Search Results (22)

I address the challenge of extracting reliable insights from large datasets using a simplified model that illustrates how hierarchical classification can distort outcomes. The model consists of discrete pixels labeled red, blue, or white. Red and blue indicate distinct properties, while white represents unclassified or ambiguous data. A macro-color is assigned only if one color holds a strict majority among the pixels. Otherwise, the aggregate is labeled white, reflecting uncertainty. This setup mimics a percolation threshold at fifty percent. Assuming that directly accessing the various proportions from the data of colors is infeasible, I implement a hierarchical coarse-graining procedure. Elements (first pixels, then aggregates) are recursively grouped and reclassified via local majority rules, ultimately producing a single super-aggregate for which the color represents the inferred macro-property of the collection of pixels as a whole. Analytical results supported by simulations show that the process introduces additional white aggregates beyond white pixels, which could be present initially; these arise from groups lacking a clear majority, requiring arbitrary symmetry-breaking decisions to attribute a color to them. While each local resolution may appear minor and inconsequential, their repetitions introduce a growing systematic bias. Even with complete data, unavoidable asymmetries in local rules are shown to skew outcomes. This study highlights a critical limitation of recursive data reduction. Insight extraction is shaped not only by data quality but also by how local ambiguity is handled, resulting in built-in biases. Thus, the related flaws are not due to the data but to structural choices made during local aggregations. Although based on a simple model, these findings expose a high likelihood of inherent flaws in widely used hierarchical classification techniques. Full article

Background: Parkinson’s disease (PD) is a progressive neurodegenerative condition that impairs motor and non-motor functions. Early and accurate diagnosis is critical for effective management and care. Leveraging machine learning (ML) techniques, this study aimed to develop a robust prediction system for PD using a stacked ensemble learning approach, addressing challenges such as imbalanced datasets and feature optimization. Methods: An open-access PD dataset comprising 22 vocal attributes and 195 instances from 31 subjects was utilized. To prevent data leakage, subjects were divided into training (22 subjects) and testing (9 subjects) groups, ensuring no subject appeared in both sets. Preprocessing included data cleaning and normalization via min–max scaling. The synthetic minority oversampling technique (SMOTE) was applied exclusively to the training set to address class imbalance. Feature selection techniques—forward search, gain ratio, and Kruskal–Wallis test—were employed using subject-wise cross-validation to identify significant attributes. The developed system combined support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN), and decision tree (DT) as base classifiers, with logistic regression (LR) as the meta-classifier in a stacked ensemble learning framework. Performance was evaluated using both recording-wise and subject-wise metrics to ensure clinical relevance. Results: The stacked ensemble learning model achieved realistic performance with a recording-wise accuracy of 84.7% and subject-wise accuracy of 77.8% on completely unseen subjects, outperforming individual classifiers including KNN (81.4%), RF (79.7%), and SVM (76.3%). Cross-validation within the training set showed 89.2% accuracy, with the performance difference highlighting the importance of proper validation methodology. Feature selection results showed that using the top 10 features ranked by gain ratio provided optimal balance between performance and clinical interpretability. The system’s methodological robustness was validated through rigorous subject-wise evaluation, demonstrating the critical impact of validation methodology on reported performance. Conclusions: By implementing subject-wise validation and preventing data leakage, this study demonstrates that proper validation yields substantially different (and more realistic) results compared to flawed recording-wise approaches. The findings underscore the critical importance of validation methodology in healthcare ML applications and provide a template for methodologically sound PD classification research. Future research should focus on validating the model with larger, multi-center datasets and implementing standardized validation protocols to enhance clinical applicability. Full article

This paper makes a critical evaluation of several of the research methods used to investigate the relationship between diet, health, and disease. The two widely used methods are randomized controlled trials (RCTs) and prospective cohort studies. RCTs are widely viewed as being more reliable than cohort studies and for that reason are placed higher in the research hierarchy. However, RCTs have inherent flaws and, consequently, they may generate findings that are less reliable than those from cohort studies. The text presents a discussion of the errors that may occur as a result of confounding. This refers to the correlation of the exposure and the outcome with other variables and can mask the true association or produce false associations. Another source of error is reverse causation, which is most commonly associated with cross-sectional studies. These studies do not allow researchers to determine the temporal sequence of lifestyle and other inputs together with health-related outcomes. As a result, it may be unclear which is cause and which is effect. This may also occur with cohort studies and can be illustrated by the inverse association between alcohol intake and coronary heart disease. Mechanistic research refers to the investigation of the intricate details of body functioning in health and disease and this research strategy is widely used in biomedical science. The evidence presented here makes the case that most of our information of practical value in the field of nutrition and disease has come from epidemiological research, including RCTs, whereas mechanistic research has been of minor value. Full article

Continuous-variable (CV) measurement-device-independent (MDI) quantum key distribution (QKD) can remove the feasible side-channel attacks on detectors based on the accurate Bell-state measurement (BSM), where an optical amplitude modulator (AM) plays a crucial role in managing the intensity of the transmitted light pulse. However, the AM-involved practical security has remained elusive as the operating frequency of the AM usually determines the actual secret key rate of the CV-MDI-QKD system. We find that an imperfect pulse generated from the AM at high speed can lead to a challenge to the practical security as a minor intensity change of the light pulse can bring about a potential information leakage. Taking advantage of this flaw, we suggest an attack strategy targeting the embedded AM in CV-MDI-QKD without sending the local oscillator (LO). This attack can damage the AM and thus decrease the estimated secret key rate of the system even when the orthogonal local LO (LLO) scheme is carried out. To assess the practical security risk resulting from the leaked information from the AM, we conduct numerical simulations to demonstrate the influence of the AM on the CVMDI-QKD system. Full article

To address the challenge of detecting tiny flaws in tile defect detection, a lightweight algorithm for identifying minor defects in tile images has been developed, referred to as CTDD-YOLO. Firstly, CAACSPELAN is proposed as the core component of the backbone network for extracting features of tile defects; secondly, full-dimensional dynamic convolution ODConv is introduced at the end of the backbone network to enhance the model’s ability to deal with tiny defects; next, a new neck network, CGRFPN, is proposed to improve the model’s ability to represent multi-scale features and enhance the model’s ability to recognize small targets in the context of large formats; finally, MPNWD is proposed to optimize the loss function to improve the model’s detection accuracy further. Experiments on the Ali Tianchi tile defect detection dataset show that the CTDD-YOLO model not only has a lower number of parameters than the original YOLOv8n but also improves the mAP by 7.2 percentage points, i.e., the proposed model can more accurately recognize and deal with minor surface defects of tiles and can significantly improve the detection effect while maintaining the light weight. Full article

This paper applies selected methodologies for the measurement of the environmental and economic sustainability of the Peggy Guggenheim Collection (PGC) in Venice with a view to assessing the PGC’s sustainability and commitment to implementing selected SDGs. To assess environmental sustainability, a life cycle assessment (LCA) has been carried out. The museum is conceptualized as a “firm” that produces several outputs and needs several inputs. The results provide the number of annual CO₂e (and other pollutants) emissions linked to the regular activity of the museum. The environmental cost (in EUR), linked to the impacts obtained from LCA, has been calculated. To assess economic sustainability, a survey and econometric methods were used to value services directly generated by the museum, and input/output methods were used to compute the direct and indirect impacts on the local economy. Nonetheless, PGC visitors (those who travel to Venice with the main objective of visiting the PGC) contribute to around 1.2%/1.4% of Venice’s GDP. The results from input–output tables show that, although the final demand generated by the PGC’s own activities amounted to about EUR 620 million in 2022, the economic benefits of the PGC beyond this final demand are significant and very positive due to carry-over effects. Specifically, the PGC leads to an increase in GDP of around EUR 1.200 million, with a multiplier of 1.9. In terms of employment, around 8200 jobs are associated with the presence of the PGC. The net public finance revenue also clearly benefits, with a net income of around EUR 150 million in 2022. Comparing both the environmental and economic impacts of the PGC, one can conclude that the annual activities performed by the museum are highly sustainable, with the economic pillar strongly offsetting the costs generated using natural resources. The creation of economic value, therefore, is generated in respect of environmental boundaries, even if some minor flaws can be highlighted. The connection between museums and sustainable development goals is highly recognized. The findings show the PGC’s commitment to achieving and implementing selected SDGs, including SDG 4, SDG 11, and SDG 16, by implementing actions and strategies that are aligned with these goals. Full article

Product quality inspection is a crucial element of industrial manufacturing, yet flaws such as blemishes and stains frequently emerge after the product is completed. Most research has utilized detection models and avoided segmenting networks due to the unequal distribution of faulty information. To overcome this challenge, this work presents a rapid segmentation-based technique for surface defect detection. The proposed model is based on a modified U-Net, which introduces a hybrid residual module (SAFM), combining an improved spatial attention mechanism and a feedforward neural network in place of the remaining downsampling layers, except for the first layer of downsampling in the encoder, and applies this residual module to the decoder structure. Dilated convolutions are also incorporated in the decoder to obtain more spatial information about the feature defects and to reduce the gradient vanishing problem of the model. An improved hybrid loss function with Dice and focal loss is introduced to alleviate the small defect segmentation problem. Comparative experiments were conducted on different segmentation-based inspection methods, revealing that the Dice coefficient (DSC) evaluated by the proposed approach is better than previous generic segmentation benchmarks on KolektorSDD, KolektorSDD2, and RSDD datasets, with fewer parameters and FLOPs. Additionally, the detection network displays higher precision in recognizing the characteristics of minor flaws. This paper proposes a practical and effective technique for anomaly segmentation in surface defect identification, delivering considerable improvements over previous methods. Full article

Original compositions based on iron micro-powders and an organic binder mixture were developed for the fabrication of sintered metallic elements with micro-powder injection molding (µPIM) and material extrusion additive manufacturing of metal powders (MEX). The binder formulation was thoroughly adjusted to exhibit rheological and thermal properties suitable for µPIM and MEX. The focus was set on adapting the proper binder composition to meet the requirements for injection/extrusion and, at the same time, to have comparable thermogravimetric characteristics for the thermal debinding and sintering process. A basic analysis of the forming process indicates that the pressure has a low influence on clogging, while the temperature of the material and mold/nozzle impacts the viscosity of the composition significantly. The influence of the Fe micro-powder content in the range of 45–60 vol.% was evaluated against the injection/extrusion process parameters and properties of sintered elements. Different debinding and sintering processes (chemical and thermal) were evaluated for the optimal properties of the final samples. The obtained sintered elements were of high quality and showed minor signs of binder-related flaws, with shrinkage in the range of 10–15% for both the injection-molded and 3D printed parts. These results suggest that, with minor modifications, compositions tailored for the PIM technique can be adapted for the additive manufacturing of metal parts, achieving comparable characteristics of the parts obtained for both forming methods. Full article

Green plums have produced significant economic benefits because of their nutritional and medicinal value. However, green plums are affected by factors such as plant diseases and insect pests during their growth, picking, transportation, and storage, which seriously affect the quality of green plums and their products, reducing their economic and nutritional value. At present, in the detection of green plum defects, some researchers have applied deep learning to identify their surface defects. However, the recognition rate is not high, the types of defects identified are singular, and the classification of green plum defects is not detailed enough. In the actual production process, green plums often have more than one defect, and the existing detection methods ignore minor defects. Therefore, this study used the vision transformer network model to identify all defects on the surfaces of green plums. The dataset was classified into multiple defects based on the four types of defects in green plums (scars, flaws, rain spots, and rot) and one type of feature (stem). After the permutation and combination of these defects, a total of 18 categories were obtained after the screening, combined with the actual situation. Based on the VIT model, a fine-grained defect detection link was added to the network for the analysis layer of the major defect hazard level and the detection of secondary defects. The improved network model has an average recognition accuracy rate of 96.21% for multiple defect detection of green plums, which is better than that of the VGG16 network, the Desnet121 network, the Resnet18 network, and the WideResNet50 network. Full article

Attempts to resolve the problem of evil often appeal to a greater good, according to which God’s permission of moral and natural evil is justified because (and just in case) the evil that is permitted is necessary for the realization of some greater good. In the extensive litany of greater good theodicies and defenses, the appeal to the greater good of an afterlife of infinite reward or pleasure has played a minor role in Christian thought but a more important role in Islamic thought. In a recent article, Seyyed Jaaber Mousavirad invites us to reconsider the greater good theodicy of compensation. He contends that not only are all evils justified in that God compensates the sufferer in an afterlife, but because the evils experienced produce some good, God has reason for bringing about or allowing evils in the first place. In what follows, I argue that this modified compensation theodicy is flawed in its premises, faces serious problems with its concept of justice, treats people as means only and not as intrinsically valuable, and ultimately fails to show that an afterlife compensation, along with some good produced here and now by evil, justify God bringing about or allowing evil. Full article

Dementia is a neurological condition that primarily affects older adults and there is still no cure or therapy available to cure it. The symptoms of dementia can appear as early as 10 years before the beginning of actual diagnosed dementia. Hence, machine learning (ML) researchers have presented several methods for early detection of dementia based on symptoms. However, these techniques suffer from two major flaws. The first issue is the bias of ML models caused by imbalanced classes in the dataset. Past research did not address this issue well and did not take preventative precautions. Different ML models were developed to illustrate this bias. To alleviate the problem of bias, we deployed a synthetic minority oversampling technique (SMOTE) to balance the training process of the proposed ML model. The second issue is the poor classification accuracy of ML models, which leads to a limited clinical significance. To improve dementia prediction accuracy, we proposed an intelligent learning system that is a hybrid of an autoencoder and adaptive boost model. The autoencoder is used to extract relevant features from the feature space and the Adaboost model is deployed for the classification of dementia by using an extracted subset of features. The hyperparameters of the Adaboost model are fine-tuned using a grid search algorithm. Experimental findings reveal that the suggested learning system outperforms eleven similar systems which were proposed in the literature. Furthermore, it was also observed that the proposed learning system improves the strength of the conventional Adaboost model by 9.8% and reduces its time complexity. Lastly, the proposed learning system achieved classification accuracy of 90.23%, sensitivity of 98.00% and specificity of 96.65%. Full article

The COVID-19 pandemic has caused unprecedented changes in the educational system, requiring students to continually switch between distance and in-person learning conditions. Recent studies have revealed that students experienced severe levels of anxiety in the COVID-19 period. Considering the close relationship that has always linked anxiety to mathematics, the present study explores the differences in the anxiety levels of students towards mathematics during distance or in-person school learning. During the second wave of COVID-19, 405 students, recruited from twelve middle schools of Catania province (Italy), completed an online version of the MeMa questionnaire, answering each item twice and imagining themselves to be, respectively, in distance and in-person learning conditions. The items explored generalized school anxiety, learning and evaluation mathematics anxiety, mental states, and the metacognitive awareness associated with mathematical tasks. The results showed a minor state of anxiety experienced during distance learning. However, the students who preferred to learn mathematics in person revealed less mathematics anxiety and better mental states and metacognitive awareness; the same results were found in those who reported higher math marks and who preferred scientific subjects. It seems that math anxiety is not one of the various flaws that are imputed to distance learning. Our findings encourage a reflection on possible interventions to reduce students’ anxiety by working on motivation and dysfunctional beliefs. Full article

Pediatric patients, particularly in neonatal and pediatric intensive care units (NICUs and PICUs), are typically at an increased risk of fatal decompensation. That being said, any delay in treatment or minor errors in medication dosage can overcomplicate patient health. Under such an environment, clinicians are expected to quickly and effectively comprehend large volumes of medical information to diagnose and develop a treatment plan for any baby. The integration of Artificial Intelligence (AI) into the clinical workflow can be a potential solution to safeguard pediatric patients and augment the quality of care. However, before making AI an integral part of pediatric care, it is essential to evaluate the technology from a human factors perspective, ensuring its readiness (technology readiness level) and ecological validity. Addressing AI accountability is also critical to safeguarding clinicians and improving AI acceptance in the clinical workflow. This article summarizes the application of AI in NICU/PICU and consecutively identifies the existing flaws in AI (from clinicians’ standpoint), and proposes related recommendations, which, if addressed, can improve AIs’ readiness for a real clinical environment. Full article

Generally, software developers make errors during the distributed software development process; therefore, software testing delay is a significant concern. Some of the software mistakes are minor, but others may be costly or harmful. Since things can still go wrong—individuals encounter mistakes from time to time—there is a need to double-check any software we develop in a distributed environment. The current global pandemic, COVID-19, has exacerbated and generated new challenges for IT organizations. Many issues exist for distributed software testing that prevent the achievement of successful and timely risk reduction when several of the mechanisms on which testing is based are disrupted. The environment surrounding COVID-19 is quickly evolving on a daily basis. Moreover, the pandemic has exposed or helped to develop flaws in production systems, which obstruct software test completion. Although some of these issues were urgent and needed to be evaluated early during the distributed software development process, this paper attempts to capture the details that represent the current pandemic reality in the software testing process. We used a Fuzzy TOPSIS-based multiple-criteria decision-making approach to evaluate the distributed software testing challenges. The statistical findings show that data insecurity is the biggest challenge for successful distributed software testing. Full article

In this paper, mechanical property tests of sandstone with multiple shapes of prefabricated fractures (single, T-shaped, and Y-shaped fractures) are carried out through the MTS815 rock mechanics testing machine and the Teledyne ISCO D-Series Pumps system. Considering the hydro-mechanical coupling effects, the experiments reveal the key thresholds, strength characteristics and deformation laws of multi-shape fractured sandstones during the progressive failure process. According to the elastic-plastic theory, the continuous damage theory and the statistical damage theory, a new damage model is constructed, which fully reflects the coupled effects among water, micro flaws and macroscopic prefabricated fractures. The crack closure stress σ_cc, crack initiation stress σ_ci and damage stress σ_cd of multi-shape fractured sandstone samples are determined by the proposed volumetric strain response method. In the range of 0–90°, the σ_cc and σ_ci of the multi-shape fractured sandstone samples are different, as well as the angles when the σ_cd and peak strength (σ_c) reach their peak values. The stress ratios (the σ_cc/σ_c, σ_ci/σ_c, and σ_cd/σ_c are collectively referred to as stress ratios) are hardly affected by the shape and inclination of the fractures inside the rock. According to strength analysis and deformation characteristics, the weakening effect of water has less of an influence on the strength than prefabricated fractures. The stress–strain curve obtained, based on the hydro-mechanical coupling test, is in good agreement with the theoretical curve generated by the damage constitutive model, verifying the rationality of the damage constitutive model. In addition, the fracture inclination only affects the numerical value of the total damage variable of multi-shape fractured sandstone samples, and has minor effects on its variation trend. Full article