Search

Lip reading is vital for robots in social settings, improving their ability to understand human communication. This skill allows them to communicate more easily in crowded environments, especially in caregiving and customer service roles. Generating a Persian Lip-reading dataset, this study integrates Persian lip-reading technology into the Surena-V humanoid robot to improve its speech recognition capabilities. Two complementary methods are explored, an indirect method using facial landmark tracking and a direct method leveraging convolutional neural networks (CNNs) and long short-term memory (LSTM) networks. The indirect method focuses on tracking key facial landmarks, especially around the lips, to infer movements, while the direct method processes raw video data for action and speech recognition. The best-performing model, LSTM, achieved 89\% accuracy and has been successfully implemented into the Surena-V robot for real-time human-robot interaction. The study highlights the effectiveness of these methods, particularly in environments where verbal communication is limited.

Estimation of patient acuity in the Intensive Care Unit (ICU) is vital to ensure timely and appropriate interventions. Advances in artificial intelligence (AI) technologies have significantly improved the accuracy of acuity predictions. However, prior studies using machine learning for acuity prediction have predominantly relied on electronic health records (EHR) data, often overlooking other critical aspects of ICU stay, such as patient mobility, environmental factors, and facial cues indicating pain or agitation. To address this gap, we present MANGO: the Multimodal Acuity traNsformer for intelliGent ICU Outcomes, designed to enhance the prediction of patient acuity states, transitions, and the need for life-sustaining therapy. We collected a multimodal dataset ICU-Multimodal, incorporating four key modalities, EHR data, wearable sensor data, video of patient's facial cues, and ambient sensor data, which we utilized to train MANGO. The MANGO model employs a multimodal feature fusion network powered by Transformer masked self-attention method, enabling it to capture and learn complex interactions across these diverse data modalities even when some modalities are absent. Our results demonstrated that integrating multiple modalities significantly improved the model's ability to predict acuity status, transitions, and the need for life-sustaining therapy. The best-performing models achieved an area under the receiver operating characteristic curve (AUROC) of 0.76 (95% CI: 0.72-0.79) for predicting transitions in acuity status and the need for life-sustaining therapy, while 0.82 (95% CI: 0.69-0.89) for acuity status prediction...

Reinforcement learning (RL) is one of the active fields in machine learning, demonstrating remarkable potential in tackling real-world challenges. Despite its promising prospects, this methodology has encountered with issues and challenges, hindering it from achieving the best performance. In particular, these approaches lack decent performance when navigating environments and solving tasks with large observation space, often resulting in sample-inefficiency and prolonged learning times. This issue, commonly referred to as the curse of dimensionality, complicates decision-making for RL agents, necessitating a careful balance between attention and decision-making. RL agents, when augmented with human or large language models' (LLMs) feedback, may exhibit resilience and adaptability, leading to enhanced performance and accelerated learning. Such feedback, conveyed through various modalities or granularities including natural language, serves as a guide for RL agents, aiding them in discerning relevant environmental cues and optimizing decision-making processes. In this survey paper, we mainly focus on problems of two-folds: firstly, we focus on humans or an LLMs assistance, investigating the ways in which these entities may collaborate with the RL agent in order to foster optimal behavior and expedite learning; secondly, we delve into the research papers dedicated to addressing the intricacies of environments characterized by large observation space.

The quantum Wigner crystal is a many-body state where Coulombic repulsion quenches the kinetic energy of electrons, causing them to crystallize into a lattice. Experimental realization of a quantum Wigner crystal at zero magnetic field has been a long-sought goal. Here, we report on the experimental evidence of a Wigner solid in ultra-thin films of cadmium arsenide (Cd3As2) at zero magnetic field. We show that a finite bias depins the domains and produces an unusually sharp threshold current-voltage behavior. Hysteresis and voltage fluctuations point to domain motion across the pinning potential and disappear at finite temperature as thermal fluctuations overcome the potential. The application of a small magnetic field destroys the Wigner solid, pointing to an unconventional origin. We use Landau level spectroscopy to show that the formation of the Wigner solid is closely connected to a topological transition as the film thickness is reduced.

Advanced algorithms based on Deep Reinforcement Learning (DRL) have been able to become a reliable tool for the Forex market traders and provide a suitable strategy for maximizing profit and reducing trading risk. These tools try to find the most profitable strategy in this market by examining past market data. Artificial intelligent agents based on the Proximal Policy Optimization (PPO) algorithm, one of the DRL algorithms, have shown a special ability to determine a profitable strategy. In this research, to increase profitability and determine the optimal strategy for the PPO, an auxiliary task has been used. The auxiliary function helps the PPO to model the reward function in a better way by recognizing and classifying patterns to obtain additional information from the problem's inputs and data. The results of simulation and backtesting on the EUR/USD currency pair have shown that the new proposed model has been able to increase the overall return from -25.25% to 14.86% and the value of sharp ratio from -2.61 to 0.24 in dataset 1 and increased overall return from 2.12% to 42.22% and the value of sharp ratio from -2.93 to 0.47 in dataset 2. These results indicate better identification of trends, timely trading, and great improvement in risk reduction while increasing the profit. It has also reduced the transaction risk and can be used as an effective tool for automatic trading in the complex forex market

The association between preoperative cognitive status and surgical outcomes is a critical, yet scarcely explored area of research. Linking intraoperative data with postoperative outcomes is a promising and low-cost way of evaluating long-term impacts of surgical interventions. In this study, we evaluated how preoperative cognitive status as measured by the clock drawing test contributed to predicting length of hospital stay, hospital charges, average pain experienced during follow-up, and 1-year mortality over and above intraoperative variables, demographics, preoperative physical status and comorbidities. We expanded our analysis to 6 specific surgical groups where sufficient data was available for cross-validation. The clock drawing images were represented by 10 constructional features discovered by a semi-supervised deep learning algorithm, previously validated to differentiate between dementia and non-dementia patients. Different machine learning models were trained to classify postoperative outcomes in hold-out test sets. The models were compared to their relative performance, time complexity, and interpretability. Shapley Additive Explanations (SHAP) analysis was used to find the most predictive features for classifying different outcomes in different surgical contexts. Relative classification performances achieved by different feature sets showed that the perioperative cognitive dataset which included clock drawing features in addition to intraoperative variables, demographics, and comorbidities served as the best dataset for 12 of 18 possible surgery-outcome combinations...

Mobile health (mHealth) apps have gained popularity over the past decade for patient health monitoring, yet their potential for timely intervention is underutilized due to limited integration with electronic health records (EHR) systems. Current EHR systems lack real-time monitoring capabilities for symptoms, medication adherence, physical and social functions, and community integration. Existing systems typically rely on static, in-clinic measures rather than dynamic, real-time patient data. This highlights the need for automated, scalable, and human-centered platforms to integrate patient-generated health data (PGHD) within EHR. Incorporating PGHD in a user-friendly format can enhance patient symptom surveillance, ultimately improving care management and post-surgical outcomes. To address this barrier, we have developed an mHealth platform, ROAMM-EHR, to capture real-time sensor data and Patient Reported Outcomes (PROs) using a smartwatch. The ROAMM-EHR platform can capture data from a consumer smartwatch, send captured data to a secure server, and display information within the Epic EHR system using a user-friendly interface, thus enabling healthcare providers to monitor post-surgical symptoms effectively., Comment: 8 pages, ACM MobiCom4AgeTech 2024

Delirium is an acute confusional state that has been shown to affect up to 31% of patients in the intensive care unit (ICU). Early detection of this condition could lead to more timely interventions and improved health outcomes. While artificial intelligence (AI) models have shown great potential for ICU delirium prediction using structured electronic health records (EHR), most of them have not explored the use of state-of-the-art AI models, have been limited to single hospitals, or have been developed and validated on small cohorts. The use of large language models (LLM), models with hundreds of millions to billions of parameters, with structured EHR data could potentially lead to improved predictive performance. In this study, we propose DeLLiriuM, a novel LLM-based delirium prediction model using EHR data available in the first 24 hours of ICU admission to predict the probability of a patient developing delirium during the rest of their ICU admission. We develop and validate DeLLiriuM on ICU admissions from 104,303 patients pertaining to 195 hospitals across three large databases: the eICU Collaborative Research Database, the Medical Information Mart for Intensive Care (MIMIC)-IV, and the University of Florida Health's Integrated Data Repository. The performance measured by the area under the receiver operating characteristic curve (AUROC) showed that DeLLiriuM outperformed all baselines in two external validation sets, with 0.77 (95% confidence interval 0.76-0.78) and 0.84 (95% confidence interval 0.83-0.85) across 77,543 patients spanning 194 hospitals. To the best of our knowledge, DeLLiriuM is the first LLM-based delirium prediction tool for the ICU based on structured EHR data, outperforming deep learning baselines which employ structured features and can provide helpful information to clinicians for timely interventions.

We present a comprehensive analysis of the periodic flares observed in the 6.7 GHz methanol transition in G22.356+0.066, utilizing the Maxwell-Bloch equations (MBEs) as a framework to model these phenomena. By solving the one-dimensional MBEs, we describe the behavior of both the quasi-steady-state maser and transient superradiance regimes. Our findings indicate that the observed periodic flares, with varying timescales across different velocities, are consistent with the characteristics of Dicke's superradiance, triggered by a common radiative pump in regions of varying inverted column densities. This work provides new insights into the physical processes governing variability in maser-hosting regions and underscores the significance of superradiance as a powerful radiation mechanism in astrophysical environments., Comment: 5 pages, 7 figures

We demonstrate that (001) grown Cd3As2 thin films with a superlattice-patterned gate can potentially realize the moir\'e Bernevig-Hughes-Zhang (BHZ) model. Our calculations identify the parameterization region necessary to achieve topological flat mini-bands with a C4z symmetric and a C6z symmetric potential. Additionally, we show that a spin-polarized state can serve as the minimal platform for hosting the moir\'e induced quantum anomalous Hall effect, supported by Hartree Fock interaction kernel analysis and self-consistent mean field calculations., Comment: 10 pages, 10 figures

Achieving precise control of photoinduced molecular charge transfer reactions underpins key emerging technologies. As such, the use of hybrid light-matter molecular exciton-polariton states has been proposed as a scheme to directly modify the efficiency and rate of such reactions. However, the efficacy of polariton-driven photochemistry remains an open question. Here, we demonstrate conditions under which photoinduced polaritonic charge transfer can be achieved and directly visualized using momentum resolved ultrafast spectroscopy. Key conditions for charge transfer are satisfied using Bloch surface wave polaritons, which exhibit favorable dispersion characteristics that permit the selective pumping of hybrid states with long lifetimes (100-400 fs) that permit vibrationally assisted molecular charge transfer. Using this approach, we tune the energetic driving force for charge separation, reducing it by as much as 0.5 eV compared to the bare exciton. These results establish that tunable and efficient polariton-driven molecular charge transfer is indeed possible using carefully considered photonic systems., Comment: 10 pages, 5 figures

Inspired by the entropy-area relation of black hole thermodynamics, we study the thermodynamics of cosmological apparent horizon in a spatially flat Friedmann-Robertson-Walker (FRW) universe in the framework of an Extended Uncertainty Principle (EUP). The adopted EUP naturally admits a minimal measurable momentum (equivalently a maximal measurable length), as an infrared cutoff in the theory. We derive the modified Friedmann equations in this setup and explore some predictions of these equations for the late time universe via distance measures. We show that in this framework it is possible to realize the late time cosmic speed-up and transition to the phantom phase of the equation of state parameter of the effective cosmic fluid without recourse to any dark energy component or modified gravity. Inspection of various distance measures in this framework shows that an EUP with a negative deformation parameter suffices for the interpretation of the late time asymptotically de Sitter universe with standard non-relativistic matter., Comment: 25 pages, 13 figures, Accepted for publication in The Astrophysical Journal

In this paper we construct a dissipative quintessential cosmic inflation. For this purpose, we add a multiplicative dissipative term in the standard quintessence field Lagrangian. We consider the specific form of dissipation as the time integral including the Hubble parameter and an arbitrary function that describes the dissipative properties of the quintessential scalar field. Inflation parameters and observables are calculated under slow-roll approximations and a detailed calculation of the cosmological perturbations is performed in this setup. We consider different forms of potentials and calculate the scalar spectral index and tensor-to-scalar ratio for a constant as well as variable dissipation function. To check the reliability of this model, a numerical analysis on the model parameters space is done in confrontation with recent observational data. By comparing the results with observational joint datasets at 68% and 95% confidence levels, we obtain some constraints on the model parameters space, specially the dissipation factor with e-folds numbers N = 55 and N = 60. As some specific results, we show that the power-law potential with a constant dissipation factor and N = 60 is mildly consistent with observational data in some restricted domains of the model parameter space with very small and negative dissipation factor and a negligible tensor-toscalar ratio. But this case with N = 55 is consistent with observation considerably. For power-law potential and variable dissipation factor as $Q = {\alpha}\phi^n$, the consistency with observation is also considerable with a reliable tensor-to-scalar ratio. The quadratic and quartic potentials with variable dissipation function as $Q = {\alpha}\phi^n$ are consistent with Planck2018 TT, TE, EE+lowE+lensing data at the 68% and 95% levels of confidence for some intervals of the parameter n., Comment: Accepted for publication in Physics of the Dark Universe, 37 pages, 14 figures, 8 tables

As more clinical workflows continue to be augmented by artificial intelligence (AI), AI literacy among physicians will become a critical requirement for ensuring safe and ethical AI-enabled patient care. Despite the evolving importance of AI in healthcare, the extent to which it has been adopted into traditional and often-overloaded medical curricula is currently unknown. In a scoping review of 1,699 articles published between January 2016 and June 2024, we identified 18 studies which propose guiding frameworks, and 11 studies documenting real-world instruction, centered around the integration of AI into medical education. We found that comprehensive guidelines will require greater clinical relevance and personalization to suit medical student interests and career trajectories. Current efforts highlight discrepancies in the teaching guidelines, emphasizing AI evaluation and ethics over technical topics such as data science and coding. Additionally, we identified several challenges associated with integrating AI training into the medical education program, including a lack of guidelines to define medical students AI literacy, a perceived lack of proven clinical value, and a scarcity of qualified instructors. With this knowledge, we propose an AI literacy framework to define competencies for medical students. To prioritize relevant and personalized AI education, we categorize literacy into four dimensions: Foundational, Practical, Experimental, and Ethical, with tailored learning objectives to the pre-clinical, clinical, and clinical research stages of medical education. This review provides a road map for developing practical and relevant education strategies for building an AI-competent healthcare workforce., Comment: 25 pages, 2 figures, 3 tables

The glycocalyx is a complex layer of carbohydrate and protein molecules that surrounds the cell membrane of many types of mammalian cells. It serves several important functions, including cell adhesion and communication, and maintain cell shape and stability, especially in the case of erythrocytes. Alteration of glycocalyx composition represents a cardiovascular health threatening. For example, in diabetes mellitus glycocalyx of erythrocytes and of endothelial cells is known to be impaired, a potential source of blood occlusion in microcirculation, which may lead to blindness, and renal failure of patients. The impact of glycocalyx impairment on erythrocyte aggregation remains a largely unexplored research area. We conduct here in vitro-experiments in microfluidic devices in order to investigate erythrocytes aggregation incubated with amylase, an enzyme that partially breaks down glycocalyx molecules. It is found that incubation of erythrocytes by amylase leads to a dramatic increase of their aggregation and stability and alters the aggregates morphologies. Confocal microscopy analysis reveals a significant degradation of the glycocalyx layer, correlated with enhanced erythrocytes aggregation. An increased erythrocyte aggregation in vivo should affect oxygen and other metabolites delivery to organs and tissues. This study brings new elements about elucidation of microscopic origins of erythrocyte aggregation and their potential impact on cardiovascular pathologies.

Distributed Deep Neural Network (DNN) training is a technique to reduce the training overhead by distributing the training tasks into multiple accelerators, according to a parallelization strategy. However, high-performance compute and interconnects are needed for maximum speed-up and linear scaling of the system. Wafer-scale systems are a promising technology that allows for tightly integrating high-end accelerators with high-speed wafer-scale interconnects, making it an attractive platform for distributed training. However, the wafer-scale interconnect should offer high performance and flexibility for various parallelization strategies to enable maximum optimizations for compute and memory usage. In this paper, we propose FRED, a wafer-scale interconnect that is tailored for the high-BW requirements of wafer-scale networks and can efficiently execute communication patterns of different parallelization strategies. Furthermore, FRED supports in-switch collective communication execution that reduces the network traffic by approximately 2X. Our results show that FRED can improve the average end-to-end training time of ResNet-152, Transformer-17B, GPT-3, and Transformer-1T by 1.76X, 1.87X, 1.34X, and 1.4X, respectively when compared to a baseline waferscale 2D-Mesh fabric.

By considering the Friedmann equations emerging from the entropy-area law of black hole thermodynamics in the context of the generalized uncertainty principle, we study tachyon inflation in the early universe. The presence of a minimal length modifies the Friedmann equations and hence the slow-roll and perturbation parameters in the tachyon model. These modifications, though small, affect the viability of the tachyon inflation in confrontation with observational data. We compare the numerical results of the model with Planck2018 TT, TE, EE +lowE+lensing+BAO+BK14(18) data and Planck2018 TT, TE,EE +lowE+lensing+BK14(18) +BAO+LIGO $\&$ Virgo2016 data at $68\%$ and $95\%$ CL. We show that while the tachyon inflation with power-law, inverse power-law and inverse exponential potentials is not observationally viable in comparison with the $1\sigma$ and $2\sigma$ confidence levels of the new joint data, in the presence of the minimal length the model becomes observationally viable., Comment: Accepted for publication in Physics of the Dark Universe,29 pages, 12 figures, 10 tables

We present significant improvements to our previous work on noise reduction in {\sl Herschel} observation maps by defining sparse filtering tools capable of handling, in a unified formalism, a significantly improved noise reduction as well as a deconvolution in order to reduce effects introduced by the limited instrumental response (beam). We implement greater flexibility by allowing a wider choice of parsimonious priors in the noise-reduction process. More precisely, we introduce a sparse filtering and deconvolution approach approach of type $l^2$-$l^p$, with $p > 0$ variable and apply it to a larger set of molecular clouds using {\sl Herschel} 250 $\mu $m data in order to demonstrate their wide range of application. In the {\sl Herschel} data, we are able to use this approach to highlight extremely fine filamentary structures and obtain singularity spectra that tend to show a significantly less $\log$-normal behavior and a filamentary nature in the less dense regions. We also use high-resolution adaptive magneto-hydrodynamic simulation data to assess the quality of deconvolution in such a simulated beaming framework., Comment: 24 pages, 32 figures, official acceptance date on 16/05/2024 (Astronomy & Astrophysics)

In today's forex market traders increasingly turn to algorithmic trading, leveraging computers to seek more profits. Deep learning techniques as cutting-edge advancements in machine learning, capable of identifying patterns in financial data. Traders utilize these patterns to execute more effective trades, adhering to algorithmic trading rules. Deep reinforcement learning methods (DRL), by directly executing trades based on identified patterns and assessing their profitability, offer advantages over traditional DL approaches. This research pioneers the application of a multi-agent (MA) RL framework with the state-of-the-art Asynchronous Advantage Actor-Critic (A3C) algorithm. The proposed method employs parallel learning across multiple asynchronous workers, each specialized in trading across multiple currency pairs to explore the potential for nuanced strategies tailored to different market conditions and currency pairs. Two different A3C with lock and without lock MA model was proposed and trained on single currency and multi-currency. The results indicate that both model outperform on Proximal Policy Optimization model. A3C with lock outperforms other in single currency training scenario and A3C without Lock outperforms other in multi-currency scenario. The findings demonstrate that this approach facilitates broader and faster exploration of different currency pairs, significantly enhancing trading returns. Additionally, the agent can learn a more profitable trading strategy in a shorter time.

Infrared emission features toward interstellar gas of the IC 348 star cluster in Perseus have been recently proposed to originate from the amino acid tryptophan. The assignment was based on laboratory infrared spectra of tryptophan pressed into pellets, a method which is known to cause large frequency shifts compared to the gas phase. We assess the validity of the assignment based on the original Spitzer data as well as new data from JWST. In addition, we report new spectra of tryptophan condensed in para-hydrogen matrices to compare with the observed spectra. The JWST MIRI data do not show evidence for tryptophan, despite deeper integration toward IC 348. In addition, we show that several of the lines attributed to tryptophan are likely due to instrumental artifacts. This, combined with the new laboratory data, allows us to conclude that there is no compelling evidence for the tryptophan assignment.

In recent years, there has been a significant increase in applications of multimodal signal processing and analysis, largely driven by the increased availability of multimodal datasets and the rapid progress in multimodal learning systems. Well-known examples include autonomous vehicles, audiovisual generative systems, vision-language systems, and so on. Such systems integrate multiple signal modalities: text, speech, images, video, LiDAR, etc., to perform various tasks. A key issue for understanding such systems is the relationship between various modalities and how it impacts task performance. In this paper, we employ the concept of mutual information (MI) to gain insight into this issue. Taking advantage of the recent progress in entropy modeling and estimation, we develop a system called InfoMeter to estimate MI between modalities in a multimodal learning system. We then apply InfoMeter to analyze a multimodal 3D object detection system over a large-scale dataset for autonomous driving. Our experiments on this system suggest that a lower MI between modalities is beneficial for detection accuracy. This new insight may facilitate improvements in the development of future multimodal learning systems., Comment: 6 pages, 7 figures, IEEE MIPR 2024

Given the sheer volume of surgical procedures and the significant rate of postoperative fatalities, assessing and managing surgical complications has become a critical public health concern. Existing artificial intelligence (AI) tools for risk surveillance and diagnosis often lack adequate interpretability, fairness, and reproducibility. To address this, we proposed an Explainable AI (XAI) framework designed to answer five critical questions: why, why not, how, what if, and what else, with the goal of enhancing the explainability and transparency of AI models. We incorporated various techniques such as Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), counterfactual explanations, model cards, an interactive feature manipulation interface, and the identification of similar patients to address these questions. We showcased an XAI interface prototype that adheres to this framework for predicting major postoperative complications. This initial implementation has provided valuable insights into the vast explanatory potential of our XAI framework and represents an initial step towards its clinical adoption., Comment: 32 pages, 7 figures, 4 supplement figures and 1 supplement table

Manipulating external stimuli is crucial for enhancing and controlling the optical response in chiral structures. This study introduces chiral structures composed of single, dimer, or trimer arrays of plasmonic and magneto-plasmonic trapezoidal and triangular nanoantennas. Through modeling, we demonstrate how these systems can (i) function as chiral metasurfaces, (ii) modulate chiral light absorption via an external magnetic field, (iii) magnetically enhance the chiro-optical response, and (iv) improve the chiro-optical and magneto-chiral response through structural engineering. By optimizing the nanoantenna geometry, we achieve enhancements larger than 100 in the chiro-optical signal of a magnetized Ni monomer compared to a non-magnetized system. Additionally, we model bimetallic systems, showing a ~2-fold increase in chiro-optical response for hybrid Au and Ni magneto-plasmonic systems compared to purely plasmonic Au systems, while simultaneously enabling magnetic tuning of the optical response. Our results highlight the potential of structurally engineered magneto-plasmonic nanoantennas for advanced applications in chiro-optical sensing and tunable photonic devices.

Modern electronic health records (EHRs) hold immense promise in tracking personalized patient health trajectories through sequential deep learning, owing to their extensive breadth, scale, and temporal granularity. Nonetheless, how to effectively leverage multiple modalities from EHRs poses significant challenges, given its complex characteristics such as high dimensionality, multimodality, sparsity, varied recording frequencies, and temporal irregularities. To this end, this paper introduces a novel multimodal contrastive learning framework, specifically focusing on medical time series and clinical notes. To tackle the challenge of sparsity and irregular time intervals in medical time series, the framework integrates temporal cross-attention transformers with a dynamic embedding and tokenization scheme for learning multimodal feature representations. To harness the interconnected relationships between medical time series and clinical notes, the framework equips a global contrastive loss, aligning a patient's multimodal feature representations with the corresponding discharge summaries. Since discharge summaries uniquely pertain to individual patients and represent a holistic view of the patient's hospital stay, machine learning models are led to learn discriminative multimodal features via global contrasting. Extensive experiments with a real-world EHR dataset demonstrated that our framework outperformed state-of-the-art approaches on the exemplar task of predicting the occurrence of nine postoperative complications for more than 120,000 major inpatient surgeries using multimodal data from UF health system split among three hospitals (UF Health Gainesville, UF Health Jacksonville, and UF Health Jacksonville-North)., Comment: 12 pages, 3 figures. arXiv admin note: text overlap with arXiv:2403.04012

Background: The accurate prediction of postoperative complication risk using Electronic Health Records (EHR) and artificial intelligence shows great potential. Training a robust artificial intelligence model typically requires large-scale and diverse datasets. In reality, collecting medical data often encounters challenges surrounding privacy protection. Methods: This retrospective cohort study includes adult patients who were admitted to UFH Gainesville (GNV) (n = 79,850) and Jacksonville (JAX) (n = 28,636) for any type of inpatient surgical procedure. Using perioperative and intraoperative features, we developed federated learning models to predict nine major postoperative complications (i.e., prolonged intensive care unit stay and mechanical ventilation). We compared federated learning models with local learning models trained on a single site and central learning models trained on pooled dataset from two centers. Results: Our federated learning models achieved the area under the receiver operating characteristics curve (AUROC) values ranged from 0.81 for wound complications to 0.92 for prolonged ICU stay at UFH GNV center. At UFH JAX center, these values ranged from 0.73-0.74 for wound complications to 0.92-0.93 for hospital mortality. Federated learning models achieved comparable AUROC performance to central learning models, except for prolonged ICU stay, where the performance of federated learning models was slightly higher than central learning models at UFH GNV center, but slightly lower at UFH JAX center. In addition, our federated learning model obtained comparable performance to the best local learning model at each center, demonstrating strong generalizability. Conclusion: Federated learning is shown to be a useful tool to train robust and generalizable models from large scale data across multiple institutions where data protection barriers are high., Comment: 57 pages. 2 figures, 3 tables, 2 supplemental figures, 8 supplemental tables