Proposed Analysis Framework Research Articles

Integrated [formula omitted] anti-disturbance synchronization control for switched neural networks with unknown delays

This research is concerned with the integrated L∞ anti-disturbance synchronization control problem for continuous-time switched neural networks (SNNs). Different from the relevant synchronization results about SNNs, multiple disturbances and unknown delays are simultaneously addressed in a uniform framework. By constructing the novel Lyapunov function and the ameliorative combined switching strategy, an L∞ performance analysis framework with less conservativeness is then established, where each individual subnetwork is unnecessary to possess a prescribed L∞ performance index. By designing an appropriate disturbance observer and transforming the delay-dependent neuron activation function into a bounded disturbance, an integrated L∞ anti-disturbance synchronization control paradigm is also presented on the basis of the proposed analysis framework. Finally, three illustrative examples are employed to substantiate the applicability and superiority of the developed theoretical results.

Assessing urban green infrastructure: A simple and practical measure of its spatial distribution equity and a comprehensive evaluation

Urban green infrastructure (UGI) that provides diverse ecosystem services are crucial components of a city. Achieving appropriate density and equitable distribution of UGI is highly significant for urban planning to ensure sustainable and high-quality development. However, there is still a lack of an applied method for effectively measuring spatial equity of UGI in cities with multiple scales, which impairs the ability to regularly monitor it. Furthermore, the extent to which UGI should be implemented in urban planning in China is still uncertain. In this study, a “Structure-Function-Zoning“ analysis framework for the spatial pattern of UGI is proposed. A monitoring index system in terms of the structure and functions of UGI is first established. Then, an approach for measurement of spatial distribution equity of UGI is developed based on a spatial zoning strategy along the urban expansion axis. And a weighted index is proposed for comprehensive evaluation of UGI pattern. Using three Chinese cities (i.e., Beijing, Tianjin, Shijiazhuang) as examples, our research shows that the UGI pattern of Tianjin is distributed more balanced from urban center to suburb among the cities and Shijiazhuang shows relatively poor spatial equity of UGI. In terms of comprehensive evaluation of UGI pattern, Beijing and Tianjin are both rated as “moderate” while Beijing was slightly better than Tianjin, and Shijiazhuang was rated as “relatively poor”. The proposed analysis framework is designed to access the structure and function of UGI in a way that can be spatially differentiated, easily described, and clearly compared for cities of different sizes. More broadly, the comprehensive index adopted in the study can be regarded as a simple and practical tool to perform regular monitoring and provide early warning for UGI planning in urban renewal process.

Modular model reduction of interconnected systems: A robust performance analysis perspective

Many complex engineering systems consist of multiple subsystems that are developed by different teams of engineers. To analyse, simulate and control such complex systems, accurate yet computationally efficient models are required. Modular model reduction, in which the subsystem models are reduced individually, is a practical and an efficient method to obtain accurate reduced-order models of such complex systems. However, when subsystems are reduced individually, without taking their interconnections into account, the effect on stability and accuracy of the resulting reduced-order interconnected system is difficult to predict. In this work, a mathematical relation between the accuracy of reduced-order linear-time invariant subsystem models and (stability and accuracy of) resulting reduced-order interconnected linear time-invariant model is introduced. This result can subsequently be used in two ways. Firstly, it can be used to translate accuracy characteristics of the reduced-order subsystem models directly to accuracy properties of the interconnected reduced-order model. Secondly, it can also be used to translate specifications on the interconnected system model accuracy to accuracy requirements on subsystem models that can be used for fit-for-purpose reduction of the subsystem models. These applications of the proposed analysis framework for modular model reduction are demonstrated on an illustrative structural dynamics example.

Towards reliable seismic fragility assessment of highway bridges with oblong columns considering the drift-based capacity directionality effect

Highway bridges in seismic zones are commonly designed with oblong columns, featuring different force-displacement relationships and seismic capacities along different response directions. However, the state-of-art seismic fragility assessment treats seismic demand and capacity models as two independent models, which cannot capture the in-between directional dependence and correlation for oblong columns. This study develops direction-dependent seismic capacity limit state models and proposes a multi-step analysis framework to assess the seismic fragility of highway bridges with flexure-controlled oblong columns. Cyclic pushover analyses are conducted along different loading directions to identify direction-sensitive column drift ratios against four damage states. These drift ratios are utilized to regress a closed-form model, namely the directional capacity ratio model, to quantify the changes in drift capacities as a function of the column response angle. This model is integrated into an analysis framework to directly simulate the demand-capacity ratio (DCR) time history, where the peak value is utilized to assess the seismic fragility of the oblong column. The case study indicates that the most severe seismic damage to an oblong column is not necessarily along its principal axis. The proposed analysis framework outperforms existing approaches in having more reliable fragility estimates for the oblong column and bridge system.

Multi-scenario dynamic prediction of ecological risk assessment in an arid area of northwest China

Ecological security is essential for the development of environmental health and human society. However, the driving mechanisms and spatial non-stationarity of ecological risks are still unknown, particularly in arid areas. Thus, this study constructed an ecological risk index (ERI) research framework in arid areas based on land use and socio-economic data from 1980 to 2020 to investigate the spatiotemporal evolution trend, driving mechanisms, and spatial non-stationarity of ecological risk and predict the ecological risk under business-as-usual, economic development, and ecological protection scenarios using a patch-generating land use simulation (PLUS) model. The results show that: (1) cropland and build-up land were expanding, whereas forestland and grassland water bodies were shrinking. Compared with the two development stages from 1980 to 2000 and from 2000 to 2020, cropland increased by 4.73% and 51.52%, respectively, whereas build-up land increased by 12.64% and 106.52%, respectively, indicating rapid development and drastic changes in the past 20 years. (2) With the change in landscape diversity, the regional ecological risks showed an increasing trend, whereas the future ecological risks showed an increasing trend because of the cumulative effect of landscape pattern changes caused by human activities. (3) Moreover, dramatic changes in ecological risk, particularly in areas of intense human activity in build-up land, agricultural land, artificial oases, and mountainous areas. Corresponding to the analysis of driving mechanisms, population, GDP, distance to government, water, and primary roads; have a significant effect on ecological risk, whereas DEM has direct and indirect effect paths. Ultimately, the proposed analysis framework can present the spatiotemporal changes of ecological risks and their influencing mechanisms, which can help develop sustainable development policies.

Material efficiency strategies across the industrial chain to secure indium availability for global carbon neutrality

A global energy transition toward carbon neutrality system requires increased critical metals, and the availability of critical byproduct metals raises concerns due to their dependence on host metals. To ensure the supply security of critical byproduct metals, material efficiency (ME) strategies are needed. Taking indium as an example, this paper explores the indium availability for global energy transition goals, and then illustrates the effects of ME strategies on securing indium availability throughout the whole industry chain. The system dynamics model serves to understand the joint production mechanism of zinc-indium and the nexus between energy transition and indium demand. Results show that the indium supply shortage for global energy transition could be intensified due to a less zinc-intensive economy. To address this challenge, focusing on indium demand reduction is more effective than emphasizing expanding indium supply for global sustainable development goals and material content reduction is the most promising strategy among the five ME strategies. Whereas, eliminating the concerns of indium availability thoroughly for facilitating a carbon neutrality transition requires coordinated ME strategies both from supply-side and demand-side. Moreover, the proposed analysis framework of ME strategies for securing indium availability for energy transition could also be applied for other byproduct metals.

Exploring the potential of material information in patent data: The case of solid-state batteries

The investigation of high-performance materials represents an essential step in the battery industry to develop innovative technologies. While patent analysis could help promote research and development (R&D) by providing insights into innovative activities, the significance of material information contained in patent documents for the advancement of battery technologies has not yet been investigated. This study explored the potential of material information derived from patent data for analyzing the patent landscape by introducing a novel analysis framework. It uses patent claims collected between 2010 and 2021 from the domain of solid-state battery technology to extract and transform material information. This study relies on 1-mode network analysis to construct a material network and a 2-mode network analysis is applied to quantify the extent of the chemical portfolios of assignees. The results indicate that the polymer-based electrolytes are highlighted as the dominant type in the solid-state batteries. A glance at the chemical portfolios of assignees reveals a heterogeneous technology profile, pointing to a highly diversified material knowledge across several industries. The proposed analysis framework is expected to offer a new empirical perspective on the patent landscape in the battery sector by exploiting material information. Broadening the view of material research within the battery sector can support R&D managers and research scientists in the design of improved material search strategies and the selection of appropriate collaborations partners for the respective process.

Investigating the Stiffness Characteristics of a Tendon-Driven Continuum Manipulator Using Sensitivity Analysis: A Case Study in Transoral Laser Microsurgery

Continuum manipulators, with their characteristics of flexibility and dexterity, have gained significant interest in various applications across industries such as inspection, manufacturing, space exploration, and medical surgery. However, because of their inherent compliance, handling payloads may prove challenging due to shape distortion and deflection. This demonstrates the need to optimize the manipulator’s stiffness. The primary objective of this work was to show the merits of sensitivity analysis in the design of flexible surgical manipulators. Such analysis can guide important design decisions and enable the more efficient use of available resources, contributing to designing more effective prototypes. A new sensitivity analysis framework based on a multi-model and a multi-method approach was proposed to achieve this. This framework was then demonstrated by studying a tendon-driven rolling contact joint hyper-redundant manipulator for transoral laser microsurgery. In this analysis, the effects of independent design parameters on the stiffness of the manipulator were examined. Then, scaled-up 3D-printed prototypes were used to validate the accuracy of the stiffness model experimentally, which enabled us to assess the outcome of the sensitivity analysis framework. The results demonstrated that only two out of five design parameters for the considered manipulator significantly impacted the device’s performance. This information could enable the designer to efficiently allocate resources toward correctly setting these two most important parameters to achieve the desired system. Overall, the proposed analysis framework is a general tool that can be applied to any design architecture, helping to develop optimal manipulators for various applications.

An EEG-fNIRS neurovascular coupling analysis method to investigate cognitive-motor interference

Background and Objective:The simultaneous execution of a motor and cognitive dual task may lead to the deterioration of task performance in one or both tasks due to cognitive-motor interference (CMI). Neuroimaging techniques are promising ways to reveal the underlying neural mechanism of CMI. However, existing studies have only explored CMI from a single neuroimaging modality, which lack built-in validation and comparison of analysis results. This work is aimed to establish an effective analysis framework to comprehensively investigate the CMI by exploring the electrophysiological and hemodynamic activities as well as their neurovascular coupling. Methods:Experiments including an upper limb single motor task, single cognitive task, and cognitive-motor dual task were designed and performed with 16 healthy young participants. Bimodal signals of electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) were recorded simultaneously during the experiments. A novel bimodal signal analysis framework was proposed to extract the task-related components for EEG and fNIRS signals respectively and analyze their correlation. Indicators including within-class similarity and between-class distance were utilized to validate the effectiveness of the proposed analysis framework compared to the canonical channel-averaged method. Statistical analysis was performed to investigate the difference in the behavior and neural correlates between the single and dual tasks. Results:Our results revealed that the extra cognitive interference caused divided attention in the dual task, which led to the decreased neurovascular coupling between fNIRS and EEG in all theta, alpha, and beta rhythms. The proposed framework was demonstrated to have a better ability in characterizing the neural patterns than the canonical channel-averaged method with significantly higher within-class similarity and between-class distance indicators. Conclusions:This study proposed a method to investigate CMI by exploring the task-related electrophysiological and hemodynamic activities as well as their neurovascular coupling. Our concurrent EEG-fNIRS study provides new insight into the EEG-fNIRS correlation analysis and novel evidence for the mechanism of neurovascular coupling in the CMI.

System Reliability Analysis for Truss Structures Based on Automatic Updated�Model

Multiple failure modes tend to be identified in the reliability analysis of a redundant truss structure. This identification process involves updating the model for identifying the next potential failure members. Herein we intend to update the finite element model automatically in the identification process of failure modes and further perform the system reliability analysis efficiently. This study presents a framework that is implemented through the joint simulation of MATLAB and APDL and consists of three parts: reliability index of a single member, identification of dominant failure modes, and system-level reliability analysis for system reliability analysis of truss structures. Firstly, RSM (response surface method) combines with a constrained optimization model to calculate the reliability indices of members. Then the β-unzipping method is adopted to identify the dominant failure modes, and the system function in MATLAB, as well as the EKILL command in APDL, is used to facilitate the automatic update of the finite element model and realize load-redistribution. Besides, the differential equivalence recursion algorithm is performed to approximate the reliability indices of failure modes efficiently and accurately. Eventually, the PNET (probabilistic network evaluation technique) is used to calculate the joint failure probability as well as the system reliability index. Two illustrative examples demonstrate the accuracy and efficiency of the proposed system reliability analysis framework through comparison with corresponding references.

Design and Analysis of Hardware-Limited Non-Uniform Task-Based Quantizers

Hardware-limited task-based quantization is a new design paradigm for data acquisition systems equipped with serial scalar analog-to-digital converters using a small number of bits. By taking into account the underlying system task, task-based quantizers can efficiently recover the desired parameters from the low-bit quantized observation. Current design and analysis frameworks for hardware-limited task-based quantization are only applicable to inputs with bounded support and uniform quantizers with non-subtractive dithering. Here, we propose a new framework based on generalized Bussgang decomposition that enables the design and analysis of hardware-limited task-based quantizers that are equipped with non-uniform scalar quantizers or that have inputs with unbounded support. We first consider the scenario in which the task is linear. Under this scenario, we derive new pre-quantization and post-quantization linear mappings for task-based quantizers with mean squared error (MSE) that closely matches the theoretical MSE. Next, we extend the proposed analysis framework to quadratic tasks. We demonstrate that our derived analytical expression for the MSE accurately predicts the performance of task-based quantizers with quadratic tasks.

Towards a hydrogeomorphological understanding of proglacial catchments: an assessment of groundwater storage and release in an Alpine catchment

Abstract. Proglacial margins form when glaciers retreat and create zones with distinctive ecological, geomorphological and hydrological properties in Alpine environments. There is extensive literature on the geomorphology and sediment transport in such areas as well as on glacial hydrology, but there is much less research into the specific hydrological behavior of the landforms that develop after glacier retreat in and close to proglacial margins. Recent reviews have highlighted the presence of groundwater stores even in such rapidly draining environments. Here, we describe the hydrological functioning of different superficial landforms within and around the proglacial margin of the Otemma glacier, a temperate Alpine glacier in the Swiss Alps; we characterize the timing and amount of the transmission of different water sources (rain, snowmelt, ice melt) to the landforms and between them, and we compare the relationship between these processes and the catchment-scale discharge. The latter is based upon a recession-analysis-based framework. In quantifying the relative groundwater storage volumes of different superficial landforms, we show that steep zones only store water on the timescale of days, while flatter areas maintain baseflow on the order of several weeks. These landforms themselves fail to explain the catchment-scale recession patterns; our results point towards the presence of an unidentified storage compartment on the order of 40 mm, which releases water during the cold months. We suggest attributing this missing storage to deeper bedrock flowpaths. Finally, the key insights gained here into the interplay of different landforms as well as the proposed analysis framework are readily transferable to other similar proglacial margins and should contribute to a better understanding of the future hydrogeological behavior of such catchments.

Evaluation of resources and environmental carrying capacity of marine ranching in China: An integrated life cycle assessment-emergy analysis

Resources and environmental carrying capacity (RECC) describes the ability of a system to achieve healthy and sustainable development. Various marine ranching enterprises have emerged in China in recent years, which have aroused concern and debate about the RECC of marine ranching systems. By taking the environmental impact calculated by life cycle assessment (LCA) into consideration in emergy analysis (EA), this study evaluated the comprehensive RECC performance of the whole system and each stage of a marine ranching system in China. The resource use efficiency (RUE) and system carrying ratio (SCR) of the system were reasonably good. However, its environmental loading ratio (ELR), emergy yield ratio (EYR), and emergy sustainability index (ESI) were unsatisfactory. First, the nonrenewable resources dominated the emergy input. Second, the emergy input from the purchased resources was much greater than that of local resources. Third, the potential environmental impact mainly came from the construction stage. Fourth, serious overload of RECC was observed in the maintenance stage. The results indicate that the system is efficient, and its RECC is in a safe state, but the system has deficiencies in environmental protection and the exploitation and utilization of local resources. The proposed analysis framework helps us comprehensively understand the marine ranching RECC performance and provides a research paradigm reference for the sustainability analysis of other complex eco-economic systems.

Reliability assessment of guyed transmission towers through active learning metamodeling and progressive collapse simulation

This manuscript presents a novel approach to the reliability of guyed transmission line towers (TLTs) considering complete geometric and material non-linear analyses, bolt slippage effects, and progressive collapse simulation, using active learning Kriging to handle the computational burden. In the existing literature, such issues have been considered separately, with the main focus on self-supporting towers. Herein, novelty arises from the combined consideration of the above and the addressing of guyed TLTs. Yet, because the wind speed variability is much higher than those related to the mechanical model, a modified initial sampling plan strategy is successfully proposed to reduce the computational time required. The application case-study addressing long transmission lines in midwest Brazil is also relevant, considering the recent updating of design wind speeds, with increases of up to 25%. Failure probabilities obtained using code resistance models indicate the need for tower retrofitting. However, the more accurate assessment via collapse simulation shows that the support is still safe, even for the updated wind speeds. Results demonstrate the accuracy and efficiency of the proposed analysis framework and show the importance of using accurate numerical models in retrofit evaluations.

Lifetime multi-hazard fragility analysis of transmission towers under earthquake and wind considering wind-induced fatigue effect

Wind-induced fatigue damage is a predominant factor eventuating in structural performance degradation of transmission towers exposed to long-term environmental wind. Moreover, transmission towers may suffer from multiple extreme natural hazards (such as earthquake and strong wind) during their service lives. However, these phenomena have not been well accounted for in current design specifications and previous studies for transmission towers. To this end, this paper proposes a comprehensive probabilistic analysis framework for achieving the lifetime multi-hazard fragility evaluation by considering the wind-induced fatigue, which is applied in a 1000 kV steel tubular transmission tower. Firstly, based on the field wind data measured at the meteorological station of Qingdao city, China, from Jan. 1, 1971 to Dec. 31, 2020, a fatigue analysis is performed on an actual transmission tower constructed in this region to compute the annual mean fatigue damage. Secondly, a series of time-dependent uncertain numerical models of the tower are established by integrating the accumulated fatigue damage and material deterioration model, and nonlinear time history analyses are imposed on these models to examine the effect of fatigue damage on lifetime responses under combined earthquake and wind loads. Analogously, the time-dependent probabilistic multi-hazard demand models are derived via the dynamic analyses and statistical regression, and these demand models are employed to further develop the lifetime multi-hazard fragility surfaces of the tower. Finally, a fast lifetime safety assessment method, i.e. the critical safety surface, is proposed to efficiently judge the collapse state of the transmission tower at various wind speeds, PGAs and service times. Numerical results indicate that the structural capacity of the transmission tower will experience significant deterioration due to wind-induced fatigue damage; the multi-hazard induced structural responses and fragilities will considerably increase with the service life although the tower is initially well-designed. The proposed analysis framework will be beneficial for the design and maintenance of transmission towers subjected to both earthquake and wind hazards for designers and decision-makers.

Analysis and optimization of thermoelastic structures with tension–compression asymmetry

Many engineering and biological materials have been found to have tension–compression asymmetry properties. An extension of the linear thermoelastic behavior called a bi-modulus constitutive model is proposed to study the impact of thermo-mechanical coupling in such materials and structures. The well-posedness of the corresponding bi-modulus thermoelasticity problems is confirmed via a variational approach. An efficient numerical analysis method is developed for such nontrivial material behavior and is further employed for structural topology optimization. The importance of the proposed analysis and design framework is demonstrated by the significant impact of the tension–compression asymmetry and thermal effect on structural responses and optimum structural configurations.

Estimating effectiveness of speed reduction measures for pedestrian crossing treatments using an empirically supported speed choice modeling framework

Pedestrian crossing treatments, such as Pedestrian Hybrid Beacons (PHBs) and Rectangular Rapid-Flashing Beacons (RRFBs), are traffic control devices implemented to help pedestrians safely cross busy or higher-speed roadways at midblock crossings and uncontrolled intersections. This study aims to assess the effectiveness of the PHB and RRFB by analyzing drivers’ speeding behavior under different roadway types with real-life traffic conditions. In order to understand the effect of pedestrian crossing treatments (i.e., PHB and RRFB) have on drivers’ speeding behavior, this paper analyzes four zones (i.e., one upstream zone and three consecutive downstream zones). For the four analysis zones, different indexes were computed which corresponds to the differences in drivers’ speed when encountered with the pedestrian crossing treatments. A grouped random effect hurdle beta regression model is estimated for the indexes, with a fixed effect hurdle beta model used for comparison and validating the importance of considering the grouped random heterogeneity across participants. The proposed analysis framework was validated by the means of an empirical driving simulator study, based on two urban roads in the Central Florida region, North Alafaya Trail (SR-434) and South Orange Blossom Trail (US-441). The results revealed that the proposed modeling framework reflects drivers’ difference in speed for the different pedestrian crossing treatments. The results suggest that with proper understanding of the PHB, the PHB can reduce drivers speed even beyond the location of the PHB. Meanwhile, the RRFB does have some effect in speed reduction beyond the location of the RRFB, however many drivers failed to acknowledge the RRFB. It is suggested that when drivers’ have proper education on the use of the PHB to reduce speed safely and for the installation of RRFB be on roads with two or less lanes and a speed limit less than 40 mph. While the main purpose of the pedestrian crossing treatments is to help pedestrians cross safely, speed reduction can be considered a byproduct as revealed in this study.

How to Measure the Network Vulnerability of Cities to Wildfires: Cases in California, U.S.A.

This paper proposes a framework to evaluate the network vulnerability of cities to wildfires. Three cities are selected from the California Public Utilities Commission (CPUC), U.S., fire-threat regions: Orinda, Paradise, and Atascadero. For each city, four different network connectivity measures are calculated, and agent-based evacuation simulations are performed by the Monte Carlo method. In the simulations, the number of isolated vehicles and evacuation time estimates are measured for the following scenarios: (i) no wildfire case with original network; and (ii) wildfire cases with randomly damaged networks that are reduced by 1%, 3%, 5%, 7%, and 10% from the original network. A city-to-city comparison is conducted in relation to network connectivity measures and evacuation simulation results. It is shown that Paradise has the worst network connectivity, and the simulation results reveal that Paradise also has the most sensitive network in relation to random roadway closures caused by wildfire propagation. Thus, among the three cities, Paradise has the most vulnerable network to wildfires as determined through the two analysis results concerning the worst network measures and the simulation results. It is expected that the proposed analysis framework can be generally applied to any city located in a fire-threat region.

Theoretical Performance Analysis of Distributed Queue for Massive Machine Type Communications: Throughput, Latency, Energy Consumption

Massive machine type communications (mMTC) is one of main application cases in 5G, which is supposed to support communications of massive number of machine-type devices (MTDs). Distributed queue (DQ) is a variant of tree splitting protocol which combines an m-ary tree splitting algorithm with a set of simple smart rules, organizing every terminal in one out of two virtual queues. Theoretically, DQ allows access to infinite terminals and is stable under any traffic condition, which alleviates the unstable problem of slotted ALOHA, and is especially suitable for mMTC. However, its theoretical comprehensive performance analysis as well as related statistical characteristics is still missing, which severely restricts the full manifestation of its performance advantages. In view of this, the paper proposes a general performance analysis framework for DQ, with which full probability space of DQ evolution process is presented for the first time. To be more specific, probability distribution function (PDF), mean and variance of throughput, latency and energy consumption of DQ is analytically derived to comprehensively evaluate performance. Taking the IEEE 802.15.4 standard for mMTC as example, numerical results validate the accuracy of the proposed analysis framework and the stability of DQ, present effects of number of MTDs, number of contention slots ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${m}$ </tex-math></inline-formula> ), and maximum number of transmissions ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${L}$ </tex-math></inline-formula> ) on DQ in terms of aforementioned performance metrics. These results together provide good reference to find appropriate value of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${m}$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${L}$ </tex-math></inline-formula> to balance the performance metrics and enable more practical network optimization.

Medical checkup data analysis method based on LiNGAM and its application to nonalcoholic fatty liver disease

Although medical checkup data would be useful for identifying unknown factors of disease progression, a causal relationship between checkup items should be taken into account for precise analysis. Missing values in medical checkup data must be appropriately imputed because checkup items vary from person to person, and items that have not been tested include missing values. In addition, the patients with target diseases or disorders are small in comparison with the total number of persons recorded in the data, which means medical checkup data is an imbalanced data analysis. We propose a new method for analyzing the causal relationship in medical checkup data to discover disease progression factors based on a linear non-Gaussian acyclic model (LiNGAM), a machine learning technique for causal inference. In the proposed method, specific regression coefficients calculated through LiNGAM were compared to estimate the causal strength of the checkup items on disease progression, which is referred to as LiNGAM-beta. We also propose an analysis framework consisting of LiNGAM-beta, collaborative filtering (CF), and a sampling approach for causal inference of medical checkup data. CF and the sampling approach are useful for missing value imputation and balancing of the data distribution. We applied the proposed analysis framework to medical checkup data for identifying factors of Nonalcoholic fatty liver disease (NAFLD) development. The checkup items related to metabolic syndrome and age showed high causal effects on NAFLD severity. The level of blood urea nitrogen (BUN) would have a negative effect on NAFLD severity. Snoring frequency, which is associated with obstructive sleep apnea, affected NAFLD severity, particularly in the male group. Sleep duration also affected NAFLD severity in persons over fifty years old. These analysis results are consistent with previous reports about the causes of NAFLD; for example, NAFLD and metabolic syndrome are mutual and bi-directionally related, and BUN has a negative effect on NAFLD progression. Thus, our analysis result is plausible. The proposed analysis framework including LiNGAM-beta can be applied to various medical checkup data and will contribute to discovering unknown disease factors.