arXiv subjects

Guilin Zhang

Publications and source records attributed to Guilin Zhang.

27 records · Page 2Linked to original sources

arXiv2025

A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice

A global shortage of radiologists has been exacerbated by the significant volume of chest X-ray workloads, particularly in primary care. Although multimodal large language models show promise, existing evaluations predominantly rely on automated metrics or retrospective analyses, lacking rigorous prospective clinical validation. Janus-Pro-CXR (1B), a chest X-ray interpretation system based on DeepSeek Janus-Pro model, was developed and rigorously validated through a multicenter prospective trial (NCT07117266). Our system outperforms state-of-the-art X-ray report generation models in automated report generation, surpassing even larger-scale models including ChatGPT 4o (200B parameters), while demonstrating reliable detection of six clinically critical radiographic findings. Retrospective evaluation confirms significantly higher report accuracy than Janus-Pro and ChatGPT 4o. In prospective clinical deployment, AI assistance significantly improved report quality scores, reduced interpretation time by 18.3% (P < 0.001), and was preferred by a majority of experts in 54.3% of cases. Through lightweight architecture and domain-specific optimization, Janus-Pro-CXR improves diagnostic reliability and workflow efficiency, particularly in resource-constrained settings. The model architecture and implementation framework will be open-sourced to facilitate the clinical translation of AI-assisted radiology solutions.

cs.AI↗

arXiv2025

AAPA: An Archetype-Aware Predictive Autoscaler with Uncertainty Quantification for Serverless Workloads on Kubernetes

Serverless platforms such as Kubernetes are increasingly adopted in high-performance computing, yet autoscaling remains challenging under highly dynamic and heterogeneous workloads. Existing approaches often rely on uniform reactive policies or unconditioned predictive models, ignoring both workload semantics and prediction uncertainty. We present AAPA, an archetype-aware predictive autoscaler that classifies workloads into four behavioral patterns -- SPIKE, PERIODIC, RAMP, and STATIONARY -- and applies tailored scaling strategies with confidence-based adjustments. To support reproducible evaluation, we release AAPAset, a weakly labeled dataset of 300,000 Azure Functions workload windows spanning diverse patterns. AAPA reduces SLO violations by up to 50% and lowers latency by 40% compared to Kubernetes HPA, albeit at 2-8x higher resource usage under spike-dominated conditions. To assess trade-offs, we propose the Resource Efficiency Index (REI), a unified metric balancing performance, cost, and scaling smoothness. Our results demonstrate the importance of modeling workload heterogeneity and uncertainty in autoscaling design.

cs.DC↗

arXiv2025

KIS-S: A GPU-Aware Kubernetes Inference Simulator with RL-Based Auto-Scaling

Autoscaling GPU inference workloads in Kubernetes remains challenging due to the reactive and threshold-based nature of default mechanisms such as the Horizontal Pod Autoscaler (HPA), which struggle under dynamic and bursty traffic patterns and lack integration with GPU-level metrics. We present KIS-S, a unified framework that combines KISim, a GPU-aware Kubernetes Inference Simulator, with KIScaler, a Proximal Policy Optimization (PPO)-based autoscaler. KIScaler learns latency-aware and resource-efficient scaling policies entirely in simulation, and is directly deployed without retraining. Experiments across four traffic patterns show that KIScaler improves average reward by 75.2%, reduces P95 latency up to 6.7x over CPU baselines, and generalizes without retraining. Our work bridges the gap between reactive autoscaling and intelligent orchestration for scalable GPU-accelerated environments.

cs.DC↗

arXiv2025

From Bench to Bedside: A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice

A global shortage of radiologists has been exacerbated by the significant volume of chest X-ray workloads, particularly in primary care. Although multimodal large language models show promise, existing evaluations predominantly rely on automated metrics or retrospective analyses, lacking rigorous prospective clinical validation. Janus-Pro-CXR (1B), a chest X-ray interpretation system based on DeepSeek Janus-Pro model, was developed and rigorously validated through a multicenter prospective trial (NCT06874647). Our system outperforms state-of-the-art X-ray report generation models in automated report generation, surpassing even larger-scale models including ChatGPT 4o (200B parameters), while demonstrating robust detection of eight clinically critical radiographic findings (area under the curve, AUC > 0.8). Retrospective evaluation confirms significantly higher report accuracy than Janus-Pro and ChatGPT 4o. In prospective clinical deployment, AI assistance significantly improved report quality scores (4.37 vs. 4.11, P < 0.001), reduced interpretation time by 18.5% (P < 0.001), and was preferred by a majority of experts (3 out of 5) in 52.7% of cases. Through lightweight architecture and domain-specific optimization, Janus-Pro-CXR improves diagnostic reliability and workflow efficiency, particularly in resource-constrained settings. The model architecture and implementation framework will be open-sourced to facilitate the clinical translation of AI-assisted radiology solutions.

cs.HC↗

arXiv2025

AMP4EC: Adaptive Model Partitioning Framework for Efficient Deep Learning Inference in Edge Computing Environments

Edge computing facilitates deep learning in resource-constrained environments, but challenges such as resource heterogeneity and dynamic constraints persist. This paper introduces AMP4EC, an Adaptive Model Partitioning framework designed to optimize deep learning inference in edge environments through real-time resource monitoring, dynamic model partitioning, and adaptive task scheduling. AMP4EC features a resource-aware model partitioner that splits deep learning models based on device capabilities, a task scheduler that ensures efficient load balancing using a weighted scoring mechanism, and a Docker-based deployment environment for validation. Experimental results show up to a 78% reduction in latency and a 414% improvement in throughput compared to baseline methods. The framework achieves consistent performance with low scheduling overhead across varying resource profiles, demonstrating adaptability in high-resource (1 CPU, 1GB RAM) and low-resource (0.4 CPU, 512MB RAM) scenarios. These results highlight AMP4EC's scalability, efficiency, and robustness for real-world edge deployments, addressing the critical need for efficient distributed inference in dynamic, resource-constrained environments.

cs.DC↗

arXiv2018

Vehicle Pose and Shape Estimation through Multiple Monocular Vision

In this paper, we present an accurate approach to estimate vehicles' pose and shape from off-board multiview images. The images are taken by monocular cameras and have small overlaps. We utilize state-of-the-art convolutional neural networks (CNNs) to extract vehicles' semantic keypoints and introduce a Cross Projection Optimization (CPO) method to estimate the 3D pose. During the iterative CPO process, an adaptive shape adjustment method named Hierarchical Wireframe Constraint (HWC) is implemented to estimate the shape. Our approach is evaluated under both simulated and real-world scenes for performance verification. It's shown that our algorithm outperforms other existing monocular and stereo methods for vehicles' pose and shape estimation. This approach provides a new and robust solution for off-board visual vehicle localization and tracking, which can be applied to massive surveillance camera networks for intelligent transportation.

cs.CV↗

arXiv2017

Simulation of Physical Parameters for a Photoneutron Source

A compact photoneutron source (PNS), based on an electron linac was designed and constructed to provide required nuclear data for the design of Thorium Molten Salt Reactor (TMSR). Many local shielding are built to reduce the background of neutron and γ rays, making the location of the time of flight (TOF) detector be fixed at 6.2 m place. Under the existing layout, some physical parameters are very difficult to get by the experiments, while can be obtained by the Monte Carlo simulation method. However, for the deep penetration problem of the neutron and γ rays transport in the channel of PNS with complex geometry, the normal Monte Carlo method is inefficient since electron transport calculation need a large amount of computing time and neutrons have little contribution to the detector in far-source region. In this work, the subsection method is applied in the simulation for PNS, which divide the simulation process in two steps, recording the neutron and γ rays information passing through the source window in the first step and adopting the covariance reduction techniques in the second step. The simulated neutron flux and energy spectrum at the TOF detector place with the relative error 1.6% are well agreement with the experimental results, achieving an efficiency 23 times better than the normal method. This method is fast and efficient in predicting the physical parameters, providing a required verification and initiating the foreseen physics experiment.

physics.comp-ph↗

arXiv2017

Measurements of the Total Cross section for Thermal Neutrons at PNS

In order to measure the total cross section for thermal neutrons, a photoneutron source (PNS, phase 1) has been developed for the acquisition of nuclear data for the Thorium Molten Salt Reactor (TMSR) at the Shanghai Institute of Applied Physics (SINAP). PNS is an electron LINAC pulsed neutron facility that uses the time-of-flight (TOF) technique. It records the neutron TOF and identifies neutrons and $γ$-rays by using a digital signal processing technique. The background is obtained by using a combination of employing 12.8 cm boron-loaded polyethylene(PEB) (5$\%$ w.t.) to block the flight path and Monte Carlo methods. The neutron total cross sections of natural beryllium are measured in the neutron energy region from 0.007 to 0.1 eV. The present measurement result is compared with the fold Harvey data with the response function of PNS.

nucl-ex↗

arXiv2016

TOF spectroscopy measurement using waveform digitizer

The photoneutron source (PNS, phase 1), an electron linear accelerator (linac)-based pulsed neutron facility that uses the time-of-flight (TOF) technique, was constructed for the acquisition of nuclear data from the thorium molten salt reactor(TMSR) at the Shanghai Institute of Applied Physics (SINAP). The neutron detector signal, with the information on the pulse arrival time, pulse shape, and pulse height, was recorded by using a waveform digitizer (WFD). By using the pulse height and pulse-shape discrimination (PSD) analysis to identify neutrons and $γ$-rays, the neutron TOF spectrum was obtained by employing a simple electronic design, and a new WFD-based DAQ system was developed and tested in this commissioning experiment. The developed DAQ system is characterized by a very high efficiency with respect to millisecond neutron TOF spectroscopy

physics.ins-det↗