搜索结果：BMC geriatrics

Amortized Bayesian model comparison (BMC) enables fast probabilistic ranking of models via simulation-based training of neural surrogates. However, the accuracy of neural surrogates deteriorates when simulation models are misspecified; the very case where model comparison is most needed. We evaluate four different amortized BMC methods. We supplement traditional simulation-based training of these methods with a \emph{self-consistency} (SC) loss on unlabeled real data to improve BMC estimates under distribution shifts. Using one artificial and two real-world case studies, we compare amortized BMC estimators with and without SC against analytic or bridge sampling benchmarks. In the \emph{closed-world} case (data is generated by one of the candidate models), BMC estimators using classifiers work acceptably well even without SC training. However, these methods also benefit the least from SC training. In the \emph{open-world} scenario (all models misspecified), SC training strongly improves BMC estimators when having access to analytic likelihoods, or when surrogate likelihoods are locally accurate near the true parameter posterior, even for severely misspecified models. We conclude wit

Bounded Model Checking (BMC) is a widely used software verification technique. Despite its successes, the technique has several limiting factors, from state-space explosion to lack of completeness. Over the years, interval analysis has repeatedly been proposed as a partial solution to these limitations. In this work, we evaluate whether the computational cost of interval analysis yields significant enough improvements in BMC's performance to justify its use. In more detail, we quantify the benefits of interval analysis on two benchmarks: the Intel Core Power Management firmware and 9537 programs in the ReachSafety category of the International Competition on Software Verification. Our results show that interval analysis is essential in solving 203 unique benchmarks.

Finding software vulnerabilities in concurrent programs is a challenging task due to the size of the state-space exploration, as the number of interleavings grows exponentially with the number of program threads and statements. We propose and evaluate EBF (Ensembles of Bounded Model Checking with Fuzzing) -- a technique that combines Bounded Model Checking (BMC) and Gray-Box Fuzzing (GBF) to find software vulnerabilities in concurrent programs. Since there are no publicly-available GBF tools for concurrent code, we first propose OpenGBF -- a new open-source concurrency-aware gray-box fuzzer that explores different thread schedules by instrumenting the code under test with random delays. Then, we build an ensemble of a BMC tool and OpenGBF in the following way. On the one hand, when the BMC tool in the ensemble returns a counterexample, we use it as a seed for OpenGBF, thus increasing the likelihood of executing paths guarded by complex mathematical expressions. On the other hand, we aggregate the outcomes of the BMC and GBF tools in the ensemble using a decision matrix, thus improving the accuracy of EBF. We evaluate EBF against state-of-the-art pure BMC tools and show that it can

Data scarcity challenges the development and implementation of innovative healthcare solutions. In geriatrics, fall-related injuries are a major cause of hospitalization, functional decline, and mortality in older adults. Optimizing post-operative discharge planning can mitigate these outcomes, but limited data hinders predictive model development. Here, we explored generative machine learning approaches to augment data from the SURGE-Ahead project (Supporting SURgery with Geriatric Co-Management and AI), an initiative addressing geriatric perioperative care. Data from the German geriatric trauma register (AltersTraumaZentrum; ATZ) were incorporated using two strategies: (i) combining SURGE-Ahead and ATZ register data with imputation (ComImp) and (ii) generating synthetic data from SURGE-Ahead alone or combined SURGE-Ahead and the ATZ register datasets with Adversarial random forests (ARF). Predictive models, including multinomial logistic regression, random forest, and a prior-fitted transformer (TabPFN), were trained and evaluated using standard performance metrics: accuracy, area under the receiver operating characteristic curve (ROC AUC), Brier score, and the logistic loss. Ran

With the skyrocketing costs of GPUs and their virtual instances in the cloud, there is a significant desire to use CPUs for large language model (LLM) inference. KV cache update, often implemented as allocation, copying, and in-place strided update for each generated token, incurs significant overhead. As the sequence length increases, the allocation and copy overheads dominate the performance. Alternate approaches may allocate large KV tensors upfront to enable in-place updates, but these matrices (with zero-padded rows) cause redundant computations. In this work, we propose a new KV cache allocation mechanism called Balancing Memory and Compute (BMC). BMC allocates, once every r iterations, KV tensors with r redundant rows, allowing in-place update without copy overhead for those iterations, but at the expense of a small amount of redundant computation. Second, we make an interesting observation that the extra rows allocated in the KV tensors and the resulting redundant computation can be repurposed for Speculative Decoding (SD) that improves token generation efficiency. Last, BMC represents a spectrum of design points with different values of r. To identify the best-performing d

Bounded model checking (BMC) and fuzzing techniques are among the most effective methods for detecting errors and security vulnerabilities in software. However, there are still shortcomings in detecting these errors due to the inability of existent methods to cover large areas in target code. We propose FuSeBMC v4, a test generator that synthesizes seeds with useful properties, that we refer to as smart seeds, to improve the performance of its hybrid fuzzer thereby achieving high C program coverage. FuSeBMC works by first analyzing and incrementally injecting goal labels into the given C program to guide BMC and Evolutionary Fuzzing engines. After that, the engines are employed for an initial period to produce the so-called smart seeds. Finally, the engines are run again, with these smart seeds as starting seeds, in an attempt to achieve maximum code coverage / find bugs. During both seed generation and normal running, coordination between the engines is aided by the Tracer subsystem. This subsystem carries out additional coverage analysis and updates a shared memory with information on goals covered so far. Furthermore, the Tracer evaluates test cases dynamically to convert cases

The rapid increase in the world's aging population to 16% by the year 2050 spurs the need for the application of digital health solutions to enhance older individuals' independence, accessibility, and well-being. While digital health technologies such as telemedicine, wearables, and mobile health applications can transform geriatric care, their adoption among older individuals is not evenly distributed. This study redefines the "digital divide" among older health care as a usability divide, contends that user experience (UX) poor design is the primary adoption barrier, rather than access. Drawing on interdisciplinary studies and design paradigms, the research identifies the main challenges: visual, cognitive, and motor impairment; complicated interfaces; and lack of co-creation with older adults, and outlines how participatory, user-focused, and inclusive notions of design can transcend them. Findings reveal that older persons easily embrace those technologies that are intuitive, accessible, and socially embedded as they promote autonomy, confidence, and equity in health. The study identifies the effects of the design attributes of high-contrast screens, lower interaction flow, mul

Satisfiability Modulo Theories (SMT) solvers have been successfully applied to solve many problems in formal verification such as bounded model checking (BMC) for many classes of systems from integrated circuits to cyber-physical systems. Typically, BMC is performed by checking satisfiability of a possibly long, but quantifier-free formula. However, BMC problems can naturally be encoded as quantified formulas over the number of BMC steps. In this approach, we then use decision procedures supporting quantifiers to check satisfiability of these quantified formulas. This approach has previously been applied to perform BMC using a Quantified Boolean Formula (QBF) encoding for purely discrete systems, and then discharges the QBF checks using QBF solvers. In this paper, we present a new quantified encoding of BMC for rectangular hybrid automata (RHA), which requires using more general logics due to the real (dense) time and real-valued state variables modeling continuous states. We have implemented a preliminary experimental prototype of the method using the HyST model transformation tool to generate the quantified BMC (QBMC) queries for the Z3 SMT solver. We describe experimental result

To support aging-in-place, adult children often provide care to their aging parents from a distance. These informal caregivers desire plug-and-play remote care solutions for privacy-preserving continuous monitoring that enabling real-time activity monitoring and intuitive, actionable information. This short paper presents insights from three iterations of deployment experience for remote monitoring system and the iterative improvement in hardware, modeling, and user interface guided by the Geriatric 4Ms framework (matters most, mentation, mobility, and medication). An LLM-assisted solution is developed to balance user experience (privacy-preserving, plug-and-play) and system performance.

The training process of generative adversarial networks (GANs) is unstable and does not converge globally. In this paper, we examine the stability of GANs from the perspective of control theory and propose a universal higher-order noise-based controller called Brownian Motion Controller (BMC). Starting with the prototypical case of Dirac-GANs, we design a BMC to retrieve precisely the same but reachable optimal equilibrium. We theoretically prove that the training process of DiracGANs-BMC is globally exponential stable and derive bounds on the rate of convergence. Then we extend our BMC to normal GANs and provide implementation instructions on GANs-BMC. Our experiments show that our GANs-BMC effectively stabilizes GANs' training under StyleGANv2-ada frameworks with a faster rate of convergence, a smaller range of oscillation, and better performance in terms of FID score.

In the era of technological revolution in healthcare, the integration of Artificial Intelligence (AI) stands out as a potentially transformative force (1). Clinicians often view the promises of AI with a healthy degree of skepticism (2), understanding the complexities of aging and the heterogeneity of physical and cognitive functions in older adults (3). This skepticism is necessary and healthy. AI engineers, on the other hand, sometimes struggle to fully grasp the expectations of clinicians and the intricate nature of aging (4). The future will require multidisciplinary collaboration among clinicians, AI engineers, and key stakeholders, including patient advocacy groups and older adults, to develop solutions that effectively address the complex nature of aging. As a geriatrician (P.A.) and an AI researcher (R.C.), we aim to provide insight into this burgeoning field, highlighting how AI can redefine, revitalize, and revolutionize care for older adults. In this editorial, we are also charting a course for an upcoming special issue in The Journals of Gerontology on AI-enabled wearables and sensors for older adults. This upcoming issue will aim to explore innovative ways these technologies enhance care for older adults, with a focus on practical applications and real-world impact. Geriatrics, with its focus on the health and care of older adults, presents unique challenges, including managing multiple chronic diseases, cognitive decline, and end-of-life care (5). AI’s role in geriatrics is not limited to clinical applications. It extends to enhancing the everyday lives of older adults through AI-powered assistive technologies, enabling better management of chronic conditions, and providing companionship through advanced robotics and virtual assistants (1). These innovations are crucial in addressing the psychological and social aspects of aging, thus contributing to a holistic approach to care for older adults. The future of geriatrics care is inherently collaborative, demanding a synergy among healthcare professionals, technologists, ethicists, and AI developers. The goal is to foster an environment where AI not only coexists with traditional healthcare practices but also enhances them, always keeping the well-being and dignity of older adults at the forefront. In that realm, the National Institute on Aging has allocated $60 million to establish 3 Artificial Intelligence and Technology Collaboratories for Aging Research at Johns Hopkins University, University of Pennsylvania, and a University of Massachusetts-led consortium. This initiative aims to accelerate the development of AI technologies tailored for older adults. The project brings together experts across various fields, including medicine, engineering, and business, alongside patient advocacy groups and older adults. These collaboratories seek to integrate AI into geriatric care effectively, enhancing healthcare practices while focusing on the well-being and dignity of older adults (6,7). AI has already made significant strides in the healthcare of older adults, particularly in predictive analytics and diagnostics. A notable example is Stanford Hospital’s AI algorithm (8), which assesses mortality risk. This deep-learning algorithm, trained on 2 million patient records, flags patients with a 75% or higher risk of dying within 3–12 months (9,10). This information prompts physicians to initiate advance care planning conversations, though the specific mortality risk is not disclosed to the patient or attending physician. This algorithm is intended as a tool for facilitating critical end-of-life discussions, often neglected due to busy clinical schedules and the discomfort of the topic (8,11). The utility of this algorithm extends beyond physicians to other healthcare staff. Therapists, for example, use it to guide conversations about end-of-life care, focusing on what is important to the patient in their final months. This broader involvement of nonphysician staff in such discussions represents a significant shift in the approach to terminal care (8,11). Likewise, the University of Pennsylvania Health System’s AI algorithm, which evaluates over 500 patient variables, has quadrupled the number of advance care conversations (8). This algorithm has also helped address potential biases in patient care, particularly in conversations about death among different ethnic or minority groups (12). AI’s impact on healthcare is not limited to predictive analytics. Google’s AI system for lung cancer detection, for instance, has shown promising results, outperforming radiologists in some aspects (13). The FDA has approved over 160 AI-driven products in recent years, mostly in medical imaging, reflecting AI’s growing role in this field (8). A recent project by computer scientists at the University of California, San Diego, CA, exemplifies this evolution. Funded by an Amazon Research Award, they are developing a voice assistant tailored to understand and respond to medical questions from adults over age 65 (8,14). This demographic often struggles with current voice assistants due to their conversational speech patterns. This initiative underscores the necessity to adapt AI to the user, emphasizing inclusivity in technology design. In terms of patient outcomes, AI is also helping with surgical decision making. A team from Johns Hopkins University (8) has developed an AI algorithm that uses imaging data to predict the best candidates for spinal surgery. This model, which considers spinal morphology, is potentially more accurate than traditional assessment methods based solely on general health indicators (15). The coronavirus disease 2019 (COVID-19) pandemic has further highlighted the importance of AI in healthcare. BlueDot’s AI methodologies were instrumental in tracking early cases of the outbreak. The Severe COVID-19 Adaptive Risk Predictor developed by Johns Hopkins researchers provided valuable insights for patient management during the pandemic, although challenges in data consistency and validation were noted (8). AI has also been pivotal in sepsis detection and management. Suchi Saria’s work at Johns Hopkins led to the development of an AI system that integrates patient data for early sepsis detection, potentially saving lives and reducing healthcare costs. Her startup, Bayesian Health, has made this technology accessible to thousands of healthcare providers (16). One of the emerging applications of AI in healthcare is its utility for clinicians, including geriatricians, in their day-to-day patient interactions. AI systems are increasingly being developed to review patient messages and analyze them in the context of past encounters and medical records. This capability allows for the preparation of a thorough analysis and assists doctors in crafting responses, enhancing the efficiency and effectiveness of patient communication. Additionally, AI is being employed to listen to patient encounters and prepare detailed encounter notes. This not only saves time for healthcare professionals but also ensures accuracy in medical documentation, which is critical for patient care continuity. AI’s potential in diagnosing rare syndromes and diseases is another promising application. By analyzing vast datasets and identifying patterns that might be overlooked by human clinicians, AI can aid in the early detection and diagnosis of rare conditions, significantly affecting patient outcomes (8). However, AI in healthcare faces challenges, including algorithmic biases and data quality issues. Studies have shown that AI tools can unintentionally perpetuate racial and gender biases present in healthcare data (17,18). Addressing these biases is crucial for the equitable application of AI in healthcare. The integration of AI for older adults is also not without challenges. Existing systems often reflect societal biases, particularly ageism, neglecting the needs and preferences of older adults. This issue is evident in the way older adults interact with technologies like voice assistants. In addition, and specifically in the older adult’s domain, there is a narrow focus on accessibility, often conflating aging with disability, which inadvertently excludes older adults from the wider benefits of technology (19). This perspective sidelines the rich life experiences, wisdom, and potential contributions of older adults in the realm of digital technology. There is a need for technologies that are not only usable by older adults with age-related limitations but are also useful and acceptable to the broader spectrum of older adults, many of whom do not have such limitations. There is a critical need to involve older adults as full stakeholders in digital society, and enable them to be active participants and co-designers in technology development. This inclusion can lead to innovations that better align with the varied needs and preferences of older adults, and help dispel stereotypes that often lead to the marginalization of this demographic in the tech industry. By embracing the diverse experiences and insights of older adults, we can develop AI technologies that truly enhance the lives of older adults, making the digital world more inclusive and beneficial for all age groups. Despite these challenges, AI’s role in healthcare is rapidly expanding. Big tech firms and startups are investing heavily in AI-driven healthcare solutions, anticipating significant growth in this sector. The future of AI in healthcare, particularly geriatrics care, lies in addressing these challenges and ensuring equitable access to AI tools across all patient demographics (1). In addressing the future of geriatrics and healthcare, a common concern arises: will AI replace physicians, and scientists? The reality, however, reveals a different narrative. The truth is, AI will not replace clinicians, including geriatricians, or scientists. These roles, steeped in years of training, experience, and human understanding, remain irreplaceable. What AI does is enhance these roles, offering tools and insights that were previously unimaginable. It brings a level of precision, efficiency, and predictive power to healthcare, changing how we approach diagnosis, treatment, and patient care. This power of AI does not diminish the value of healthcare professionals; rather, it elevates it. Those who embrace AI and integrate it into their practice will find themselves at the forefront of modern medicine, especially in the field of geriatrics where the complexities of care are manifold. Thus, as we navigate the inevitable tide of integrating AI into geriatrics and healthcare at large, it’s not a question of replacement but of enhancement and evolution. The key lies in harnessing AI as a powerful ally in our quest to provide the best possible care for older adults. Those who adapt, learn, and grow with these technological advancements will be the ones shaping the future of geriatrics care. They are the ones who will not be replaced, but will instead become irreplaceable. This work was supported by grants from the National Institute on Aging, part of the National Institutes of Health (P30AG073104 to Johns Hopkins University). None.

Bayesian model comparison (BMC) offers a principled approach for assessing the relative merits of competing computational models and propagating uncertainty into model selection decisions. However, BMC is often intractable for the popular class of hierarchical models due to their high-dimensional nested parameter structure. To address this intractability, we propose a deep learning method for performing BMC on any set of hierarchical models which can be instantiated as probabilistic programs. Since our method enables amortized inference, it allows efficient re-estimation of posterior model probabilities and fast performance validation prior to any real-data application. In a series of extensive validation studies, we benchmark the performance of our method against the state-of-the-art bridge sampling method and demonstrate excellent amortized inference across all BMC settings. We then showcase our method by comparing four hierarchical evidence accumulation models that have previously been deemed intractable for BMC due to partly implicit likelihoods. Additionally, we demonstrate how transfer learning can be leveraged to enhance training efficiency. We provide reproducible code for

Bounded model checking (BMC) is an effective technique for hunting bugs by incrementally exploring the state space of a system. To reason about infinite traces through a finite structure and to ultimately obtain completeness, BMC incorporates loop conditions that revisit previously observed states. This paper focuses on developing loop conditions for BMC of HyperLTL- a temporal logic for hyperproperties that allows expressing important policies for security and consistency in concurrent systems, etc. Loop conditions for HyperLTL are more complicated than for LTL, as different traces may loop inconsistently in unrelated moments. Existing BMC approaches for HyperLTL only considered linear unrollings without any looping capability, which precludes both finding small infinite traces and obtaining a complete technique. We investigate loop conditions for HyperLTL BMC, where the HyperLTL formula can contain up to one quantifier alternation. We first present a general complete automata-based technique which is based on bounds of maximum unrollings. Then, we introduce alternative simulation-based algorithms that allow exploiting short loops effectively, generating SAT queries whose satisfia

Traumatic Brain Injury (TBI) is a major contributor to mortality among older adults, with geriatric patients facing disproportionately high risk due to age-related physiological vulnerability and comorbidities. Early and accurate prediction of mortality is essential for guiding clinical decision-making and optimizing ICU resource allocation. In this study, we utilized the MIMIC-III database to identify geriatric TBI patients and applied a machine learning framework to develop a 30-day mortality prediction model. A rigorous preprocessing pipeline-including Random Forest-based imputation, feature engineering, and hybrid selection-was implemented to refine predictors from 69 to 9 clinically meaningful variables. CatBoost emerged as the top-performing model, achieving an AUROC of 0.867 (95% CI: 0.809-0.922), surpassing traditional scoring systems. SHAP analysis confirmed the importance of GCS score, oxygen saturation, and prothrombin time as dominant predictors. These findings highlight the value of interpretable machine learning tools for early mortality risk stratification in elderly TBI patients and provide a foundation for future clinical integration to support high-stakes decision

Monitoring respiration parameters such as respiratory rate could be beneficial to understand the impact of training on equine health and performance and ultimately improve equine welfare. In this work, we compare deep learning-based methods to an adapted signal processing method to automatically detect cyclic respiratory events and extract the dynamic respiratory rate from microphone recordings during high intensity exercise in Standardbred trotters. Our deep learning models are able to detect exhalation sounds (median F1 score of 0.94) in noisy microphone signals and show promising results on unlabelled signals at lower exercising intensity, where the exhalation sounds are less recognisable. Temporal convolutional networks were better at detecting exhalation events and estimating dynamic respiratory rates (median F1: 0.94, Mean Absolute Error (MAE) $\pm$ Confidence Intervals (CI): 1.44$\pm$1.04 bpm, Limits Of Agreements (LOA): 0.63$\pm$7.06 bpm) than long short-term memory networks (median F1: 0.90, MAE$\pm$CI: 3.11$\pm$1.58 bpm) and signal processing methods (MAE$\pm$CI: 2.36$\pm$1.11 bpm). This work is the first to automatically detect equine respiratory sounds and automatically

In order to address the need for more capacity and coverage in the 5th generation (5G) of wireless networks, ultra-dense wireless networks are introduced which mainly consist of indoor small cells. This new architecture has paved the way for the advent of a new concept called Micro Operator. A micro operator is an entity that provides connections and local 5G services to the customers and relies on local frequency resources. We discuss business models of micro operators in a 5G coopetitive environment and develop a framework to indicate the business model canvas (BMC) of this new concept. Providing BMC for new businesses is a strategic approach to offer value to customers. In this research study, BMC and its elements are introduced and explained for 5G micro operators.

Resolving the exploration-exploitation trade-off remains a fundamental problem in the design and implementation of reinforcement learning (RL) algorithms. In this paper, we focus on model-free RL using the epsilon-greedy exploration policy, which despite its simplicity, remains one of the most frequently used forms of exploration. However, a key limitation of this policy is the specification of $\varepsilon$. In this paper, we provide a novel Bayesian perspective of $\varepsilon$ as a measure of the uniformity of the Q-value function. We introduce a closed-form Bayesian model update based on Bayesian model combination (BMC), based on this new perspective, which allows us to adapt $\varepsilon$ using experiences from the environment in constant time with monotone convergence guarantees. We demonstrate that our proposed algorithm, $\varepsilon$-\texttt{BMC}, efficiently balances exploration and exploitation on different problems, performing comparably or outperforming the best tuned fixed annealing schedules and an alternative data-dependent $\varepsilon$ adaptation scheme proposed in the literature.

This paper investigates approaches to parallelizing Bounded Model Checking (BMC) for shared memory environments as well as for clusters of workstations. We present a generic framework for parallelized BMC named Tarmo. Our framework can be used with any incremental SAT encoding for BMC but for the results in this paper we use only the current state-of-the-art encoding for full PLTL. Using this encoding allows us to check both safety and liveness properties, contrary to an earlier work on distributing BMC that is limited to safety properties only. Despite our focus on BMC after it has been translated to SAT, existing distributed SAT solvers are not well suited for our application. This is because solving a BMC problem is not solving a set of independent SAT instances but rather involves solving multiple related SAT instances, encoded incrementally, where the satisfiability of each instance corresponds to the existence of a counterexample of a specific length. Our framework includes a generic architecture for a shared clause database that allows easy clause sharing between SAT solver threads solving various such instances. We present extensive experimental results obtained with multip

Bayesian model comparison (BMC) offers a principled probabilistic approach to study and rank competing models. In standard BMC, we construct a discrete probability distribution over the set of possible models, conditional on the observed data of interest. These posterior model probabilities (PMPs) are measures of uncertainty, but -- when derived from a finite number of observations -- are also uncertain themselves. In this paper, we conceptualize distinct levels of uncertainty which arise in BMC. We explore a fully probabilistic framework for quantifying meta-uncertainty, resulting in an applied method to enhance any BMC workflow. Drawing on both Bayesian and frequentist techniques, we represent the uncertainty over the uncertain PMPs via meta-models which combine simulated and observed data into a predictive distribution for PMPs on new data. We demonstrate the utility of the proposed method in the context of conjugate Bayesian regression, likelihood-based inference with Markov chain Monte Carlo, and simulation-based inference with neural networks.