Reliable Estimation Performance Research Articles

Machine learning-driven predictions and interventions for cardiovascular occlusions.

Cardiovascular diseases remain a leading cause of global morbidity and mortality, with heart attacks and strokes representing significant health challenges. The accurate, early diagnosis and management of these conditions are paramount in improving patient outcomes. The specific disease, cardiovascular occlusions, has been chosen for the study due to the significant impact it has on public health. Cardiovascular diseases are a leading cause of mortality globally, and occlusions, which are blockages in the blood vessels, are a critical factor contributing to these conditions. By focusing on cardiovascular occlusions, the study aims to leverage machine learning to improve the prediction and management of these events, potentially helping to reduce the incidence of heart attacks, strokes, and other related health issues. The use of machine learning in this context offers the promise of developing more accurate and timely interventions, thus improving patient outcomes. We analyze diverse datasets to assess the efficacy of various machine learning algorithms in predicting heart attacks and strokes, comparing their performance to pinpoint the most accurate and reliable models. Additionally, we classify individuals by their predicted risk levels and examine key features that correlate with the incidence of cardiovascular events. The PyCaret machine learning library's Classification Module was key in developing predictive models which were evaluated with stratified cross-validation for reliable performance estimates. Our findings suggest that machine learning can significantly improve the prediction accuracy for heart attacks and strokes, facilitating earlier and more precise interventions. We also discuss the integration of machine learning models into clinical practice, addressing potential challenges and the need for healthcare professionals to interpret and apply these predictions effectively. The use of machine learning for risk stratification and the identification of modifiable factors may empower preemptive approaches to cardiovascular care, ultimately aiming to reduce the occurrence of life-threatening events and improve long-term patient health trajectories.

Experimental investigations of upscaling effects of high-temperature heat pumps with R1233zd(E)

High-temperature heat pumps (HTHP) are crucial for decarbonizing district heating and industry. Hence, a reliable performance estimation considering effects of scaling is of interest for cost-efficient system integration. This study compares two laboratory HTHPs with the same refrigerant R1233zd(E). These systems have a similar plant design but a capacity scale factor of 3.2. The aim is to evaluate scale effects and make general valid statements about the challenges of upscaling laboratory test results. The analyses show that it is essential to distinguish between design effects like temperature differences at the pinch point or different heat exchanger surfaces and scale effects like relative heat losses and efficiencies of components and the system. At the component level, large surface reserves in the heat exchangers lead to 3–5 K smaller approach temperature differences in the lower capacity HTHP. At the system level, the relative heat losses are about 15 % higher than in the system with a larger capacity. The coefficient of performance of both HTHPs shows a similar trend as a function of the temperature lift from the heat source to the sink. Cross-referencing these performance results with over 200 data points from the literature, validated the possibility of upscaling.

An event‐triggered zonotopic Gaussian state estimator for discrete‐time stochastic multiplicative systems with zonotopic set‐membership and stochastic uncertainties

SummaryThis article addresses the challenge of estimating the state of a discrete‐time stochastic multiplicative system with zonotopic set‐membership and stochastic uncertainties. The stochastic system model incorporates additive and multiplicative Gaussian and zonotopic set‐membership uncertainties, posing challenges in accurately managing these sources of uncertainty. In practical scenarios with limited communication resources, it is crucial to reduce communication overhead, save resources, and prevent unnecessary information transmission. To achieve these objectives, this article proposes an event‐triggered mechanism and a corresponding state estimator. The optimal gain matrix of the state estimator is derived to minimize the estimation uncertainty represented by the size of zonotope and Gaussian confidence ellipsoid regions. Conditions are established to guarantee the mean square stability of the estimation error, ensuring reliable estimation performance. The configuration of the triggering condition parameters, which play a vital role in determining the event‐triggered updates, is presented. is presented. The effectiveness of the proposed event‐triggered mechanism and the corresponding estimator is demonstrated through simulation results and analyses, using specific metrics and evaluation criteria.

RDCP: A Real Time Sea Fog Intensity and Visibility Estimation Algorithm

A number of accidents at sea are primarily caused by low visibility due to sea fog. Therefore, it is important to estimate sea fog intensity and visibility in real-time in the ocean environment. Visibility meters utilize optical sensors rather than visional information, so that the measured visibility data occasionally includes errors. Moreover, visibility meters have significant costs so that it is not viable to install them at various locations. Therefore, this paper proposes an algorithm called RDCP (Reduced Dark Channel Prior), which provides reliable estimation at a low cost by processing images captured from cameras, since they receive identical information as human eyes. For the estimation, the RDCP algorithm firstly acquires dark channels from an ocean image, then applies an optimized threshold value to the dark channels and crops out the sky region in the ocean image. For the estimation performance evaluation, 320 raw images captured from cameras at four different ports in Republic of Korea are used. Considering facilities in oceans, which are usually operated in power-limited environment, the processing time performance of PDCP is also examined. The experiments demonstrate that the RDCP algorithm provides reliable estimation performance in real-time.

Identification of Uncertain Parameter in Flight Vehicle Using Physics-Informed Deep Learning

This paper presents the estimation method for uncertain parameters in flight vehicles, especially missile systems, based on physics-informed neural networks (PINNs) augmented with a novel integration-based loss. The proposed method identifies four types of structured uncertainty: burnout time, rocket motor tilt angle, location of the center of pressure, and control fin bias, which significantly affect the missile performance. In the estimation framework, as neural networks (NNs) are updated, these uncertainties are also identified simultaneously because they are also included in the structure of NNs. After testing 100 simulation data, the average estimation errors are within 1% of the mean value for each type of uncertainty. The methodology is able to identify the parameters despite noise corruption in the time-series data. Compared with the conventional PINNs, adding the new loss based on the integration of differential equations yields a more reliable estimation performance for all types of uncertainty. This approach can be effective for complex systems and ill-posed inverse problems, which makes it applicable to other aerospace systems.

Dissociating sensitivity from bias in the Mini Profile of Music Perception Skills.

The Mini Profile of Music Perception Skills (Mini-PROMS) is a rapid performance-based measure of musical perceptual competence. The present study was designed to determine the optimal way to evaluate and score the Mini-PROMS results. Two traditional methods for scoring the Mini-PROMS, the weighted composite score and the parametric sensitivity index (d'), were compared with nonparametric alternatives, also derived from signal detection theory. Performance estimates using the traditional methods were found to depend on response bias (e.g., confidence), making them suboptimal. The simple nonparametric alternatives provided unbiased and reliable performance estimates from the Mini-PROMS and are therefore recommended instead.

Optimizing pricing and inventory strategies for dietary supplement production under stochastic demand

PurposeThe increasing popularity of ERP solutions has provided dietary supplement manufacturing companies with modules to manage pricing and inventory. However, the decisions made by these modules are often independent and rely on deterministic forecasts. This paper studies a multi-product dietary supplement manufacturing system under stochastic demands. The purpose is to maximize the long-run expected profit by jointly considering pricing and inventory strategies.Design/methodology/approachThe authors investigate both the general cases and three special cases including stable demand, negligible backlog and instantaneous replenishment. A two-stage algorithm named PAS is proposed. In the strategy construction stage, the constructed objective bounds are combined to provide estimates which then help to derive the optimal product prices. In the system operation stage, replenishment decisions are further made based on the prices generated from the previous stage.FindingsIt is proved that base-stock policy is optimal for the studied system, and the optimal based-stock level is provided. The global optimal strategies are obtained for three important special cases. For the general case, theoretical objective bounds are established. These bounds provide quick and reliable performance estimates for practical applications.Originality/valueVery few studies have jointly considered pricing and inventory strategies with uncertainty demands in the dietary supplement industry. The PAS algorithm developed integrates these decisions and consistently generates high-quality solutions even under highly varying demands. Such algorithm could be a valuable add-on to the pricing and inventory management modules in ERP systems.

High Performance on a Pragmatic Task May Not Be the Result of Successful Reasoning: On the Importance of Eliciting Participants' Reasoning Strategies.

Formal probabilistic models, such as the Rational Speech Act model, are widely used for formalizing the reasoning involved in various pragmatic phenomena, and when a model achieves good fit to experimental data, that is interpreted as evidence that the model successfully captures some of the underlying processes. Yet how can we be sure that participants' performance on the task is the result of successful reasoning and not of some feature of experimental setup? In this study, we carefully manipulate the properties of the stimuli that have been used in several pragmatics studies and elicit participants' reasoning strategies. We show that certain biases in experimental design inflate participants' performance on the task. We then repeat the experiment with a new version of stimuli which is less susceptible to the identified biases, obtaining a somewhat smaller effect size and more reliable estimates of individual-level performance.

Acceptability of wearable inertial sensors, completeness of data, and day-to-day variability of everyday life motor activities in children and adolescents with neuromotor impairments.

Monitoring the patients' motor activities in a real-world setting would provide essential information on their functioning in daily life. In this study, we used wearable inertial sensors to monitor motor activities of children and adolescents with congenital and acquired brain injuries. We derived a set of clinically meaningful performance measures and addressed the following research questions: Is the target population willing to wear the sensors in their habitual environment? Which factors lead to missing data, and can we avoid them? How many measurement days are needed to obtain reliable estimates of the children's and adolescents' motor performance? The study participants wore our sensor system for seven consecutive days during waking hours. First, we derived the daily hand use of all participants, the duration of different body positions and the wheeling activity of individuals using a manual wheelchair, and walking-related measures in individuals being able to walk. Then, we analyzed the reasons for missing data and determined the reliability of the performance measures mentioned above. The large majority (41 of 43 participants) was willing to wear the sensor system for a week. However, forgetting to reattach the sensors after charging them overnight and taking them off during bathing and swimming was the main contributor to missing data. Consequently, improved battery life and waterproofness of the sensor technology are essential requirements for measurements in daily life. Besides, 5 of 11 performance measures showed significant differences between weekdays and weekend days. The reliability, measured with the intraclass correlation coefficient, ranged between 0.82 and 0.98. Seven measurement days were enough to obtain significantly higher reliability scores than the desired level of 0.8 for all but two performance measures. In children and adolescents with neuromotor impairments, we recommend monitoring everyday life motor activities on seven consecutive days. The target population accepted this measurement protocol, it covers school days and weekend days, and the number of measurement days is sufficient to obtain reliable estimates of motor performance.

Adaptive Control of Wire-Borne Underactuated Brachiating Robots Using Control Lyapunov and Barrier Functions

Executing safe brachiation maneuvers with a cable-suspended underactuated robot is a challenging problem due to the complications induced by the cable dynamics. We present and experimentally implement an online adaptive controller for a wire-borne brachiating robot swinging on a vibrating cable. An adaptive function approximation approach is proposed to estimate the unknown dynamics of the flexible cable as an external force applied to the robot. Robust control Lyapunov and barrier functions are designed and incorporated into quadratic programs to synthesize a unified adaptive control framework, which formally guarantees the stability and safety of the brachiating robot in the presence of dynamic uncertainties, actuator constraints, and obstacles in the environment. The stability analysis and derivation of the adaptation law are carried out through a Lyapunov analysis. We demonstrate and validate the proposed control framework using extensive hardware experiments with an underactuated brachiating robot operating on a flexible cable. Simulation results, hardware experiments, and comparisons with a baseline controller show that the proposed quadratic programming-based controller achieves reliable tracking performance and disturbance estimation in the presence of model uncertainties, actuator limits, and safety constraints.

Nonideality-Aware Training for Accurate and Robust Low-Power Memristive Neural Networks.

Recent years have seen a rapid rise of artificial neural networks being employed in a number of cognitive tasks. The ever‐increasing computing requirements of these structures have contributed to a desire for novel technologies and paradigms, including memristor‐based hardware accelerators. Solutions based on memristive crossbars and analog data processing promise to improve the overall energy efficiency. However, memristor nonidealities can lead to the degradation of neural network accuracy, while the attempts to mitigate these negative effects often introduce design trade‐offs, such as those between power and reliability. In this work, authors design nonideality‐aware training of memristor‐based neural networks capable of dealing with the most common device nonidealities. The feasibility of using high‐resistance devices that exhibit high I‐V nonlinearity is demonstrated—by analyzing experimental data and employing nonideality‐aware training, it is estimated that the energy efficiency of memristive vector‐matrix multipliers is improved by almost three orders of magnitude (0.715 TOPs−1W−1 to 381 TOPs−1W−1) while maintaining similar accuracy. It is shown that associating the parameters of neural networks with individual memristors allows to bias these devices toward less conductive states through regularization of the corresponding optimization problem, while modifying the validation procedure leads to more reliable estimates of performance. The authors demonstrate the universality and robustness of this approach when dealing with a wide range of nonidealities.

Efficient framework for the simulation of translational and rotational laser speckle displacement in optical sensor assemblies

The optimization and uncertainty analysis of laser-based optical sensors in the design phase is a challenging task due to the presence of stochastic laser speckle effects. We present an accurate, efficient, and versatile simulation framework for the design of optical sensor assemblies, capable of handling objective as well as subjective speckle effects. The framework integrates the stochastic nature of laser speckle with the deterministic properties of ray-tracing simulations, enabling the simulation of sensitivities to translational as well as rotational target motion and reliable performance estimation, even for more complex optical assemblies. To validate the simulation results for translation and rotation, they are compared against the experimental data of four speckle-based optical sensor assemblies as well as against analytical relations for speckle pattern motion. The accuracy of the developed framework is demonstrated by simulation errors for correlation peak shift of the speckle pattern of less than 2 μm rms and 2.4 μm rms for translation and rotation, respectively. For the center of gravity shift as additional simulation output for an integrated laser sensor for sensing translations in all three degrees of freedom, a simulation error of 2.6 μm rms was obtained, which also lies well below the resolution of the designed optical sensor assemblies.

A Logistic Regression Approach to Field Estimation Using Binary Measurements

In this letter, we consider the problem of field estimation using binary measurements. Previous work has formulated the problem as a parameter estimation problem, which can be solved in an online manner using sequential Monte Carlo (SMC) techniques. In the current work, we consider an alternative approach to the parameter estimation problem based on online logistic regression. The developed algorithm is less computationally intensive than the SMC approach, while having more reliable estimation performance.

Reliability of claims-based measures used to assess cancer care quality.

42 Background: As value-based payment models for cancer care expand, the need for measures which reliably assess the quality of care provided increases. This is especially true for models like the Oncology Care Model (OCM) that rely on quality rankings to determine potential shared savings. Under models like these, unreliable measures may result in arbitrary application of value-based payments. The goal of this project is to evaluate the extent to which measures used within the OCM are reliable indicators of provider performance. Methods: Data for this project came from North Carolina Medicare claims from 2015-2017. Episodes were attributed to physician practices at the tax identification number (TIN) level, lasted 6 months, and were divided into two performance years beginning 1/1/2016 and 7/1/2016. TINs with fewer than 20 attributed patients were excluded. Three claims-based OCM measures were used in this evaluation: 1) proportion of episodes with all-cause hospital admissions; 2) proportion of episodes with all-cause emergency department (ED) visits or observation stays; and 3) proportion of patients that died who were admitted to hospice for 3 days or more. Risk adjustment followed the method described by measure specifications from the OCM. Reliability was calculated as the ratio of between practice variation (e.g. signal) to the sum of between practice variation and within practice variation (e.g. noise). Variance estimates were derived from hierarchical logistic regression models used for risk adjustment. Results: For the hospitalization and ED visit measures, episode counts for years 1 and 2 were 30,746 and 28,430 and TIN counts were 86 and 84, respectively. Hospice use measures had fewer episodes (2,677 and 2,428) and TINs (36 and 33). Across all measures, median reliability scores failed to achieve the recommended 0.7 threshold and only hospice had a median reliability score above 0.5 (Table). Conclusions: These findings suggest claims-based measures included in the OCM may produce imprecise estimates of provider performance and are vulnerable to random variation. Consideration should be given to developing alternative measures which may be more reliable estimates of provider performance and to increasing minimum denominator requirements for existing measures.[Table: see text]

Optimization of scintillator-reflector optical interfaces for the LUT Davis model.

Designing and optimizing scintillator-based gamma detector using Monte Carlo simulation is of great importance in nuclear medicine and high energy physics. In scintillation detectors, understanding the light transport in the scintillator and the light collection by the photodetector plays a crucial role in achieving high performance. Thus, accurately modeling them is critical. In previous works, we developed a model to compute crystal reflectance from the crystal 3D surface measurement and store it in look-up tables to be used in the Monte Carlo simulation software GATE. The relative light output comparison showed excellent agreement between simulations and experiments for both polished and rough surfaces in several configurations, that is, without and with reflector. However, when comparing them at the irradiation depth closest to the photodetector face, rough crystals with a reflector overestimated the predicted light output. Investigating the cause of this overestimation, we optimized the LUT algorithm to improve the reflectance computation accuracy, especially for rough surfaces. However, optical Monte Carlo simulations carried out with these newly generated LUTs still overestimate the light output. Based on previous observations, one probable cause is the erroneous assumption of perfect couplings between the reflector and crystal and between the crystal and photodetector, which likely results in an important overestimation of the light output compared to experimental values. In practice, several factors could degrade it. Here, we investigated possible suboptimal optical experimental configurations that could lead to a degraded light collection when using Teflon or ESR reflectors coupled to the crystal with air or grease. We generated look-up tables with a mixture of air and grease and showed the effect of three possible sources of light loss: the presence of a small gap between the crystal and the reflector edges close to the photodetector face, the infiltration of grease in the crystal-reflector coupling, and the presence of inhomogeneities in the photodetector-crystal interface. The strongest effect is linked to the presence of a small gap of grease between the edges of the reflector material and the crystal (light loss of 10%-12% for 0.2mm gap). The optical grease infiltrating the crystal-reflector air coupling decreases the light output, depending on the infiltration's extent and the amount of grease infiltrated. Five percent of air in the crystal-photodetector coupling can cause a light output decrease of 2% to 4%. The individual and combined effect of these advanced models can explain the discrepancy of the relative light output obtained with ESR in simulations and experiments. With Teflon, the study indicates that the light output loss strongly depends on the reflectance deterioration caused by grease absorption. Our results indicate that when studying scintillation detector performance with different finishes, performing simulations in ideal coupling conditions can lead to light output overestimation. To perform an accurate light output comparison and ultimately have a reliable detector performance estimation, all potential sources of practical limitations must be carefully considered. To broadly enable high-fidelity modeling, we developed an interface for users to compute their own LUTs, using their surface, scintillator, and reflector characteristics.

Impact of train/test sample regimen on performance estimate stability of machine learning in cardiovascular imaging

As machine learning research in the field of cardiovascular imaging continues to grow, obtaining reliable model performance estimates is critical to develop reliable baselines and compare different algorithms. While the machine learning community has generally accepted methods such as k-fold stratified cross-validation (CV) to be more rigorous than single split validation, the standard research practice in medical fields is the use of single split validation techniques. This is especially concerning given the relatively small sample sizes of datasets used for cardiovascular imaging. We aim to examine how train-test split variation impacts the stability of machine learning (ML) model performance estimates in several validation techniques on two real-world cardiovascular imaging datasets: stratified split-sample validation (70/30 and 50/50 train-test splits), tenfold stratified CV, 10 × repeated tenfold stratified CV, bootstrapping (500 × repeated), and leave one out (LOO) validation. We demonstrate that split validation methods lead to the highest range in AUC and statistically significant differences in ROC curves, unlike the other aforementioned approaches. When building predictive models on relatively small data sets as is often the case in medical imaging, split-sample validation techniques can produce instability in performance estimates with variations in range over 0.15 in the AUC values, and thus any of the alternate validation methods are recommended.

Stability, Adaptability Analysis of Wheat Genotypes by AMMI with BLUP for Restricted Irrigated Multi Location Trials in Peninsular Zone of India

Highly significant effects of the environment (E), GxE interaction, and genotypes (G) had been observed by AMMI analysis for wheat genotypes evaluated under restricted irrigated timely sown multi-location trials in Peninsular zone of the country during 2018-19 and 2019-20 cropping seasons. Ranking of genotypes had changed with the number of interaction Principal Component Axes (IPCA’s) included for the calculation of Additive Main and Multiplicative Interaction (AMMI) based as well as Weighted Average of Absolute Scores (WAASB) stability measures. The Superiority indexes while assigning more weights to yield as compared to stability measures pointed out wheat genotypes MACS6695, HI1605, NIAW3170 & MACS6696 would maintain high yield and stable performance for the first year. Adaptability measures as per various averages expressed deviation from other measures and maintained right angle with MASV1 and stability measures. Moreover, the Superiority indexes as per various averages clustered in the same quadrant. Second-year of the study observed MP1358, NIAW3170, NIDW1149, MACS4087 wheat genotypes pointed by Superiority indexes. Adaptability measures as per arithmetic, geometric and harmonic means expressed strong bondage and grouped in a separate quadrant. This cluster maintained the right angle with stability measures and cluster of Superiority indexes as per various averages placed in a different quadrant. Superiority indexes would provide the reliable estimates of genotype performance in future studies in a biplot as considered all of significant IPCA’s.

Combining Radar and Optical Sensor Data to Measure Player Value in Baseball.

Evaluating a player’s talent level based on batted balls is one of the most important and difficult tasks facing baseball analysts. An array of sensors has been installed in Major League Baseball stadiums that capture seven terabytes of data during each game. These data increase interest among spectators, but also can be used to quantify the performances of players on the field. The weighted on base average cube model has been used to generate reliable estimates of batter performance using measured batted-ball parameters, but research has shown that running speed is also a determinant of batted-ball performance. In this work, we used machine learning methods to combine a three-dimensional batted-ball vector measured by Doppler radar with running speed measurements generated by stereoscopic optical sensors. We show that this process leads to an improved model for the batted-ball performances of players.

The effects of infill walls and basement shear walls on the seismic performance of existing damaged building

Many buildings have basement and some of these basements partially surrounded by shear walls. This partial surrounding causes torsional irregularity, which may increase the damage under the effect of a severe earthquake. A reliable seismic performance estimation is an important issue and there are different methods in the estimation of this performance such as nonlinear static procedure and time history analysis. Nonlinear static procedure does not give accurate results for irregular buildings, thus nonlinear time history analysis is a more convenient method in the estimation of seismic performance of a building, which has such irregularities. Additionally, recent researches showed that infill walls have significant effects on the seismic behavior of a building. In this study, a 6-story RC building with partial surrounded basement, which suffered damage in Kocaeli 1999 earthquake, was considered. The seismic performance of the building was estimated according to Turkish Seismic Code for Buildings (TSC) 2007 using nonlinear time history analysis. Buildings, which are partially and fully surrounded with shear walls in the basement, were modeled with and without infill walls and existing damage was tried to be estimated analytically.

Brace-to-frame connection modelling effects on seismic loss assessment of steel concentrically-braced frames

The main objective of this study is to investigate the implications of the explicit brace-to-frame gusset connection modelling on the seismic performance assessment of steel concentrically-braced frames (CBFs). Two key modelling alternatives are considered, namely through the simulation of CBFs with: i) a benchmark experimentally-validated model, in which the gusset plates are explicitly designed and simulated; and ii) a more simplified option, whereby diagonals are simulated through a centreline-to-centreline approach with fully-pinned boundary conditions. Anticipating the consequences of a less refined modelling approach, a simplified regression-based design process to Eurocode 8 for these gusset connections is proposed, therefore avoiding the need for an explicit connection design for seismic assessment purposes. A third modelling option, an identical approach to the benchmark but with the gusset dimensions being determined through the aforementioned regression-based design process, is also considered. The importance of the explicit modelling of the gusset plate connections to obtain reliable estimates of seismic performance is shown through the comparison of the seismic performance of various building archetypes, considering different metrics (e.g. lateral deformations, earthquake-induced collapse risk and economic losses). Furthermore, the validity of the proposed gusset-design methodology is discussed, by comparing the resulting gusset dimensions against those associated with a rigorous seismic design process, as well as by comparing the seismic performance assessed with numerical models incorporating explicitly-designed or simplified-designed gusset geometries.