To comply with privacy regulations and self-protection from cloud outages, increasingly more users are leveraging multiple cloud providers for service deployment. However, due to the isolation of highly Non-Identically and Independently Distributed (Non-IID) workload datasets and deficiencies of existing methods as reflected in our experimental studies, no cloud providers can single-handedly capture the workload patterns of such users for accurate workload prediction. Accordingly, we proposeP3Forecast, aPersonalizedPrivacy-Preserving Cloud workload prediction framework based on Federated Generative Adversarial Networks (GANs), which allows cloud providers with Non-IID workload data to collaboratively train workload prediction models as preferred while protecting privacy. We first design a data synthesis quality assessment method based on Dynamic Time Warping (calledpattern-aware DTW), which is insusceptible to time series length and reliable for the comparison of temporal patterns. By usingpattern-aware DTWas the model aggregation weights, we adopt the Federated Learning (FL) of a GAN model for the augmentation of IID workload training datasets per cloud provider. Then, we further design a post-training method of local workload prediction models, which consists of a query mechanism based on comprehensive evaluation of data synthesis informativeness and an adaptive learning rate adjustment strategy for stable convergence. Extensive experiments driven by real-world workloads demonstrate that compared with the state-of-the-art,P3Forecastimproves workload prediction accuracy by 23.7%-64.5% on average over all cloud providers, while ensuring the fastest convergence in Federated GAN training.

Zhu Xu, Zhiqiang Zhao, Zihan Zhang, Yuchi Liu, Quanwei Shen, Fei Liu, Yu Kuang, Jian He, Conglin Liu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

To comply with privacy regulations and selfprotection from cloud outages, increasingly more users are leveraging multiple cloud providers for service deployment. However, due to the isolation of highly Non-Identically and Independently Distributed (Non-IID) workload datasets and deficiencies of existing methods as reflected in our experimental studies, no cloud providers can single-handedly capture the workload patterns of such users for accurate workload prediction. Accordingly, we propose$\boldsymbol{P}^{3}$Forecast, a Personalized Privacy-Preserving Cloud workload prediction framework based on Federated Generative Adversarial Networks (GANs), which allows cloud providers with Non-IID workload data to collaboratively train workload prediction models as preferred while protecting privacy. We first design a data synthesis quality assessment method based on Dynamic Time Warping (called pattern-aware DTW), which is insusceptible to time series length and reliable for the comparison of temporal patterns. By using pattern-aware DTW as the model aggregation weights, we adopt the Federated Learning (FL) of a GAN model for the augmentation of IID workload training datasets per cloud provider. Then, we further design a post-training method of local workload prediction models, which consists of a query mechanism for extracting the most informative synthesized data for training dataset augmentation and a learning rate adjustment strategy for stable convergence. Extensive experiments driven by real-world workloads demonstrate that compared with the state-of-the-art,$P^{3}$Forecast improves workload prediction accuracy by$19.5 \%-46.7 \%$in average over all cloud providers, while ensuring the fastest convergence in Federated GAN training. The codes can be found at https://github.com/liyan2015/P3Forecast.

A number of deep neural network (DNN)-based models have been applied to help classify and detect severe cardiovascular diseases using 12-lead electrocardiogram (ECG) signals. These models, however, suffer from poor performance in detecting one or two specific cardiac abnormalities, like ST-segment abnormalities, with their accuracy lingering at only around 60%, which in turn limits their applicability in clinical practice. In this paper, we show that the convolution layers of these DNN models can cause the diminishment of the key features of ST-segment abnormalities, making it, compared to cardiac arrhythmia and normal sinus rhythm (NSR), hard to be classified out from the ECG. Correspondingly, we, in this paper, propose a novel DNN-based model that is able to achieve high accuracy of detecting 9 classes of rhythms. In specific, the moving averages of the ECG signals are used as the second input to the proposed model so that it can help fully mine the features relevant to the ST abnormalities. In addition, as opposed to the convolution layers, our model utilizes the inception-residual layers to preserve features from the shallow layers for reuse in the deep layers. On top of these network architecture improvements, we further introduce a customized differential layer so that all the relevant features can be preserved and/or amplified for classification purposes. Trained with CPSC 2018 dataset, our proposed model is able to accurately classify the 9 rhythm classes, with an F1 score as high as 89.7%, up by merely 4.2% from the best result reported in the literature. As far as the ST segment abnormalities are concerned, the proposed model achieves an 89.6% F1 score for STD and an 80.8% F1 score for STE, which is 7.8% and 13.1% respectively higher than that of the best result. The proposed method is thus poised to become a viable solution for cardiovascular health monitoring with the increasing availability of portable and home-based ECG devices.

This study investigated the feasibility of a highly specific multiplexed image-guided small animal radiation therapy (SART) platform based on triple imaging from on-board single-photon emission computed tomography (SPECT), spectral-CT, and cone-beam CT (CBCT) guidance in radiotherapy treatment. As a proof-of-concept, the SART system was built with the capability of triple on-board image guidance by utilizing an x-ray tube and a single cadmium zinc telluride (CZT) semiconductor photon-counting imager via a Monte Carlo simulation study. The x-ray tube can be set at a low tube current for imaging mode and a high tube current for radiation therapy mode, respectively. In the imaging mode, both x-ray and gamma-ray projection data were collected by the imager to reconstruct CBCT, SPECT and spectral CT images of small animals being treated. The modulation transfer function (MTF) of the pixelated CZT imager measured was 8.6 lp/mm. The overall performances of the CBCT and SPECT imaging of the system were evaluated with sufficient spatial resolution and imaging quality to be fitted into the SART platform. The material differentiation and decomposition capacities of spectral CT within the system were verified using K-edge imaging, image-based optimal energy weighted imaging, and image-based linear material decomposition methods. The triple imaging capability of the system was demonstrated using a PMMA phantom containing gadolinium, iodine and radioisotope99mTc inserts. All the probes were clearly identified in the registered image. The results demonstrated that a novel SART platform with high-quality on-board CBCT, spectral-CT, SPECT image guidance is technically feasible by using a single semiconductor imager, thus affording comprehensive image guidance from anatomical, functional, and molecular levels for radiation treatment beam delivery.

The GEANT4-based GATE is a unique and powerful Monte Carlo (MC) platform, which provides a single code library allowing the simulation of specific medical physics applications. However, this rigorous yet flexible platform is used only sparingly in the clinic due to its lengthy calculation time and significant computational overhead. By accessing the much more powerful computational resources of a cloud computing environment, GATE's run time can be significantly reduced to clinically feasible levels without the sizable investment of a local high performance cluster. This study investigated a reliable and efficient execution of GATE MC simulation using a commercial cloud computing services. A Monte Carlo cloud computing framework, Gate Cloud, for medical physics applications was proposed. Amazon's Elastic Compute Cloud (EC2) was used to launch several nodes equipped with GATE V6.1. The Positron emission tomography (PET) Benchmark included in the GATE software was repeated for various cluster sizes between 1 and 100 nodes in order to establish the ideal cluster size in terms of cost and time efficiency. The study shows that increasing the number of nodes in the cluster resulted in a decrease in calculation time that could be expressed with an inverse power model. Simulation results were not affected by the cluster size, indicating that integrity of a calculation is preserved in a cloud computing environment. With high power computing continuing to lower in price and accessibility, implementing Gate Cloud for clinical applications will continue to become more attractive.

Simultaneous imaging of multiple probes or biomarkers represents a critical step toward high specificity molecular imaging. In this work, we propose to utilize the element-specific nature of the X-ray fluorescence (XRF) signal for imaging multiple elements simultaneously (multiplexing) using XRF computed tomography (XFCT). A 5-mm-diameter pencil beam produced by a polychromatic X-ray source (150 kV, 20 mA) was used to stimulate emission of XRF photons from 2% (weight/volume) gold (Au), gadolinium (Gd), and barium (Ba) embedded within a water phantom. The phantom was translated and rotated relative to the stationary pencil beam in a first-generation CT geometry. The X-ray energy spectrum was collected for 18 s at each position using a cadmium telluride detector. The spectra were then used to isolate the K shell XRF peak and to generate sinograms for the three elements of interest. The distribution and concentration of the three elements were reconstructed with the iterative maximum likelihood expectation maximization algorithm. The linearity between the XFCT intensity and the concentrations of elements of interest was investigated. We found that measured XRF spectra showed sharp peaks characteristic of Au, Gd, and Ba. The narrow full-width at half-maximum (FWHM) of the peaks strongly supports the potential of XFCT for multiplexed imaging of Au, Gd, and Ba ( FWHM(Au,Kα1) = 0.619 keV, FWHM(Au,Kα2)=1.371 keV , FWHM(Gd,Kα)=1.297 keV, FWHM(Gd,Kβ)=0.974 keV , FWHM(Ba,Kα)=0.852 keV, and FWHM(Ba,Kβ)=0.594 keV ). The distribution of Au, Gd, and Ba in the water phantom was clearly identifiable in the reconstructed XRF images. Our results showed linear relationships between the XRF intensity of each tested element and their concentrations ( R(2)(Au)=0.944 , R(Gd)(2)=0.986, and R(Ba)(2)=0.999), suggesting that XFCT is capable of quantitative imaging. Finally, a transmission CT image was obtained to show the potential of the approach for providing attenuation correction and morphological information. In conclusion, XFCT is a promising modality for multiplexed imaging of high atomic number probes.

This work provides a comprehensive Monte Carlo study of X-ray fluorescence computed tomography (XFCT) and K-edge imaging system, including the system design, the influence of various imaging components, the sensitivity and resolution under various conditions. We modified the widely used EGSnrc/DOSXYZnrc code to simulate XFCT images of two acrylic phantoms loaded with various concentrations of gold nanoparticles and Cisplatin for a number of XFCT geometries. In particular, reconstructed signal as a function of the width of the detector ring, its angular coverage and energy resolution were studied. We found that XFCT imaging sensitivity of the modeled systems consisting of a conventional X-ray tube and a full 2-cm-wide energy-resolving detector ring was 0.061% and 0.042% for gold nanoparticles and Cisplatin, respectively, for a dose of ∼ 10 cGy. Contrast-to-noise ratio (CNR) of XFCT images of the simulated acrylic phantoms was higher than that of transmission K-edge images for contrast concentrations below 0.4%.