Human Subjective Judgment Research Articles

Fuzzy Logic-based Current Control Logic for Proportional Flow Control Valve System of Medical Ventilators

The critical control parameters in ventilators that provide life support to patients are the flow of air and oxygen. Ventilators typically use Proportional Flow Control Valves (PFCV) and flow sensors to achieve precise flow accuracy. Maintaining accurate flow from the PFCV is essential to minimize the thrust exerted on the patient's lungs. Fuzzy logic is particularly well-suited for this task, as it effectively captures the subjective human judgments often required in clinical settings. This study aimed to develop a fuzzy logic algorithm to control pressure support ventilation. The research describes the fuzzy logic algorithm and presents experimental studies that evaluated the flow accuracy of a commercially used PFCV in ventilators. Additionally, the reduction in output flow due to thermal effects was investigated. A correlation between the solenoid coil current (measured as voltage across the coil) and the actual output flow through the PFCV was experimentally established using a current sensing circuit. Based on the characteristic curves obtained from the test setup, a Fuzzy Logic system was developed to minimize the error in the PFCV's flow. The implementation of PFCV using this Fuzzy Logic-based current control approach demonstrated a significant reduction in error, from 10% to 3%, in airflow delivery across its operating range, with additional sample results discussed. This reduction in flow error is crucial for assessing the suitability of PFCVs in infant ventilators. The described methodology can be experimentally applied to any PFCV variant before its use in medical ventilator applications to enhance delivered flow accuracy.

The Construction of Fuzzy Prediction Model of Stock Price Rise and Fall Based on Machine Learning Technology

In recent years, the use of smart data analysis method to predict the stock price is financial technology; important issues in the field of finch. However, there are many technical indicators and human subjective factors will affect the stock price forecast, so we must effectively grasp the important influence indicators to improve the accuracy of stock price forecast. Therefore, this study uses four machine learning algorithms to predict and analyze the stock price fluctuation through the screening process of technical indicators, and then selects the important technical indicators. In addition, due to the uncertainty and fuzziness of the attributes of technical indicators and human subjective judgment, this study uses the fuzzy inference method to construct the fuzzy inference system to predict the rise and fall of stock price, and proposes the prediction method of the range of the rise and fall of stock price. Finally, this paper makes an empirical analysis on the stock price data of three companies. The results show that the accuracy of stock price forecast is more than 82.13%, and the average accuracy of stock price forecast is more than 83%. Therefore, the fuzzy inference prediction system proposed in this study not only has the theoretical basis, but also can effectively predict the trend and range of stock price, which has practical value and contribution to investors.

The Safe Catch: AI Protects Your Health from Formalin-Laced Fish

In Bangladesh, where fish is a staple food, ensuring its safety from formalin contamination poses a critical challenge due to its perishable nature. This study introduces an intelligent application employing digital image processing for the rapid and non-intrusive detection of formalin in fish. Leveraging image analysis of fish eyes, the system distinguishes between formalin and non-formalin treated fish. The proposed architecture, utilizing EfficientNet-B3 and VGG-16 models, achieved a 98.05% and 98% accuracy rate in training and validation on the dataset. This method offers a swift and accurate means of examination without damaging sample preparation, particularly beneficial in large-scale operations where manual inspection is impractical. Unlike human senses, digital image processing algorithms remain impartial, overcoming human biases and subjective judgments. Challenges persist, such as the diverse appearance of fish and external factors like varying illumination, which may impact the reliability and effectiveness of image processing programs for formalin detection. Nonetheless, this technology holds promise in addressing the pressing need for dependable and automated formalin detection in the fish supply chain, ensuring food safety and public health.

Development of web-based software for environmental impact assessment focusing on road traffic noise

In the Republic of Korea, Environmental Impact Assessment (EIAs) precedes development projects to predict and analyze potential environmental effects. Generally, EIA noise evaluations utilize 2D noise prediction equations and correction coefficients. This method, however, offers only a sectional noise evaluation and has limitations in complex environments with diverse noise sources. Moreover, the determination of various variables during the EIA process based on subjective human judgment raises concerns about the reliability of the results. Thus, this study aims to develop software accessible via a web environment for user-friendly EIA noise evaluations. This software supports integrated data management and generates a 3D noise prediction model for more precise and realistic analysis of noise impacts, specifically focusing on road-traffic noise at this stage of development. The 3D noise prediction model and noise map generated by the developed software have been validated against through comparison with the results of onsite noise measurements and commercial EIA software, SoundPLAN. This validation aimed to assess the practical utility of the application.

Spatial Image-Based Walkability Evaluation Using Regression Model

Spatial Image-Based Walkability Evaluation Using Regression Model

Dual-Level Viewpoint-Learning for Cross-Domain Vehicle Re-Identification

The definition of vehicle viewpoint annotations is ambiguous due to human subjective judgment, which makes the cross-domain vehicle re-identification methods unable to learn the viewpoint invariance features during source domain pre-training. This will further lead to cross-view misalignment in downstream target domain tasks. To solve the above challenges, this paper presents a dual-level viewpoint-learning framework that contains an angle invariance pre-training method and a meta-orientation adaptation learning strategy. The dual-level viewpoint-annotation proposal is first designed to concretely redefine the vehicle viewpoint from two aspects (i.e., angle-level and orientation-level). An angle invariance pre-training method is then proposed to preserve identity similarity and difference across the cross-view; this consists of a part-level pyramidal network and an angle bias metric loss. Under the supervision of angle bias metric loss, the part-level pyramidal network, as the backbone, learns the subtle differences of vehicles from different angle-level viewpoints. Finally, a meta-orientation adaptation learning strategy is designed to extend the generalization ability of the re-identification model to the unseen orientation-level viewpoints. Simultaneously, the proposed meta-learning strategy enforces meta-orientation training and meta-orientation testing according to the orientation-level viewpoints in the target domain. Extensive experiments on public vehicle re-identification datasets demonstrate that the proposed method combines the redefined dual-level viewpoint-information and significantly outperforms other state-of-the-art methods in alleviating viewpoint variations.

A machine vision approach with temporal fusion strategy for concrete vibration quality monitoring

In concrete construction, ensuring the quality of vibration is paramount for maintaining the strength, durability, and quality of structures. This study proposed a method for monitoring the vibration quality of vibrating robots to replace subjective human judgment. The method employed an improved EfficientNetV2 model to classify the vibration quality into four levels: unqualified, middle, qualified, and over-vibration. A temporal fusion strategy was introduced, employing time-domain probability fusion to enhance the accuracy and stability of the results. Additionally, image patching was applied to reduce computational complexity while preserving feature integrity. Experimental results demonstrate that the proposed method outperforms common mainstream models, achieving an accuracy of 96.47%, with a relatively small parameter size of 13.8 M. Compared to non-temporal fusion strategy, the accuracy is improved by 2.06%. This research has been successfully applied in practical engineering, providing a reliable means of quality assurance for concrete structures and demonstrating potential application prospects in the field of engineering construction.

Machine Vision-based Trajectory Tracking of Honey Bee Crawling Motion

With the maturity of machine vision technology in artificial intelligence and the rise of the concept of "smart agriculture", machine vision-based methods are widely interested in studying biological behavior, but the technology for acquiring biological behavior information is still unsound. In this paper, we propose a machine vision-based bee crawling trajectory tracking method, which automatically records the bee crawling trajectory in the video information and obtains the description of bee crawling behavior. The method establishes the description of bee crawling motion through bee target detection, bee crawling motion tracking and bee crawling motion prediction as a method to study bee crawling behavior through artificial intelligence techniques. The method has the advantages of high timeliness and low labor cost, and reduces the problems of traditional research through manual observation records and bioassays methods with high influence of human subjective judgment factors and high application costs.

Node search space reduction for optimal placement of pressure sensors in water distribution networks for leakage detection

This study presents a methodological framework for optimal placement of pressure sensors in Water Distribution Networks (WDNs) for leakage monitoring under uncertainty. Monte Carlo simulation is utilized to simulate leakages of different magnitudes at various nodes in the WDN taking into consideration background noise and minimum resolution of pressure sensors. A novel sensor preselection algorithm based on community detection and maximum entropy computation to reduce the search space of the pressure Sensor Placement Problem (SPP) is presented. The pressure SPP is formulated as a multi-objective optimization problem that seeks to maximize Joint Entropy, Coverage, and minimize Total Correlation. NSGA-II is used to solve the SPP and the solutions in the optimal Pareto front are ranked using a hybrid Entropy TOPSIS to eliminate potential bias and subjective human judgement in optimal sensor configuration implementation. The sensor preselection algorithm achieved a 67% reduction in the search space (possible sensor positions) of the case study, C-TOWN WDN, with only 2.78% reduction in coverage. The result of the pressure SPP indicates only 21 pressure sensors are needed to cover 95.45% of the WDN under study. Finally, the overall performance of the proposed methodological framework is presented and compared with other related works.

AI-luminating Artificial Intelligence in Inflammatory Bowel Diseases: A Narrative Review on the Role of AI in Endoscopy, Histology, and Imaging for IBD.

Endoscopy, histology, and cross-sectional imaging serve as fundamental pillars in the detection, monitoring, and prognostication of inflammatory bowel disease (IBD). However, interpretation of these studies often relies on subjective human judgment, which can lead to delays, intra- and interobserver variability, and potential diagnostic discrepancies. With the rising incidence of IBD globally coupled with the exponential digitization of these data, there is a growing demand for innovative approaches to streamline diagnosis and elevate clinical decision-making. In this context, artificial intelligence (AI) technologies emerge as a timely solution to address the evolving challenges in IBD. Early studies using deep learning and radiomics approaches for endoscopy, histology, and imaging in IBD have demonstrated promising results for using AI to detect, diagnose, characterize, phenotype, and prognosticate IBD. Nonetheless, the available literature has inherent limitations and knowledge gaps that need to be addressed before AI can transition into a mainstream clinical tool for IBD. To better understand the potential value of integrating AI in IBD, we review the available literature to summarize our current understanding and identify gaps in knowledge to inform future investigations.

Biomechanical comparison of two surgical methods for Hallux Valgus deformity: Exploring the use of artificial neural networks as a decision-making tool for orthopedists.

Hallux Valgus foot deformity affects gait performance. Common treatment options include distal oblique metatarsal osteotomy and chevron osteotomy. Nonetheless, the current process of selecting the appropriate osteotomy method poses potential biases and risks, due to its reliance on subjective human judgment and interpretation. The inherent variability among clinicians, the potential influence of individual clinical experiences, or inherent measurement limitations may contribute to inconsistent evaluations. To address this, incorporating objective tools like neural networks, renowned for effective classification and decision-making support, holds promise in identifying optimal surgical approaches. The objective of this cross-sectional study was twofold. Firstly, it aimed to investigate the feasibility of classifying patients based on the type of surgery. Secondly, it sought to explore the development of a decision-making tool to assist orthopedists in selecting the optimal surgical approach. To achieve this, gait parameters of twenty-three women with moderate to severe Hallux Valgus were analyzed. These patients underwent either distal oblique metatarsal osteotomy or chevron osteotomy. The parameters exhibiting differences in preoperative and postoperative values were identified through various statistical tests such as normalization, Shapiro-Wilk, non-parametric Wilcoxon, Student t, and paired difference tests. Two artificial neural networks were constructed for patient classification based on the type of surgery and to simulate an optimal surgery type considering postoperative walking speed. The results of the analysis demonstrated a strong correlation between surgery type and postoperative gait parameters, with the first neural network achieving a remarkable 100% accuracy in classification. Additionally, cases were identified where there was a mismatch with the surgeon's decision. Our findings highlight the potential of artificial neural networks as a complementary tool for surgeons in making informed decisions. Addressing the study's limitations, future research may investigate a wider range of orthopedic procedures, examine additional gait parameters and use more diverse and extensive datasets to enhance statistical robustness.

Signature Verification System Using SSIM In Image Processing

The verification of signatures is an essential function in several domains, including financial, legal, and administrative processes. Thanks to advancements in image processing, automatic signature verification methods have become more popular. Using structural similarities and image analysis, the proposed research offers a novel approach to signature verification. To compare and assess signatures, it uses the SSIM index. The procedure begins with pre-processing the signature pictures to improve their quality and eliminate any artifacts or noise that may have been obtained from Adobe's stock library. Then, the structural similarity between the reference signature and the input signature is calculated. The perceptual resemblance of two images is measured using structure, contrast, and brightness. The goal of the proposed research is to use this measure to record the signature's structural features and spot changes or deviations. The SSIM value that comes out of the comparison is checked against a threshold that has already been set. To validate an input signature, the calculated similarity must be greater than a certain threshold. The document is marked as suspicious or possibly falsified if it does not comply. Experimental results have shown that the method is effective in differentiating between authentic and counterfeit signatures. By doing away with the need for subjective human judgment and physical examination, this technology provides a reliable and unbiased way to authenticate signatures. Increased automation and trust in signature authentication systems are possible because to the proposed method's encouraging results in accurately differentiating genuine signatures from fakes.

3D-PSSIM: Projective Structural Similarity for 3D Mesh Quality Assessment Robust to Topological Irregularities.

Despite acceleration in the use of 3D meshes, it is difficult to find effective mesh quality assessment algorithms that can produce predictions highly correlated with human subjective opinions. Defining mesh quality features is challenging due to the irregular topology of meshes, which are defined on vertices and triangles. To address this, we propose a novel 3D projective structural similarity index ( 3D- PSSIM) for meshes that is robust to differences in mesh topology. We address topological differences between meshes by introducing multi-view and multi-layer projections that can densely represent the mesh textures and geometrical shapes irrespective of mesh topology. It also addresses occlusion problems that occur during projection. We propose visual sensitivity weights that capture the perceptual sensitivity to the degree of mesh surface curvature. 3D- PSSIM computes perceptual quality predictions by aggregating quality-aware features that are computed in multiple projective spaces onto the mesh domain, rather than on 2D spaces. This allows 3D- PSSIM to determine which parts of a mesh surface are distorted by geometric or color impairments. Experimental results show that 3D- PSSIM can predict mesh quality with high correlation against human subjective judgments, across the presence of noise, even when there are large topological differences, outperforming existing mesh quality assessment models.

Explainability and Evaluation of Vision Transformers: An In-Depth Experimental Study

In the era of artificial intelligence (AI), the deployment of intelligent systems for autonomous decision making has surged across diverse fields. However, the widespread adoption of AI technology is hindered by the risks associated with ceding control to autonomous systems, particularly in critical domains. Explainable artificial intelligence (XAI) has emerged as a critical subdomain fostering human understanding and trust. It addresses the opacity of complex models such as vision transformers (ViTs), which have gained prominence lately. With the expanding landscape of XAI methods, selecting the most effective method remains an open question, due to the lack of a ground-truth label for explainability. To avoid subjective human judgment, numerous metrics have been developed, with each aiming to fulfill certain properties required for a valid explanation. This study conducts a detailed evaluation of various XAI methods applied to the ViT architecture, thereby exploring metrics criteria like faithfulness, coherence, robustness, and complexity. We especially study the metric convergence, correlation, discriminative power, and inference time of both XAI methods and metrics. Contrary to expectations, the metrics of each criterion reveal minimal convergence and correlation. This study not only challenges the conventional practice of metric-based ranking of XAI methods but also underscores the dependence of explanations on the experimental environment, thereby presenting crucial considerations for the future development and adoption of XAI methods in real-world applications.

Innovative Multi-Criteria Decision-Making Approach for Supplier Evaluation: Combining TLF, Fuzzy BWM, and VIKOR

When confronted with underperforming suppliers, the need to evaluate and improve supplier performance becomes apparent. However, the inherent inaccuracies in information introduce complexity, especially when subjective human judgment is involved in the supplier evaluation process. Associated with such problem, this study presents a novel methodology for supplier performance evaluation in the crumb rubber industry, integrating the Taguchi Loss Function (TLF), fuzzy Best-Worst Method (BWM), and VIKOR technique in group decision-making environment. Aimed at addressing the challenges in industries with variable supplier quality and performance, such as the crumb rubber industry in Indonesia, the methodology was empirically tested to demonstrate its practical utility. The process involved identifying evaluation criteria through literature review tailored to the needs of decision makers (DMs), applying TLF to quantify losses from supplier performance deviations, using fuzzy BWM to determine criteria weights based on the DMs judgment, and employing the VIKOR technique for comprehensive supplier ranking. The findings underscore the methodology's effectiveness in enhancing decision-making, offering a unified metric that accommodates diverse criteria and balances precise data with subjective assessments. This approach simplifies the evaluation process, particularly in situations with conflicting interests among decision-makers. Demonstrating its practical application in the crumb rubber industry, the study highlights the methodology's potential for broader industrial applicability. Future research could explore comparative analyses with other analytical methods, further establishing the methodology's robustness and adaptability in different management contexts.

Precise control mode for concrete vibration time based on attention-enhanced machine vision

In concrete construction, vibration is an essential process used to ensure the structural soundness and long-term durability of concrete structures. Automating concrete vibration monitoring enables far more precise control compared to subjective human judgments. In this paper, a control mode for concrete vibration time used on the vibrating robot is proposed. The control mode utilizes YOLOv8 model to recognize the best-vibrating position and eliminate the rebar part of concrete surface image and employs attention-enhanced squeeze-and-excitation neural network to regress the vibrating completion degree. After deployment to the embedded system of vibrating robot, experimental results demonstrate superior performance of the proposed method over state-of-the-art algorithms with the highest recognition rate and maximum R2 value. The control mode enhances vibration quality while liberating workers from manual tasks. A commercial mall construction project validates the practicality of the presented control model.

Pengolahan Citra Untuk Mengetahui Derajat Korosi Pada Insulator Dengan Metode Decision Tree

One of the important parts that ensures the availability of electrical energy supplies which takes place from the central power plant and then through the transmission line to distribution to the load is the insulator. However, over time, insulators can experience anomalies such as corrosion. Insulator corrosion can be caused by many environmental factors, including pollution, humidity, high temperatures and exposure to chemicals. Previously, maintenance personnel usually identified the level of corrosion on insulators visually by using their knowledge and experience through routine inspection activities to check the condition of the insulators. This technique, however, takes a long time and is often subjective. Therefore, in this research, the way to detect the degree of corrosion on insulators is by using image processing and using active contour segmentation to determine the degree of corrosion on insulators and using decision trees as a method to create a classification model for the degree of corrosion. This can help reduce reliance on subjective human judgment, increase the effectiveness of the recognition process, and optimize overall isolator maintenance efforts. The application created in this research has a user interface that makes it easier to process images of corrosion insulators. The front page, practice page, and test page comprise this interface. Meanwhile, the results of classification using a decision tree for the low category are in the ratio x1 < 0.082104, the medium category is in the ratio 0.082104 =< x1 < 0.15061, and the high category is in the ratio x1 >= 0.15061 with an accuracy result obtained of 93.33% for test image data and recall and precision values ​​obtained 100% results.

NIMG-36. IMAGE LOCALIZATION OF CLINICAL SAMPLES IN BRAIN CANCER: A PROSPECTIVE PROTOCOL

Abstract It has been widely acknowledged that limited access to the human brain, along with the importance of its structures, presents significant challenges in obtaining brain tumor tissue. This poses difficulties for clinical decision-making, as healthcare professionals often have to rely solely on imaging to address complex problems. There is currently no definitive method to make these decisions other than relying on informed yet subjective human judgment. That said, imaging abnormalities are a result of a presence of particular tumor-related biology. Because of this, and the heterogeneity frequently observed in brain tumors, we have developed an image-localized biopsy protocol that has been employed for research purposes at Mayo Clinic Arizona for over five years. This multidisciplinary approach involves the collaborative efforts from neurosurgeons, neuro-oncologists, scrub-techs, nurses, radiologists, as well as multiple research labs, all working within a well-established pipeline. The implementation of this protocol has allowed us to obtain image-localized biopsies during surgical procedures, enabling the correlation of local imaging findings with corresponding tissue features through the utilization of machine learning models. The primary objective of this endeavor is to enhance our understanding of the underlying biological mechanisms of brain tumors based on patients' imaging data. This protocol has been successful for our research purposes, and with minor adjustments, it can be readily adopted by any hospital in order to collect more data about these incredibly heterogeneous diseases. While the use of neuronavigation during surgery has become commonplace in many medical institutions, localization of tissue samples is not. Utilizing pre-existing infrastructure in the operating room is a small burden with big potential positive outcomes. Sample locations can be useful to clinicians during tumor board discussions, to better assess disease heterogeneity. These minor refinements in the surgical workflow have the capacity to provide clinicians with additional clinically-relevant data without disrupting patient care.

Neural and behavioral signatures of the multidimensionality of manipulable object processing

Understanding how we recognize objects requires unravelling the variables that govern the way we think about objects and the neural organization of object representations. A tenable hypothesis is that the organization of object knowledge follows key object-related dimensions. Here, we explored, behaviorally and neurally, the multidimensionality of object processing. We focused on within-domain object information as a proxy for the decisions we typically engage in our daily lives – e.g., identifying a hammer in the context of other tools. We extracted object-related dimensions from subjective human judgments on a set of manipulable objects. We show that the extracted dimensions are cognitively interpretable and relevant – i.e., participants are able to consistently label them, and these dimensions can guide object categorization; and are important for the neural organization of knowledge – i.e., they predict neural signals elicited by manipulable objects. This shows that multidimensionality is a hallmark of the organization of manipulable object knowledge.

An Austrian Critique of Passive Investing

This article reviews numerous arguments Austrian scholars have made against the efficient market hypothesis and asserts that these arguments are sufficient grounds for dismissal of many of the theoretical justifications which have been advanced in favor of passive investing, and in favor of the traditional value-weighted index fund in particular. In the absence of theoretical justification, empirical test results are the primary evidence to substantiate passive investing’s claim to superiority relative to alternative investment strategies—results which, this article contends, are insufficient grounds for acceptance of the claim. Finally, it is argued that the much-emphasized contradistinction between active and passive investing is lacking in substance. The passive portfolio’s construction is the product of subjective human judgment just as the actively managed portfolio is; passive investing merely involves a different kind of judgment.