Searcharxiv⌕ Search

arXiv subjects

Songlin Zhao

Publications and source records attributed to Songlin Zhao.

18 recordsLinked to original sources

Interpreting Selective-Review Coverage for Head-and-Neck Auto-Contours

Selective-review systems aim to reduce human review by passing candidates predicted to be low risk. However, coverage (the fraction passed) can be misleading when measurements needed to interpret risk are unavailable. We separate structural evaluability (a nonempty mask with all required measurements available), computational-endpoint status, and pass-or-review decisions using archived Mayo records and controlled OpenKBP analyses. In 3,968 Mayo records, all 1,167 recorded empty predictions (predicted target-organ masks containing zero voxels despite nonempty reference masks) were endpoint-negative and passed both modules, accounting for approximately 76% of each pass set. In a held-out OpenKBP test, a logistic-regression router achieved an area under the receiver operating characteristic curve of 0.833 and average precision of 0.449 but passed all 544 non-evaluable candidates, including 542 prespecified empty-mask controls and 2 erosions that became empty. A structural gate removed those passes without changing scores or the threshold, preserved all evaluable passes, and assigned 98.71% of endpoint-positive candidates to review. Under the original preprocessing and in patient-disjoint development evaluation, a router trained only on evaluable candidates and a router modeling an unavailable state likewise passed no non-evaluable candidates, with similar evaluable discrimination and at least 99.0% of endpoint-positive candidates assigned to review in pooled outer-fold predictions. These results show that coverage should be interpreted jointly with candidate eligibility, endpoint scope, missingness handling, and routing rules; they do not establish clinical validity, deployment safety, or natural-failure prevalence.

eess.IV↗

Supervision-Aware Evaluation of MedGemma for Reference-Assisted Radiotherapy Contour Ranking

Radiotherapy contour ranking can support quality assurance, but vision-language model (VLM) gains may reflect supervision, score construction, available inputs, or representation access. We evaluated frozen MedGemma 1.5 in a single-institution study, a 12-patient held-out HaN-Seg test, five-fold cross-fitting on 30 non-test HaN-Seg patients, and transfer without refitting to 120 SegRap2023 patients. On held-out HaN-Seg, a supervised readout of the final prompt state exceeded a supervised readout of the model's yes/no output probabilities by +0.3047 pooled average precision (AP; 95% patient-bootstrap interval [+0.2822, +0.3269]); the advantage persisted against a parameter-count-matched nonlinear control. A readout of projected visual tokens exceeded the final-state readout in cross-fitting and transfer (+0.0706 AP averaged equally across patients on SegRap; 95% interval [+0.0607, +0.0804]), while a structured panel-feature readout remained competitive with visual tokens. The anatomy-conditioned readout did not demonstrate an advantage over the native-feature comparator. These results motivate visual-token and structured-feature controls in medical-imaging VLM evaluations; outcomes were computational or synthetic, not clinical acceptability.

eess.IV↗

RadOnc-Agent: An Integrated Framework for AI Workflows Across the Radiotherapy Care Pathway

Artificial intelligence has advanced radiotherapy tasks, but these capabilities remain fragmented across the care pathway. Here we present RadOnc-Agent, an integrated framework connecting 17 phase-specific functions and 9 cross-cutting utilities across four phases through a conversational interface. A language-model controller maps clinical intent to structured calls, maintains longitudinal context and validates dispatch. We evaluated request-expression diversity, function coverage, cross-stage composition, information transfer and failure recovery using 2,600 single-function requests (7,800 executions), 200 cross-stage scenarios (600 executions), and 120 workflows from 60 de-identified patient records (360 executions). RadOnc-Agent selected the intended function in 98.79% of single-function executions and completed 96.50% of scripted and 96.67% of real-patient workflows. Removing longitudinal state reduced cross-stage completion from 96.50% to 84.00%; bypassing schema and identity validation increased mismatched dispatch from 0% to 95.28% in replay/test execution. Single-function performance did not by itself predict workflow-level capability. Information continuity and cross-stage dependencies emerged as central constraints. These results establish the technical feasibility of longitudinal integration across radiotherapy workflows and provide a foundation for prospective clinical evaluation.

cs.AI↗

Personalising Deformable Image Registration with Clinical Context and Deformation History for Adaptive Radiotherapy

Adaptive radiotherapy must account for anatomical changes during treatment, but doing so depends on reliably mapping targets, organs at risk and dose across serial images. Existing registration methods treat each image pair independently, without using radiotherapy-specific context or the patient's preceding anatomical trajectory. We developed Clinical-context-Aware and Patient-adaptive Deformable Image Registration (CAP-DIR), integrating source contours, dose, clinical text and prior-registration features. We evaluated 16,361 CT examinations from 3,473 patients across four institutional cohorts and five public benchmarks. CAP-DIR increased internal-test organ-at-risk-region normalized cross-correlation from 84.76% to 96.92% relative to the strongest learning-based comparator, with low deformation folding, and remained superior across seven external cohorts. In a blinded review of 20 longitudinal cases, clinically acceptable contours increased from 45.0% to 82.5% for clinical target volumes and from 22.5% to 62.5% for organs at risk, while estimated editing time decreased. In a controlled benchmark, CAP-DIR reduced global dose-mapping error by 34.5% relative to the strongest comparator. These findings show that clinically informed, longitudinal registration can improve contour usability and dose mapping for adaptive radiotherapy.

eess.IV↗

EfficientXpert: Efficient Domain Adaptation for Large Language Models via Propagation-Aware Pruning

Deploying domain-specialized large language models on resource-constrained hardware motivates reducing both adaptation cost and the number of retained weights. LoRA lowers adaptation cost but leaves a dense backbone, while separate pruning can discard weights that become important during adaptation. We propose EfficientXpert, a framework that co-adapts low-rank updates and sparse support. Its ForeSight Mask scores weights through a downstream reconstruction surrogate using the evolving LoRA-augmented weights; Partial Brain Surgeon (PBS) realigns the adapter through a closed-form correction. At 40% sparsity, the strongest variant retains 98.55--108.99% of dense-LoRA aggregate performance across the evaluated LLaMA settings. On Qwen3-8B, adaptation adds 17.9--27.0% training time and at most 4.8% peak GPU memory. Our analysis connects the domain- and rank-dependent effects of PBS to its constrained reconstruction objective, providing guidance for selecting adapter recovery capacity. Together, these results demonstrate efficient production of sparse, domain-specialized experts. Code is available at https://github.com/TagoreZhao/Efficient_Domain_Adaptation.

cs.LG↗

Supervised Deep Multimodal Matrix Factorization for Interpretable Brain Network Analysis

Multimodal brain network analysis faces a persistent trade-off between predictive accuracy and interpretability. Deep neural networks achieve high accuracy but behave as black boxes that reveal little about the brain modules driving their decisions, whereas matrix factorization methods provide parts-based interpretability yet remain largely shallow, unsupervised, and restricted to a single view, integrating modalities through predefined or heuristic fusion rules. To bridge this gap with a formulation that couples hierarchical modeling capacity with structured, interpretable representations and data-driven fusion, we present Supervised Deep Multimodal Matrix Factorization (SD3MF), an interpretable framework for integrative brain network analysis that generalizes Symmetric Nonnegative Matrix Tri-Factorization (SNMTF) from unsupervised single-graph clustering to supervised prediction over populations of multimodal graphs. SD3MF learns deep hierarchical factorizations for each modality together with a shared latent representation that aligns subjects across modalities. An encoder-decoder formulation jointly optimizes graph reconstruction and supervised prediction, while adaptive weights enable data-driven multimodal fusion. By representing each subject through community-level interaction matrices, the model yields interpretable and discriminative features. Experiments on multimodal connectome datasets show that SD3MF consistently outperforms strong deep learning baselines such as {Convolutional Neural Networks and Graph Neural Networks}, while enabling biologically interpretable insights. Code for reproducibility is available at https://github.com/amjadseyedi/SD3MF.

cs.LG↗

MedGPT-oss: Training a General-Purpose Vision-Language Model for Biomedicine

Biomedical multimodal assistants have the potential to unify radiology, pathology, and clinical-text reasoning, yet a critical deployment gap remains: top-performing systems are either closed-source or computationally prohibitive, precluding the on-premises deployment required for patient privacy and PHI compliance. We introduce MEDGPT-OSS, an open-weight, 20B-parameter generalist vision-language model designed to facilitate open research in clinical AI. Rather than relying on architectural complexity, MEDGPT-OSS pairs the GPT-oss language backbone with a visual front-end via a optimized, three-stage training curriculum. By progressively domain-adapting these modules through rigorous data curation and long-context multimodal alignment, we demonstrate that a 20B model can bridge the capacity gap. It successfully outperforms larger open medical models on out-of-distribution (OOD) multimodal reasoning and complex text-only clinical tasks. By unifying diverse modalities under a single instruction-following interface, MEDGPT-OSS maintains a parameter-efficient footprint fully compatible with commodity GPUs. We release the complete training recipe, open-weight checkpoints, and a rigorous evaluation harness to serve as a verifiable foundation for privacy-preserving, institution-specific clinical AI research.

cs.CL↗

Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination

Recent video generators can fabricate realistic depictions of wars, disasters, public emergencies, and other real-world crises, creating substantial risks of misinformation. Existing benchmarks, however, provide limited evidence on detector and generator behavior in such settings, including how detectability varies with generation conditions, how people perceive generated videos, and whether detectors remain reliable during social dissemination. To address this gap, we introduce RA-Bench, a benchmark for AI-generated video detection that uses Real videos as Anchors. RA-Bench contains 17,886 videos, comprising 1,830 real-video anchors across 10 social-risk categories and 16,056 generated clips from four open-source and five closed-source generators. Based on RA-Bench, we organize our evaluation along three dimensions. We first assess detector generalization across seven traditional detectors, ten zero-shot multimodal models under three review settings, and two MLLMs specifically fine-tuned on AI-generated video detection. Across these methods, none of the three detector families generalizes consistently across RA-Bench instances. We then examine how detectability varies with generation quality, conditioning information, and sampling seeds. These analyses show that generation properties affect detector families differently, while source-level detection patterns remain stable across seeds. Finally, we study human authenticity judgments and detector reliability during social dissemination. We find that videos that mislead people are also difficult for current detectors, and that social dissemination makes detection harder. Together, these findings show that current methods struggle to detect realistic AI-generated videos, highlighting the need for detectors robust to evolving video generators.

cs.CV↗

Towards a Virtual Neuroscientist: Autonomous Neuroimaging Analysis via Multi-Agent Collaboration

Transforming neuroimaging data into clinically actionable biomarkers is a knowledge-intensive and labor-intensive process. Standardized workflows such as fMRIPrep have improved robustness and efficiency, but they are statically configured and cannot reason about downstream objectives, deliberate over alternative strategies, or close the loop between intermediate evidence and subsequent decisions in the way a human researcher would. This lack of closed-loop adaptation often leaves domain experts trapped in a cycle of manual trial-and-error to tune parameters and remediate pipeline failures, severely constraining the scalability of clinical biomarker development. To bridge this gap, we introduce NEXUS, an autonomous multi-agent framework that integrates neuroimaging workflow execution with scientific-objective understanding. Unlike conventional flat toolcalling agents, NEXUS adopts a code-centric execution paradigm where specialist agents collaboratively synthesize and optimize executable programs over composable domain-specific primitives. This design enables robust, long-horizon workflow construction that adapts dynamically to runtime observations. Furthermore, we propose a hierarchical verification framework for autonomous quality control, integrating cohort-level metric screening with agentic visual inspection to drive evidence-grounded workflow remediation. Experiments on ADHD-200 and ADNI demonstrate that NEXUS outperforms standard workflow-based baselines in predictive performance while exhibiting sophisticated agentic behaviors, including strategy exploration and adaptive refinement. The code is available at https://github.com/LearningKeqi/Virtual-Neuroscientist-NEXUS.

cs.AI↗

Homodyne Photonic Tensor Processor exceeds 1,000-TOPS

High-performance computing underpins modern artificial intelligence (AI), enabling foundation models, real-time inference and perception in autonomous systems, and data-intensive scientific simulations. Recent advances in quantization techniques utilizing low-precision computation without degrading model accuracy, create new opportunities for analog photonic computing characterized by ultra-high clock rates and low energy consumption. Here we propose and demonstrate a coherent homodyne integrated circuit capable of general matrix multiplication (GEMM) with aggregate throughput that exceeds 1,000 TOPS (tera-operations per second), enabled by massive on-chip optical fanout and parallelism. By leveraging time multiplexing, the required modulator count is reduced from O($N^2$) to O(N), allowing dense integration of record-scale 256 $\times$ 256 homodyne units (each <0.0064 $mm^2$) within a single reticle. We employ wafer-scale fabricated 64 thin-film lithium niobate (TFLN) transmitters (each over 40-GHz bandwidth with propagation loss of 0.2 dB/cm) to encode data and chip-to-chip coupled to Si/SiN computing circuits (64 channels). Our system achieves up to 7-bit computational accuracy across 8 $\times$ 8 parallel channels at record computing clockrate 120 Gbaud/s, and 6-bit statistical accuracy across 256 $\times$ 100 channels at 20-128 Gbaud/s, representing a total throughput of 1,000-6,000 TOPS. Massive parallelism amortizes the optoelectronic (OE) conversion to allow 330-TOPS/W efficiency using foundry-available packaging technology. The system throughput is benchmarked with Qwen2.5-0.5 billion parameter models that generate accurate tokens. High throughput and energy efficiency establish a near-term pathway toward light-based accelerators for large-scale training and low-latency inference from datacenters to edges, accelerating new models toward artificial general intelligence.

cs.ET↗

Quantization-aware Photonic Homodyne computing for Accelerated Artificial Intelligence and Scientific Simulation

Modern problems in high-performance computing, ranging from training and inferencing deep learning models in computer vision and language models to simulating complex physical systems with nonlinearly-coupled equations, require exponential growth of computational resources. Photonic analog systems are emerging with solutions of intrinsic parallelism, high bandwidth, and low propagation loss. However, their application has been hindered by the low analog accuracy due to the electro-optic distortion, material nonlinearities, and signal-to-noise ratios. Here we overcome this barrier with a quantization-aware digital-photonic mixed-precision framework across chiplets for accelerated AI processing and physical simulation. Using Lithium Niobate photonics with channel equalization techniques, we demonstrate linear multiplication (9-bit amplitude-phase decoupling) in homodyne optical logics with 6-bit precision at the clock rate of 128 giga-symbol-per-second (128 GS/s), enabling AI processing with 6 ns latency. Codesign hardware-algorithms, including iterative solvers, sparse-dense quantization, and bit-sliced matrix multiplication, explore photonic amplitude and phase coherence for complex-valued, physics-inspired computation. In electromagnetic problems, our approach yields 12-bit solutions for partial differential equations (PDEs) in scattering problems that would conventionally require up to 32-bit and often even 64-bit precision. These results preserve digital-level fidelity while leveraging the high-speed low-energy photonic hardware, establishing a pathway toward general-purpose optical acceleration for generative artificial intelligence, real-time robotics, and accurate simulation for climate challenges and biological discoveries.

cs.ET↗

Is the medical image segmentation problem solved? A survey of current developments and future directions

Medical image segmentation has advanced rapidly over the past two decades, largely driven by deep learning, which has enabled accurate and efficient delineation of cells, tissues, organs, and pathologies across diverse imaging modalities. This progress raises a fundamental question: to what extent have current models overcome persistent challenges, and what gaps remain? In this work, we provide an in-depth review of medical image segmentation, tracing its progress and key developments over the past decade. We examine core principles, including multiscale analysis, attention mechanisms, and the integration of prior knowledge, across the encoder, bottleneck, skip connections, and decoder components of segmentation networks. Our discussion is organized around seven key dimensions: (1) the shift from supervised to semi-/unsupervised learning, (2) the transition from organ segmentation to lesion-focused tasks, (3) advances in multi-modality integration and domain adaptation, (4) the role of foundation models and transfer learning, (5) the move from deterministic to probabilistic segmentation, (6) the progression from 2D to 3D and 4D segmentation, and (7) the trend from model invocation to segmentation agents. Together, these perspectives provide a holistic overview of the trajectory of deep learning-based medical image segmentation and aim to inspire future innovation. To support ongoing research, we maintain a continually updated repository of relevant literature and open-source resources at https://github.com/apple1986/medicalSegReview

eess.IV↗

Segment Anything for Video: A Comprehensive Review of Video Object Segmentation and Tracking from Past to Future

Video Object Segmentation and Tracking (VOST) presents a complex yet critical challenge in computer vision, requiring robust integration of segmentation and tracking across temporally dynamic frames. Traditional methods have struggled with domain generalization, temporal consistency, and computational efficiency. The emergence of foundation models like the Segment Anything Model (SAM) and its successor, SAM2, has introduced a paradigm shift, enabling prompt-driven segmentation with strong generalization capabilities. Building upon these advances, this survey provides a comprehensive review of SAM/SAM2-based methods for VOST, structured along three temporal dimensions: past, present, and future. We examine strategies for retaining and updating historical information (past), approaches for extracting and optimizing discriminative features from the current frame (present), and motion prediction and trajectory estimation mechanisms for anticipating object dynamics in subsequent frames (future). In doing so, we highlight the evolution from early memory-based architectures to the streaming memory and real-time segmentation capabilities of SAM2. We also discuss recent innovations such as motion-aware memory selection and trajectory-guided prompting, which aim to enhance both accuracy and efficiency. Finally, we identify remaining challenges including memory redundancy, error accumulation, and prompt inefficiency, and suggest promising directions for future research. This survey offers a timely and structured overview of the field, aiming to guide researchers and practitioners in advancing the state of VOST through the lens of foundation models.

cs.CV↗

SegImgNet: Segmentation-Guided Dual-Branch Network for Retinal Disease Diagnoses

Retinal image plays a crucial role in diagnosing various diseases, as retinal structures provide essential diagnostic information. However, effectively capturing structural features while integrating them with contextual information from retinal images remains a challenge. In this work, we propose segmentation-guided dual-branch network for retinal disease diagnosis using retinal images and their segmentation maps, named SegImgNet. SegImgNet incorporates a segmentation module to generate multi-scale retinal structural feature maps from retinal images. The classification module employs two encoders to independently extract features from segmented images and retinal images for disease classification. To further enhance feature extraction, we introduce the Segmentation-Guided Attention (SGA) block, which leverages feature maps from the segmentation module to refine the classification process. We evaluate SegImgNet on the public AIROGS dataset and the private e-ROP dataset. Experimental results demonstrate that SegImgNet consistently outperforms existing methods, underscoring its effectiveness in retinal disease diagnosis. The code is publicly available at https://github.com/hawk-sudo/SegImgNet.

eess.IV↗

Normative Modeling for AD Diagnosis and Biomarker Identification

In this paper, we introduce a novel normative modeling approach that incorporates focal loss and adversarial autoencoders (FAAE) for Alzheimer's Disease (AD) diagnosis and biomarker identification. Our method is an end-to-end approach that embeds an adversarial focal loss discriminator within the autoencoder structure, specifically designed to effectively target and capture more complex and challenging cases. We first use the enhanced autoencoder to create a normative model based on data from healthy control (HC) individuals. We then apply this model to estimate total and regional neuroanatomical deviation in AD patients. Through extensive experiments on the OASIS-3 and ADNI datasets, our approach significantly outperforms previous state-of-the-art methods. This advancement not only streamlines the detection process but also provides a greater insight into the biomarker potential for AD. Our code can be found at \url{https://github.com/soz223/FAAE}.

eess.IV↗

Rational solutions for three semi-discrete modified Korteweg-de Vries type equations

In this paper, we consider three semi-discrete modified Korteweg-de Vries type equations which are the nonlinear lumped self-dual network equation,the semi-discrete lattice potential modified Korteweg-de Vries equation and a semi-discrete modified Korteweg-de Vries equation. We derive several kinds of exact solutions, in particular rational solutions, in terms of the Casorati determinant for these three equations respectively. For some rational solutions, we present the related asymptotic analysis to understand their dynamics better.

nlin.SI↗

Regularized Kernel Recursive Least Square Algoirthm

In most adaptive signal processing applications, system linearity is assumed and adaptive linear filters are thus used. The traditional class of supervised adaptive filters rely on error-correction learning for their adaptive capability. The kernel method is a powerful nonparametric modeling tool for pattern analysis and statistical signal processing. Through a nonlinear mapping, kernel methods transform the data into a set of points in a Reproducing Kernel Hilbert Space. KRLS achieves high accuracy and has fast convergence rate in stationary scenario. However the good performance is obtained at a cost of high computation complexity. Sparsification in kernel methods is know to related to less computational complexity and memory consumption.

cs.LG↗

Generalized Cauchy matrix approach for non-autonomous discrete Kadomtsev-Petviashvili system

In this paper, we investigate the non-autonomous discrete Kadomtsev-Petviashvili (KP) system in terms of generalized Cauchy matrix approach. These equations include non-autonomous bilinear lattice KP equation, non-autonomous lattice potential KP equation, non-autonomous lattice potential modified KP equation, non-autonomous asymmetric lattice potential modified KP equation, non-autonomous lattice Schwarzian KP equation and non-autonomous lattice KP-type Nijhoff-Quispel-Capel equation. By introducing point transformations, all the equations are described as simplified forms, where the lattice parameters are absorbed. Several kinds of solutions more than multi-soliton solutions to these equations are derived by solving determining equation set. Lax representations for these equations are also discussed.

math-ph↗