurement with Schrödinger-Poisson (SP) calculations, we demonstrate that argon milling is capable of varying the band bending strength in the semiconducting wire as the electrons tend to accumulate on the etched surface for longer milling time. Finally, we perform nonlocal measurements on advanced devices to demonstrate the coexistence and tunability of crossed Andreev reflection (CAR) and elastic co-tunneling (ECT) -- key ingredients for building the prototype setup for realization of Kitaev chain and quantum entanglement probing. Such a versatile method, compatible with the standard fabrication process and accompanied by the well-controlled modification of the interface, will definitely boost the creation of more sophisticated hybrid devices for exploring physics in solid-state systems.

cond-mat.mes-hall↗

arXiv2024

Probing Perfection: The Relentless Art of Meddling for Pulmonary Airway Segmentation from HRCT via a Human-AI Collaboration Based Active Learning Method

In pulmonary tracheal segmentation, the scarcity of annotated data is a prevalent issue in medical segmentation. Additionally, Deep Learning (DL) methods face challenges: the opacity of 'black box' models and the need for performance enhancement. Our Human-Computer Interaction (HCI) based models (RS_UNet, LC_UNet, UUNet, and WD_UNet) address these challenges by combining diverse query strategies with various DL models. We train four HCI models and repeat these steps: (1) Query Strategy: The HCI models select samples that provide the most additional representative information when labeled in each iteration and identify unlabeled samples with the greatest predictive disparity using Wasserstein Distance, Least Confidence, Entropy Sampling, and Random Sampling. (2) Central line correction: Selected samples are used for expert correction of system-generated tracheal central lines in each training round. (3) Update training dataset: Experts update the training dataset after each DL model's training epoch, enhancing the trustworthiness and performance of the models. (4) Model training: The HCI model is trained using the updated dataset and an enhanced UNet version. Experimental results confirm the effectiveness of these HCI-based approaches, showing that WD-UNet, LC-UNet, UUNet, and RS-UNet achieve comparable or superior performance to state-of-the-art DL models. Notably, WD-UNet achieves this with only 15%-35% of the training data, reducing physician annotation time by 65%-85%.

eess.IV↗

arXiv2024

Discrete inverse problems with internal functionals

We study the problem of finding the resistors in a resistor network from measurements of the power dissipated by the resistors under different loads. We give sufficient conditions for local uniqueness, i.e. conditions that guarantee that the linearization of this non-linear inverse problem admits a unique solution. Our method is inspired by a method to study local uniqueness of inverse problems with internal functionals in the continuum, where the inverse problem is reformulated as a redundant system of differential equations. We use our method to derive local uniqueness conditions for other discrete inverse problems with internal functionals including a discrete analogue of the inverse Schrödinger problem and problems where the resistors are replaced by impedances and dissipated power at the zero and a positive frequency are available. Moreover, we show that the dissipated power measurements can be obtained from measurements of thermal noise induced currents.

math.AP↗

arXiv2024

Can Generative AI Replace Immunofluorescent Staining Processes? A Comparison Study of Synthetically Generated CellPainting Images from Brightfield

Cell imaging assays utilizing fluorescence stains are essential for observing sub-cellular organelles and their responses to perturbations. Immunofluorescent staining process is routinely in labs, however the recent innovations in generative AI is challenging the idea of IF staining are required. This is especially true when the availability and cost of specific fluorescence dyes is a problem to some labs. Furthermore, staining process takes time and leads to inter-intra technician and hinders downstream image and data analysis, and the reusability of image data for other projects. Recent studies showed the use of generated synthetic immunofluorescence (IF) images from brightfield (BF) images using generative AI algorithms in the literature. Therefore, in this study, we benchmark and compare five models from three types of IF generation backbones, CNN, GAN, and diffusion models, using a publicly available dataset. This paper not only serves as a comparative study to determine the best-performing model but also proposes a comprehensive analysis pipeline for evaluating the efficacy of generators in IF image synthesis. We highlighted the potential of deep learning-based generators for IF image synthesis, while also discussed potential issues and future research directions. Although generative AI shows promise in simplifying cell phenotyping using only BF images with IF staining, further research and validations are needed to address the key challenges of model generalisability, batch effects, feature relevance and computational costs.

eess.IV↗

arXiv2024

Groupwise Deformable Registration of Diffusion Tensor Cardiovascular Magnetic Resonance: Disentangling Diffusion Contrast, Respiratory and Cardiac Motions

Diffusion tensor based cardiovascular magnetic resonance (DT-CMR) offers a non-invasive method to visualize the myocardial microstructure. With the assumption that the heart is stationary, frames are acquired with multiple repetitions for different diffusion encoding directions. However, motion from poor breath-holding and imprecise cardiac triggering complicates DT-CMR analysis, further challenged by its inherently low SNR, varied contrasts, and diffusion induced textures. Our solution is a novel framework employing groupwise registration with an implicit template to isolate respiratory and cardiac motions, while a tensor-embedded branch preserves diffusion contrast textures. We have devised a loss refinement tailored for non-linear least squares fitting and low SNR conditions. Additionally, we introduce new physics-based and clinical metrics for performance evaluation. Access code and supplementary materials at: https://github.com/ayanglab/DTCMR-Reg

eess.SP↗

arXiv2024

Fuzzy Attention-based Border Rendering Network for Lung Organ Segmentation

Automatic lung organ segmentation on CT images is crucial for lung disease diagnosis. However, the unlimited voxel values and class imbalance of lung organs can lead to false-negative/positive and leakage issues in advanced methods. Additionally, some slender lung organs are easily lost during the recycled down/up-sample procedure, e.g., bronchioles & arterioles, causing severe discontinuity issue. Inspired by these, this paper introduces an effective lung organ segmentation method called Fuzzy Attention-based Border Rendering (FABR) network. Since fuzzy logic can handle the uncertainty in feature extraction, hence the fusion of deep networks and fuzzy sets should be a viable solution for better performance. Meanwhile, unlike prior top-tier methods that operate on all regular dense points, our FABR depicts lung organ regions as cube-trees, focusing only on recycle-sampled border vulnerable points, rendering the severely discontinuous, false-negative/positive organ regions with a novel Global-Local Cube-tree Fusion (GLCF) module. All experimental results, on four challenging datasets of airway & artery, demonstrate that our method can achieve the favorable performance significantly.

eess.IV↗

arXiv2024

Data-driven methods for flow and transport in porous media: a review

This review examined the current advancements in data-driven methods for analyzing flow and transport in porous media, which has various applications in energy, chemical engineering, environmental science, and beyond. Although there has been progress in recent years, the challenges of current experimental and high-fidelity numerical simulations, such as high computational costs and difficulties in accurately representing complex, heterogeneous structures, can still potentially be addressed by state-of-the-art data-driven methods. We analyzed the synergistic potential of these methods, addressed their limitations, and suggested how they can be effectively integrated to improve both the fidelity and efficiency of current research. A discussion on future research directions in this field was conducted, emphasizing the need for collaborative efforts that combine domain expertise in physics and advanced computationald and data-driven methodologies.

physics.flu-dyn↗

arXiv2024

DiffExplainer: Unveiling Black Box Models Via Counterfactual Generation

In the field of medical imaging, particularly in tasks related to early disease detection and prognosis, understanding the reasoning behind AI model predictions is imperative for assessing their reliability. Conventional explanation methods encounter challenges in identifying decisive features in medical image classifications, especially when discriminative features are subtle or not immediately evident. To address this limitation, we propose an agent model capable of generating counterfactual images that prompt different decisions when plugged into a black box model. By employing this agent model, we can uncover influential image patterns that impact the black model's final predictions. Through our methodology, we efficiently identify features that influence decisions of the deep black box. We validated our approach in the rigorous domain of medical prognosis tasks, showcasing its efficacy and potential to enhance the reliability of deep learning models in medical image classification compared to existing interpretation methods. The code will be publicly available at https://github.com/ayanglab/DiffExplainer.

cs.CV↗

arXiv2024

Diff3Dformer: Leveraging Slice Sequence Diffusion for Enhanced 3D CT Classification with Transformer Networks

The manifestation of symptoms associated with lung diseases can vary in different depths for individual patients, highlighting the significance of 3D information in CT scans for medical image classification. While Vision Transformer has shown superior performance over convolutional neural networks in image classification tasks, their effectiveness is often demonstrated on sufficiently large 2D datasets and they easily encounter overfitting issues on small medical image datasets. To address this limitation, we propose a Diffusion-based 3D Vision Transformer (Diff3Dformer), which utilizes the latent space of the Diffusion model to form the slice sequence for 3D analysis and incorporates clustering attention into ViT to aggregate repetitive information within 3D CT scans, thereby harnessing the power of the advanced transformer in 3D classification tasks on small datasets. Our method exhibits improved performance on two different scales of small datasets of 3D lung CT scans, surpassing the state of the art 3D methods and other transformer-based approaches that emerged during the COVID-19 pandemic, demonstrating its robust and superior performance across different scales of data. Experimental results underscore the superiority of our proposed method, indicating its potential for enhancing medical image classification tasks in real-world scenarios.

eess.IV↗