Probabilistic Approach Research Articles

Development and validation of a biomechanically fidelic surgical training knee model.

Knee arthroplasty technique is constantly evolving and the opportunity for surgeons to practice new techniques is currently highly dependent on the availability of cadaveric specimens requiring certified facilities. The high cost, limited supply, and heterogeneity of cadaveric specimens has increased the demand for synthetic training models, which are currently limited by a lack of biomechanical fidelity. Here, we aimed to design, manufacture, and experimentally validate a synthetic knee surgical training model which reproduces the flexion dependent varus-valgus (VV) and anterior-posterior (AP) mechanics of cadaveric knees, while maintaining anatomic accuracy. A probabilistic finite element modeling approach was employed to design physical models to exhibit passive cadaveric VV and AP mechanics. Seven synthetic models were manufactured and tested in a six-degree-of-freedom hexapod robot. Overall, the synthetic models exhibited cadaver-like VV and AP mechanics across a wide range of flexion angles with little variation between models. In the extended position, two models showed increased valgus rotation (<0.5°), and three models showed increased posterior tibial translation (<1.7 mm) when compared to the 95% confidence interval (CI) of cadaveric measurements. At full flexion, all models showed VV and AP mechanics within the 95% CI of cadaveric measurements. Given the repeatable mechanics exhibited, the knee models developed in this study can be used to reduce the current reliance on cadaveric specimens in surgical training.

Three-Body Binary Formation in Clusters: Analytical Theory

Abstract Binary formation in clusters through triple encounters between three unbound stars, ‘three-body’ binary formation, is one of the main dynamical formation processes of binary systems in dense environments. In this paper, we use an analytical probabilistic approach to study the process for the equal mass case and calculate a probability distribution for the orbital parameters of three-body-formed binaries, as well as their formation rate. For the first time, we give closed-form analytical expressions to the full orbital parameter distribution, accounting for both energy and angular momentum conservation. This calculation relies on the sensitive dependence of the outcomes of three-body scatterings on the initial conditions: here we compute the rate of three-body binaries from ergodic interactions, which allow for an analytical derivation of the distribution of orbital parameters of the binaries thus created. We find that soft binaries are highly favoured in this process and that these binaries have a super-thermal eccentricity distribution, while the few hard three-body binaries have an eccentricity distribution much closer to thermal. The analytical results predict and reproduce simulation results of three-body scattering experiments in the literature well.

Innovative decision-making for solar photovoltaic system installation in residential energy hubs: A probabilistic dynamic planning approach

This paper introduces an innovative probabilistic dynamic planning model tailored to optimize photovoltaic (PV) system installation within Residential Energy Hubs (REHs). The REH amalgamates diverse components, including PV systems, Combined Heat and Power (CHP) units, and gas boilers, to holistically fulfill residential thermal and electrical energy demands. The model's primary objective is to achieve an optimal equilibrium between minimizing initial investment costs associated with PV systems and managing ongoing energy expenses within the REH framework, encompassing potential PV energy losses over the planning horizon. To augment the flexibility and benefits of PV systems for residential consumers, the model integrates several pioneering features. Particularly notable is the introduction of a heating energy storage system, which complements the REH's thermal requirements and adeptly reacts to fluctuations in gas and electricity prices, thereby maximizing PV system benefits and ensuring homeowners' cost efficiency. Moreover, the model addresses the inherent uncertainty tied to PV generation by employing a two-point estimation method, offering a more realistic depiction of system performance and potential outcomes. To resolve the two-objective problem, an initial conversion into a single-objective problem via a weighted summation method precedes the application of an optimization algorithm inspired by symbiotic organisms. The paper demonstrates the practical application and efficacy of the proposed model through extensive simulations spanning a 10-year planning horizon. The results underscore the model's effectiveness and versatility in guiding residential consumers to make informed and economically viable decisions regarding PV system installation within the broader spectrum of their energy requirements. This research represents a significant stride towards empowering homeowners with invaluable tools for embracing sustainable energy solutions.

Application of STBC in MIMO-OFDM Broadband Wireless Communications

The Space Time Block Coding (STBC) is applied in multiple transmit antennas using data transfer for fading channels. After being encoded, the data is broken into n streams of continuously broadcast strings over n transmit antennas using STBC. The received signal is the combination of n broadcast signals and it becomes erroneous due to the entry of noise at receiver end. The probabilistic decoding approach is applied instead of joint detection approach for recovering data via decoupling of signals delivered from various antennas. The maximum likelihood decoding scheme employs the orthogonal structure of STBC called as OSTBC (Orthogonal Space Time Block Code) for providing the maximum likelihood decoding algorithm depending upon linear computation at receiver. In the present work, a MATLAB/SIMULINK environment of OSTBC has been used in order to acquire the highest dimension of variability for specific number of broadcasts as well as receive antennas utilizing a simple decoding techniques. With or without Grey Coding, OSTBC / STBC is used in MATLAB / SIMULINK blocs. The OSTBC algorithm calculates the highest achievable transmission rate for every amount of transmit antennas in any constellations including M-PSK array. M-PSK STBCs are used for different complex constellations to achieve ½ as well as ¾ of highest allowable transmission rate for Multiple Input Multiple Output (MIMO) transmit antennas utilizing various constellations

Probabilistic wind power forecasting resilient to missing values: An adaptive quantile regression approach

Probabilistic wind power forecasting approaches have significantly advanced in recent decades. However, forecasters often assume data completeness and overlook the challenge of missing values resulting from sensor failures, network congestion, etc. Traditionally, this issue is addressed during the data preprocessing procedure using methods such as deletion and imputation. Nevertheless, these ad-hoc methods pose challenges to probabilistic wind power forecasting at both parameter estimation and operational forecasting stages. In this paper, we propose a resilient probabilistic forecasting approach that smoothly adapts to missingness patterns without requiring preprocessing or retraining. Specifically, we design an adaptive quantile regression model with parameters capable of adapting to missing patterns, comprising two modules. The first is a feature extraction module where weights are kept static and biases are designed as a function of missingness patterns. The second is a non-crossing quantile neural network module, ensuring monotonicity of quantiles, with higher quantiles derived by adding non-negative amounts to lower quantiles. The proposed approach is applicable to cases under all missingness mechanisms including missing-not-at-random cases. Case studies demonstrate that our proposed approach achieves state-of-the-art results in terms of the continuous ranked probability score, with acceptable computational cost.

Reliability analysis of steel truss bars with imprecise statistical data

The article describes the probabilistic approach of reliability analysis for steel truss bars with imprecise statistical data. An approach to recovery the probability density of a random variable by the kernel density estimation method is illustrated for a small number of tests/measurements cases. The estimation of the distribution density parameters is presented in an interval form, which makes it possible to form a p-box for the random variable. Evaluation of the probability of non-failure allow to get a quantitative assessment for the safety of steel trusses, to identify the most unreliable trusses for monitoring their technical condition or to calculate the effect of strengthening the truss according to the reliability criterion. The probability of non-failure is estimated in the interval form. If the lower limit of the interval does not allow to get a conclusion about the truss safety level, then it can be increased by obtaining additional statistical data or by strengthening the truss or reducing the permissible load.

An enhanced method for evaluating the effectiveness of protective devices for road safety application

This paper presents an enhanced probabilistic approach to estimate the real-world safety performance of new device concepts for road safety applications from the perspective of Powered Two-Wheeler (PTW) riders who suffer multiple injuries in different body regions. The proposed method estimates the overall effectiveness of safety devices for PTW riders by correlating computer simulations with various levels of actual injuries collected worldwide from accident databases. The study further develops the methodology initially presented by Johnny Korner in 1989 by introducing a new indicator, Global Potential Damage (GPD), that overcomes the limitations of the original method, encompassing six biomechanical injury indices estimated in five body regions. A Weibull regression model was fit to the field data using the Maximum Likelihood Method with boundaries at the 90% confidence level for the construction of novel injury risk curves for PTW riders. The modified methodology was applied for the holistic evaluation of the effectiveness of a new safety system, the Belted Safety Jacket (BSJ), in head-on collisions across multiple injury indices, body regions, vehicle types, and speed pairs without sub-optimizing it at specific crash severities. A virtual multi-body environment was employed to reproduce a selected set of crashes. The BSJ is a device concept comprising a vest with safety belts to restrict the rider’s movements relative to the PTW during crashes. The BSJ exhibited 59% effectiveness, with an undoubted benefit to the head, neck, chest, and lower extremities. The results show that the proposed methodology enables an overall assessment of the injuries, thus improving the protection of PTW users. The novel indicator supports a robust evaluation of safety systems, specifically relevant in the context of PTW accidents.

PROTAX-GPU: a scalable probabilistic taxonomic classification system for DNA barcodes.

DNA-based identification is vital for classifying biological specimens, yet methods to quantify the uncertainty of sequence-based taxonomic assignments are scarce. Challenges arise from noisy reference databases, including mislabelled entries and missing taxa. PROTAX addresses these issues with a probabilistic approach to taxonomic classification, advancing on methods that rely solely on sequence similarity. It provides calibrated probabilistic assignments to a partially populated taxonomic hierarchy, accounting for taxa that lack references and incorrect taxonomic annotation. While effective on smaller scales, global application of PROTAX necessitates substantially larger reference libraries, a goal previously hindered by computational barriers. We introduce PROTAX-GPU, a scalable algorithm capable of leveraging the global Barcode of Life Data System (>14 million specimens) as a reference database. Using graphics processing units (GPU) to accelerate similarity and nearest-neighbour operations and the JAX library for Python integration, we achieve over a 1000 × speedup compared with the central processing unit (CPU)-based implementation without compromising PROTAX's key benefits. PROTAX-GPU marks a significant stride towards real-time DNA barcoding, enabling quicker and more efficient species identification in environmental assessments. This capability opens up new avenues for real-time monitoring and analysis of biodiversity, advancing our ability to understand and respond to ecological dynamics. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'.

A market driven identification of current and emerging skills requiring an MBA degree: a topic modelling approach

ABSTRACT The emerging field of Labour Market Intelligence focuses on extracting real-time information for professional skills. Online job posting websites provide unique opportunities to gather such intelligence, useful for job seekers and academic administrators in charge of preparing the future labour force. We propose a probabilistic topic modelling approach, incorporating occupational ontologies, to extract skillsets required for jobs needing a Master of Business Administration (MBA) degree. Analysing LinkedIn job postings in Pennsylvania, USA, we identified 28 job functions and associated skillsets confirming known demands and revealing new ones not yet identified in government and private labour market reports. By aggregating these skills, we can align MBA curricula with market needs, offering a superior alternative to conventional surveys. Our results are comparable with other popular neural models. Our approach provides improved computational efficiency with simpler methods Additionally, this methodology can be applied globally in other fields with existing occupational ontologies.

A new self-adaptive mission aborting policy for systems operating in uncertain random shock environment

For systems operating in random shock environment, mission aborting policies based on the number of shocks experienced during a certain time interval are common and have received intensive attention. However, the number of shocks-based aborting policy (NSAP) is not effective when the shock rate during the primary mission is uncertain because it cannot adapt itself to the shock rate. This paper puts forward a new, self-adaptive inter-shock interval-based aborting policy (ISIAP), which determines the inter-shock interval within which the primary mission should be aborted upon the next shock based on the previous inter-shock interval. A probabilistic approach is suggested for assessing the mission success probability (MSP) and the system survival probability (SSP) under a given ISIAP. The optimal ISIAP problem is formulated and solved with the objective to maximize the expected MSP while meeting certain requirement on the SSP. A detailed case study of an unmanned aerial vehicle performing a payload delivery mission is provided to demonstrate the proposed mission aborting model and compare the effectiveness of the proposed ISIAP and the conventional NSAP. Impacts of the SSP requirement and the shock rate parameter on the mission performance metrics and optimal solutions are also examined in the case study.

A new look at contraceptive prevalence plateaus in sub-Saharan Africa: A probabilistic approach

A new look at contraceptive prevalence plateaus in sub-Saharan Africa: A probabilistic approach

A probabilistic approach to chronic effects assessments for listed species in a vernal pool case study.

Ecological risk assessments for potential pesticide impacts on species listed as threatened or endangered must ensure that decisions to grant registration or establish water quality standards will not jeopardize species or their critical habitats. Pesticides are designed to affect pest species via physiological pathways that may be shared by some nontarget species for which toxicity data are usually unavailable, creating a need for robust methods to estimate acute and chronic toxicity with minimal data. We used a unique probabilistic approach to estimate the risk of chronic effects of two organophosphate (OP) pesticides on the vernal pool fairy shrimp Branchinecta lynchi. Acute toxicity estimates were derived from Monte Carlo (MC) sampling of acute toxicity distributions developed from interspecies relationships using surrogate species. Within each MC draw, acute values were divided by an acute to chronic ratio (ACR) sampled from a distribution of ACRs for OP pesticides and invertebrates, producing a distribution of chronic effects concentrations. The estimated exposure concentrations (EECs) were sampled from distributions representing different environmental conditions. Risk was characterized using probability distributions of acute toxicity, ACRs, and EECs in a probabilistic analysis, as well as partial probabilistic variations that used only some distributions whereas some variables were used deterministically. A deterministic risk quotient (RQ) was compared with the results of probabilistic methods to compare the approaches. Risk varied across exposure scenarios and the number of variables that were handled probabilistically, increasing as the number of variables drawn from distributions increased. The magnitude of RQs was not correlated with the probability that EECs would exceed chronic thresholds, and comparison of the two approaches demonstrates the limited interpretability of RQs. Our novel probabilistic approach to estimating chronic risk with minimal data incorporates uncertainty underlying both exposure and effects assessments for listed species. Integr Environ Assess Manag 2024;00:1-13. Published 2024. This article is a U.S. Government work and is in the public domain in the USA.

Assessment of the prevalence of inadequate iron intakes in premenopausal females based on the reference values of the European Food Safety Authority using cross-sectional food consumption data

BackgroundThe iron intake requirement distribution for premenopausal females is not symmetric, which invalidates the EAR (Estimated Average Requirement) cut-point approach for assessing prevalence of iron inadequacy. Therefore, Beaton’s Full Probability Approach (PA) must be used. While the PA requires information on the entire iron intake requirement distribution, the European Food Safety Authority (EFSA) only provided the EAR (50th percentile), 90th, 95th (PRI) and 97.5th percentiles. ObjectiveThis study aimed to reliably estimate the prevalence of iron inadequacy in premenopausal females using the PA, based on the intake requirements established by EFSA and compare the results with those obtained from the EAR cut-point method. MethodsHabitual iron intakes were calculated using Statistical Program to Assess Dietary Exposure (SPADE) with data from 484 females (20-45 y) from the Dutch National Food Consumption Survey 2012-2016. Iron requirement of EFSA (including additionally obtained information) were applied. Results from the PA were compared to results obtained with the EAR cut-point method. Sensitivity analyses examined the impact of lower iron intake distributions on differences in estimated inadequate intakes between PA and EAR-cut point method. ResultsA twofold higher prevalence of iron inadequacy among Dutch premenopausal females was observed when employing the PA compared to the EAR cut-point method, using EFSA’s reference values. Sensitivity analysis showed that the EAR cut-point method could also result in large overestimations for populations with lower intake distributions. ConclusionsThis study provided an example for using the PA method by using additionally derived information on the full requirement distribution underlying EFSA’s reference values. Results showed once more the unsuitability of the EAR cut-point method to calculate the prevalence of iron inadequacy in premenopausal females. Hence, we recommend that institutions deriving dietary reference values provide all information needed to use the correct method to determine inadequate intakes in the population.

Building consumption anomaly detection: A comparative study of two probabilistic approaches

This paper investigates the performance of two probabilistic approaches, Ensemble batch Prediction Intervals (EnbPI) a conformal prediction approach and XGBoost Location, Scale and Shape (XGBoostLSS), in predicting building energy consumption and in detecting systemic anomalies with proposed alarm matrix. The research questions focus on the effectiveness of these models in providing both point and probabilistic predictions and their utility in identifying collective anomalies. Both models showed good point and distribution prediction performance. For example, the observed point prediction had CV-RMSE in the range of 9 to 17%, outperforming recommendations from ASHRAE. Furthermore, a post-processing stage, the alarm matrix, effectively flags collective, repetitive anomalies, thus shifting focus from conventional point anomalies. The EnbPI-based method yields higher recall rates, with a trade-off of having more false alarms, while XGBoostLSS-based method excels in precision, minimizing overlooked alarms. Moreover, a robustness analysis was carried out to evaluate how these models performed when faced with training datasets containing anomalies. The robustness analysis revealed that the EnbPI-based method was more prone to overfitting, meaning its performance actually improved when the training data included some noise. On the other hand, the XGBoostLSS-based method was more stable, performing well with low levels of noise with performance drop when the noise level was high. While the findings contribute significantly to building energy consumption prediction and anomaly detection, future research could address performance in dynamic environments for the methodology and explore continual learning strategies.

Variation of total alpha and beta activities and Rn-222 concentrations in the water supply system of an Italian volcanic region: How safe is tap water for human consumption?

Total alpha and beta activities and Rn-222 concentrations were determined in water from different sections of seven aqueducts belonging to the water supply system of Campania region (Italy), known worldwide for its volcanism. Statistical analysis was performed on data to account for their variability across the aqueduct sections, and results were discussed considering the geology of reservoirs, the potential mixing processes occurring along the pipe network, the building/constituting materials of the aqueduct sections, and the integrity of the infrastructure. Guidelines proposed by Italian and international regulation entities were considered to determine if total alpha and beta activities and Rn-222 concentrations found at the taps of the different aqueducts should be considered detrimental to public health. Based on a deterministic and a stochastic approach, a health risk assessment was also tested for Rn-222, assuming direct ingestion and showering as potential exposure pathways. Results showed that applying guidelines returned an absence of hazard, whereas risk assessment returned a high probability of exposure to unacceptable Rn-222 doses for some aqueducts. Beyond the usefulness of obtained results to plan actions to improve the safety of drinking water in Campania, our outcomes represent a warning for bodies dealing with public health at any level: the use of guidelines can bring an underestimation of the risks exerted by the exposure to Rn-222 on human health. Further, using a probabilistic approach in risk assessment accounting for uncertainty can favor risk forecasts based on more “realistic” scenarios. Environmental implicationWaters along supply systems show variable radioactivity values and Rn content due to the geology of reservoirs and mixing, dilution and losses occurring during their journey to the taps. Drinking this water (and showering) can expose the population to low doses of ionizing radiation and Rn concentrations, increasing the risk of long-term effects on health. In many countries, the hazard assessment is usually based on applying guidelines. We demonstrated how this approach underestimates the risk for Rn, showing that an exposure assessment is needed to have more reliable and health-conservative results.

Epstein-Barr Virus in the Cerebrospinal Fluid and Blood Compartments of Patients With Multiple Sclerosis and Controls.

Epstein-Barr virus (EBV) infection is a major risk factor of multiple sclerosis (MS). We examined the presence of EBV DNA in the CSF and blood of patients with MS and controls. We analyzed whether EBV DNA is more common in the CSF of patients with MS than in controls and estimated the proportions of EBV-positive B cells in the CSF and blood. CSF supernatants and cells were collected at diagnostic lumbar punctures from 45 patients with MS and 45 HLA-DR15 matched controls with other conditions, all participants were EBV seropositive. Cellular DNA was amplified by Phi polymerase targeting both host and viral DNA, and representative samples were obtained in 28 cases and 28 controls. Nonamplified DNA from CSF cells (14 cases, 14 controls) and blood B cells (10 cases, 10 controls) were analyzed in a subset of participants. Multiple droplet digital PCR (ddPCR) runs were performed per sample to assess the cumulative EBV positivity rate. To detect viral RNA as a sign of activation, RNA sequencing was performed in blood CD4-positive, CD8-positive, and CD19-positive cells from 21 patients with MS and 3 controls. One of the 45 patients with MS and none of the 45 controls were positive for EBV DNA in CSF supernatants (1 mL). CSF cellular DNA was analyzed in 8 independent ddPCRs: EBV DNA was detected at least once in 18 (64%) of the 28 patients with MS and in 15 (54%) of the 28 controls (p = 0.59, Fisher test). The cumulative EBV positivity increased steadily up to 59% in the successive ddPCRs, suggesting that all individuals would have reached EBV positivity in the CSF cells, if more DNA would have been analyzed. The estimated proportion of EBV-positive B cells was >1/10,000 in both the CSF and blood. We did not detect viral RNA, except from endogenous retroviruses, in the blood lymphocyte subpopulations. EBV-DNA is equally detectable in the CSF cells of both patients with MS and controls with ddPCR, and the probabilistic approach indicates that the true positivity rate approaches 100% in EBV-positive individuals. The proportion of EBV-positive B cells seems higher than previously estimated.

A Novel Probabilistic Approach for Driving the Sustainable Energy Circular Economy: Innovating Efficiency in Renewable Resource Markets

Our research introduces a novel hybrid model aimed at driving the sustainable economy by innovating efficiency in forecasting renewable resource market trends for recovery. The fusion of Hierarchical Recurrent Neural Networks (RNNs) and LeNet-5 Convolutional Neural Networks (CNNs) capitalizes on the distinct advantages of both architectures. RNNs capture temporal dependencies, while CNNs identify spatial patterns, providing a holistic understanding of market dynamics. This symbiosis significantly enhances predictive accuracy, empowering stakeholders with more reliable insights into the trajectory of renewable resource markets. Efficient feature selection is achieved through the incorporation of the Modified Hill Climbing (MHC) algorithm. As an evolutionary feature selection technique, MHC dynamically refines input features during the learning process, adapting to the evolving nature of renewable resource markets. This adaptive feature selection not only improves computational efficiency and interpretability but also ensures a focus on the most relevant information, aligning with the goal of driving sustainable circular economic recovery. The fortified hybrid model, harmonizing RNNs, CNNs, and MHC, showcases heightened robustness and generalization across diverse market conditions. By effectively merging deep learning architectures with an evolutionary feature selection process, the model adeptly captures intricate patterns while mitigating overfitting. This contribution underscores the applicability of the proposed approach to real-world scenarios, positioning it as a catalyst for innovative efficiency in renewable resource markets—a pivotal driver for the recovery of the sustainable economy. Our research marks a crucial advancement in the realm of reliable and adaptable forecasting models, essential for informed decision-making in the dynamic landscape of renewable energy markets.

Use of proper sampling techniques to research studies

This article extensively examines several sampling strategies, emphasising probability and non-probability sampling approaches. Sampling is essential in research as it influences the degree to which research findings accurately represent and can be applied to a larger population. This article provides a definition of probability sampling and explores its different approaches. The text highlights the benefits of probability sampling, such as its capacity to guarantee impartial selection and facilitate statistical inference. Furthermore, this article delves into the constraints and factors to be taken into account regarding probability sampling, including the need for resources and the possibility of mistakes in sample frames. Subsequently, this paper examined non-probability sampling strategies and investigated methodologies such as convenience sampling, purposive sampling, and quota sampling. This article examines the fundamental concepts of non-probability sampling and highlights its specific applicability in investigating groups that are rare or challenging to access. A comprehensive examination was undertaken to assess the benefits and drawbacks of non probability sampling, with a particular focus on problems such as researcher bias and the absence of statistical generalizability. This article offers a thorough examination of probability and non-probability sampling approaches, allowing scholars to acquire a more profound comprehension of the benefits, constraints, and suitable uses of each approach. It acknowledges the significance of choosing the most suitable sampling method according to research goals, available resources, and the attributes of the target population. In conclusion, this work offers significant resources for researchers who wish to make well-informed choices on sampling procedures in their research.

Slope Stability and Reliability Analysis of Earth Embankment Constructed for the Doubling of Railway Track

Embankments of greater height pose threat to its stability and require geotechnical investigation and intervention. Conventionally, limit equilibrium approach is used to estimate the factor of safety of the slope to assess its stability. Over a period of time, with experience and engineering judgments, a factor of safety of 1.4 is deemed sufficient for Indian practice. In the approach, it is inherently assumed that different sources of uncertainty due to testing errors, model transformation, and inherent variability of the geo-material are taken care of with the use of single value of factor of safety. Still, a question remains, whether the suggested factor of safety is acceptable in a given environment of uncertainty or there is a need to further utilize the probabilistic approach. The present study demonstrates how combination of First Order Reliability Method (FORM), Finite Element Method (FEM) and Response Surface Method (RSM) can be useful in the probabilistic assessment of proposed construction of new earth embankment for the doubling of a railway track. It is further highlighted that conventional factor of safety approach when used in conjunction with probabilistic analysis brings rationality in decision making. Also, based on results of slope stability and reliability analysis, it is proposed to incorporate soil nailing to further improve the stability of the earth embankment.

Fatigue Crack and Residual Life Prediction Based on an Adaptive Dynamic Bayesian Network

Monitoring the health status of aerospace structures during their service lives is a critical endeavor, aimed at precisely evaluating their operational condition through observation data and physical modeling. This study proposes a probabilistic assessment approach utilizing Dynamic Bayesian Networks (DBNs), enhanced by an improved adaptive particle filtering technique. This approach combines physical modeling with various predictive sources, encompassing cognitive uncertainties inherent in stochastic predictions and crack propagation forecasts. By employing crack observation data, it facilitates predictions of crack growth and the residual life of metal structure. To demonstrate the efficacy of this method, the research leverages data from three-point bending and single-edge tension fatigue tests. It gathers data on crack length during the fatigue crack progression, integrating these findings with digital twin theory to forecast the residual fatigue life of the specimens. The outcomes show that the adaptive DBN model can precisely predict fatigue crack propagation in test specimens, offering a potential tool for the online health assessment and life evaluation for aerospace structures.