Search Results (134)

Gesture recognition based on conventional machine learning is the main control approach for advanced prosthetic hand systems. Its primary limitation is the need for feature extraction, which must meet real-time control requirements. On the other hand, deep learning models could potentially overfit when trained on small datasets. For these reasons, we propose a hybrid Linear Discriminant Analysis–convolutional neural network (LDA-CNN) framework to improve the gesture recognition performance of sEMG-based prosthetic hand control systems. Within this framework, 1D-CNN filters are trained to generate latent representation that closely approximates Fisher’s (LDA’s) discriminant subspace, constructed from handcrafted features. Under the train-one-test-all evaluation scheme, our proposed hybrid framework consistently outperformed the 1D-CNN trained with cross-entropy loss only, showing improvements from 4% to 11% across two public datasets featuring hand gestures recorded under various limb positions and arm muscle contraction levels. Furthermore, our framework exhibited advantages in terms of induced spectral regularization, which led to a state-of-the-art recognition error of 22.79% with the extended 23 feature set when tested on the multi-limb position dataset. The main novelty of our hybrid framework is that it decouples feature extraction in regard to the inference time, enabling the future incorporation of a more extensive set of features, while keeping the inference computation time minimal. Full article

This research paper focuses on dimensionality reduction, which is a major subproblem in any data processing operation. Dimensionality reduction based on principal components is the most used methodology. Our paper examines three heuristics, namely Kaiser’s rule, the broken stick, and the conditional number rule, for selecting informative principal components when using principal component analysis to reduce high-dimensional data to lower dimensions. This study uses 22 classification datasets and three classifiers, namely Fisher’s discriminant classifier, logistic regression, and K nearest neighbors, to test the effectiveness of the three heuristics. The results show that there is no universal answer to the best intrinsic dimension, but the conditional number heuristic performs better, on average. This means that the conditional number heuristic is the best candidate for automatic data pre-processing. Full article

The growing volume and complexity of gene expression data necessitate biologically meaningful and statistically robust methods for feature selection to enhance the effectiveness of disease diagnosis systems. The present study addresses this challenge by proposing a pipeline that integrates RNA-seq data preprocessing, differential gene expression analysis, Gene Ontology (GO) enrichment, and ensemble-based machine learning. The pipeline employs the non-parametric Kruskal–Wallis test to identify differentially expressed genes, followed by dual enrichment analysis using both Fisher’s exact test and the Kolmogorov–Smirnov test across three GO categories: Biological Process (BP), Molecular Function (MF), and Cellular Component (CC). Genes associated with GO terms found significant by both tests were used to construct multiple gene subsets, including subsets based on individual categories, their union, and their intersection. Classification experiments using a random forest model, validated via 5-fold cross-validation, demonstrated that gene subsets derived from the CC category and the union of all categories achieved the highest accuracy and weighted F1-scores, exceeding 0.97 across 14 cancer types. In contrast, subsets derived from BP, MF, and especially their intersection exhibited lower performance. These results confirm the discriminative power of spatially localized gene annotations and underscore the value of integrating statistical and functional information into gene selection. The proposed approach improves the reliability of biomarker discovery and supports downstream analyses such as clustering and biclustering, providing a strong foundation for developing precise diagnostic tools in personalized medicine. Full article

The tropism of the Human Immunodeficiency Virus type 1 (HIV-1) is determined by the use of either or both chemokine coreceptors CCR5 (R5) and CXCR4 (X4) for entry into the target cell. The ability of HIV-1 to bind R5 or X4 is determined primarily by the third variable loop (V3) of the viral envelope glycoprotein gp120. HIV-1 strains of pandemic group M contain an antisense gene termed asp, which overlaps env outside the region encoding the V3 loop. We previously showed that the ASP protein localizes on the envelope of infectious HIV-1 virions, suggesting that it may play a role in viral entry. In this study, we first developed a statistical method to predict coreceptor tropism based on Fisher’s linear discriminant analysis. We obtained three linear discriminant functions able to predict coreceptor tropism with high accuracy (94.4%) when applied to a training dataset of V3 sequences of known tropism. Using these functions, we predicted the tropism in a dataset of HIV-1 strains containing a full-length asp gene. In the amino acid sequence of ASP proteins expressed from these asp genes, we identified five positions with substitutions significantly associated with viral tropism. Interestingly, we found that these substitutions correlate significantly with substitutions at six amino acid positions of the V3 loop domain associated with tropism. Altogether, our computational analyses identify ASP amino acid signatures coevolving with V3 and potentially affecting HIV-1 tropism, which can be validated through in vitro and in vivo experiments. Full article

This study proposes a novel assessment model to evaluate the occurrence potential of channelized debris flows (CDFs) in sedimentary rock regions of Central and Southern Taiwan, with a particular emphasis on the characteristics of hillslope debris flows (HDFs) within watersheds. CDFs are significantly related to the occurrence of HDFs in the upper reaches of watersheds, suggesting a high correlation between the potential for both phenomena. The study initially developed a hillslope debris flow (HDF) recognition model utilizing Fisher’s discriminant analysis, based on data from 40 HDF events and 40 landslide events. This model was subsequently applied to identify HDF units within channelized debris flow (CDF) watersheds. Subsequently, a CDF potential assessment model was constructed using data from 32 streams, which included 16 CDFs and 16 non-debris flow streams. Two key indicators emerged as the most effective: “Total WA(_>8)” and “number of WA(_>10).” These indicators achieved an accuracy rate exceeding 84%, significantly outperforming the official assessment model, which had an accuracy rate of 60%. The newly developed assessment models offer substantial improvements in predicting CDF occurrences, enhancing disaster preparedness and sustainable environment development. Full article

Fusarium wilt of banana is a soil-borne vascular disease caused by Fusarium oxysporum f. sp. cubense. The rapid and accurate detection of this disease is of great significance to controlling its spread. The research objective was to explore rapid banana Fusarium wilt latency and onset detection methods and establish a disease severity grading model. Visible/near-infrared spectroscopy analysis combined with machine learning methods were used for the rapid in vivo detection of banana Fusarium wilt. A portable visible/near-infrared spectrum acquisition system was constructed to collect the spectra data of banana Fusarium wilt leaves representing five different disease grades, totaling 106 leaf samples which were randomly divided into a training set with 80 samples and a test set with 26 samples. Different data preprocessing methods were utilized, and Fisher discriminant analysis (FDA), an extreme learning machine (ELM), and a one-dimensional convolutional neural network (1D-CNN) were used to establish the classification models of the disease grades. The classification accuracies of the FDA, ELM, and 1D-CNN models reached 0.891, 0.989, and 0.904, respectively. The results showed that the proposed visible/near infrared spectroscopy detection method could realize the detection of the incubation period of banana Fusarium wilt and the classification of the disease severity and could be a favorable tool for the field diagnosis of banana Fusarium wilt. Full article

Heywood cases and other improper solutions occur frequently in latent variable models, e.g., factor analysis, item response theory, latent class analysis, multilevel models, or structural equation models, all of which are models with response variables taken from an exponential family. They have important consequences for scoring with the latent variable model and are indicative of issues in a model, such as poor identification or model misspecification. In the context of the 2PL and 3PL models in IRT, they are more frequently known as Guttman items and are identified by having a discrimination parameter that is deemed excessively large. Other IRT models, such as the newer asymmetric item response theory (AsymIRT) or polytomous IRT models often have parameters that are not easy to interpret directly, so scanning parameter estimates are not necessarily indicative of the presence of problematic values. The graphical examination of the IRF can be useful but is necessarily subjective and highly dependent on choices of graphical defaults. We propose using the derivatives of the IRF, item Fisher information functions, and our proposed Item Fraction of Total Information (IFTI) decomposition metric to bypass the parameters, allowing for the more concrete and consistent identification of Heywood cases. We illustrate the approach by using empirical examples by using AsymIRT and nominal response models. Full article

Introduction: Leucine-rich alpha-2-glycoprotein 1(LRG-1) is a human protein that has shown potential usefulness as a biomarker for diagnosing pediatric acute appendicitis (PAA). This study aims to validate the diagnostic performance of serum LRG-1 in PAA. Material and Methods: This work is a subgroup analysis from BIDIAP (BIomarkers for DIagnosing Appendicitis in Pediatrics), a prospective single-center observational cohort, to validate serum LRG-1 as a diagnostic tool in PAA. This analysis included 200 patients, divided into three groups: (1) healthy patients undergoing major outpatient surgery (n = 56), (2) patients with non-surgical abdominal pain (n = 52), and (3) patients with a confirmed diagnosis of PAA (n = 92). Patients in group 3 were divided into complicated and uncomplicated PAA. In all patients, a serum sample was obtained during recruitment, and LRG-1 concentration was determined by Enzyme-Linked ImmunoSorbent Assay (ELISA). Comparative statistical analyses were performed using the Mann–Whitney U, Kruskal–Wallis, and Fisher’s exact tests. The area under the receiver operating characteristic curves (AUC) was calculated for all pertinent analyses. Results: Serum LRG-1 values, expressed as median (interquartile range) were 23,145 (18,246–27,453) ng/mL in group 1, 27,655 (21,151–38,795) ng/mL in group 2 and 40,409 (32,631–53,655) ng/mL in group 3 (p < 0.0001). Concerning the type of appendicitis, the serum LRG-1 values obtained were 38,686 (31,804–48,816) ng/mL in the uncomplicated PAA group and 51,857 (34,013–64,202) ng/mL in the complicated PAA group (p = 0.02). The area under the curve (AUC) obtained (group 2 vs. 3) was 0.75 (95% CI 0.67–0.84). For the discrimination between complicated and uncomplicated PAA, the AUC obtained was 0.66 (95% CI 0.52–0.79). Conclusions: This work establishes normative health ranges for serum LRG-1 values in the pediatric population and shows that serum LRG-1 could be a potentially helpful tool for diagnosing PAA in the future. Future prospective multicenter studies, with the parallel evaluation of urinary and salivary LRG-1, are necessary to assess the implementability of this molecule in actual clinical practice. Full article

Background: Over the years, it was noticed that patients with diabetes have reached an alarming number worldwide. Diabetes presents many complications, including diabetic kidney disease (DKD), which can be considered the leading cause of end-stage renal disease. Current biomarkers such as serum creatinine and albuminuria have limitations for early detection of DKD. Methods: In our study, we used UHPLC-QTOF-ESI+-MS techniques to quantify previously analyzed metabolites. Based on one-way ANOVA and Fisher’s LSD, untargeted analysis allowed the discrimination of six metabolites between subgroups P1 versus P2 and P3: tryptophan, kynurenic acid, taurine, l-acetylcarnitine, glycine, and tiglylglycine. Results: Our results showed several metabolites that exhibited significant differences among the patient groups and can be considered putative biomarkers in early DKD, including glycine and kynurenic acid in serum (p < 0.001) and tryptophan and tiglylglycine (p < 0.001) in urine. Conclusions: Although we identified metabolites as potential biomarkers in the present study, additional studies are needed to validate these results. Full article

Along with climate warming, extreme heat events have become more frequent, severe, and seriously threaten rice production. Precisely evaluating rice heat levels based on heat duration and a cumulative intensity index dominated by temperature and humidity is of great merit to effectively assess regional heat risk and minimize the deleterious impact of rice heat along the middle and lower reaches of the Yangtze River (MLRYR). This study quantified the response mechanism of daytime heat accumulation, night-time temperature, and relative humidity to disaster-causing intensity in three categories of single-season rice heat (dry, medium, and wet conditions) using Fisher discriminant analysis to obtain the Heat Comprehensive Intensity Index daily (HCIId). It is indicated that relative humidity exhibited a negative contribution under dry heat, i.e., heat disaster-causing intensity increased with decreasing relative humidity, with the opposite being true for medium and wet heat. The Kappa coefficient, combined with heat duration and cumulative HCIId, was implemented to determine classification thresholds for different disaster levels (mild, moderate, and severe) to construct heat evaluation levels. Afterwards, spatiotemporal changes in heat risk for single-season rice through the periods of 1986–2005, 2046–2065 and 2080–2099 under SSP2-4.5 and SSP5-8.5 were evaluated using climate scenario datasets and heat evaluation levels carefully constructed. Regional risk projection explicitly revealed that future risk would reach its maximum at booting and flowering, followed by the tillering stage, and its minimum at filling. The future heat risk for single-season rice significantly increased under SSP5-8.5 than SSP2-4.5 in MLRYR. The higher risk would be highlighted in eastern Hubei, eastern Hunan, most of Jiangxi, and northern Anhui. As time goes on, the heat risk for single-season rice in eastern Jiangsu and southern Zhejiang will progressively shift from low to mid-high by the end of the twenty-first century. Understanding the potential risk of heat exposure at different growth stages can help decision-makers guide the implementation of targeted measures to address climate change. The proposed methodology also provides the possibility of assessing other crops exposure to heat stress or other extreme events. Full article

Introduction: The study of Candida glabrata genes associated with fluconazole resistance, from a molecular perspective, increases the understanding of the phenomenon with a view to its clinical applicability. Objective: We sought to establish the predictive molecular profile of fluconazole resistance in Candida glabrata by analyzing the ERG11, ERG3, CgCDR1, and CgSNQ2 genes. Method: Expression was quantified using RT-qPCR. Metrics were obtained through molecular docking and Fisher discriminant functions. Additionally, a predictive classification was made against the susceptibility of C. glabrata to fluconazole. Results: The relative expression of the ERG3, CgCDR1, and CgSNQ2 genes was higher in the fluconazole-resistant strains than in the fluconazole-susceptible, dose-dependent strains. The gene with the highest relative expression in the fluconazole-exposed strains was CgCDR1, and in both the resistant and susceptible, dose-dependent strains exposed to fluconazole, this was also the case. The molecular docking model generated a median number of contacts between fluconazole and ERG11 that was lower than the median number of contacts between fluconazole and ERG3, -CgCDR1, and -CgSNQ2. The predicted classification through the multivariate model for fluconazole susceptibility achieved an accuracy of 73.5%. Conclusion: The resistant strains had significant expression levels of genes encoding efflux pumps and the ERG3 gene. Molecular analysis makes the identification of a low affinity between fluconazole and its pharmacological target possible, which may explain the lower intrinsic susceptibility of the fungus to fluconazole. Full article

Rapid and reliable identification of mineral species is a challenging but crucial task with promising application prospects in mineralogy, metallurgy, and geology. Spectroscopic techniques such as laser-induced breakdown spectroscopy (LIBS) and Raman spectroscopy (RS) efficiently capture the elemental composition and structural information of minerals, making them a potential tool for in situ and real-time analysis of minerals. This study introduces an integrated LIBS-RS system and the fusion of LIBS and RS spectra coupled with machine learning to classify six different types of natural mineral. In order to visualize the separability of different mineral species clearly, the spectral data were projected into low-dimensional space through t-distributed stochastic neighbor embedding (t-SNE). Additionally, the Fisher score (FS) was used to identify important variables that contribute to the data classification, and the corresponding chemical elements and molecular bonds were then interpreted. The between-minerals difference in the feature spectral intensity of LIBS and RS variables could also be observed. After the minerals spectra were pre-processed, the relationship between spectral intensity and the mineral category was modeled using machine learning methods, including partial least squares–discriminant analysis (PLS-DA) and kernel extreme learning machine (K-ELM). The results show that K-ELM and PLS-DA based on the fusion LIBS-RS data achieved the highest accuracy of 98.4%. These findings demonstrate the feasibility of the integrated LIBS-RS system combined with machine learning for the fast and reliable classification of minerals. Full article

This study introduces a novel method for classifying sets of images, called Riemannian geodesic discriminant analysis–minimum Riemannian mean distance (RGDA-MRMD). This method first converts image data into symmetric positive definite (SPD) matrices, which capture important features related to the variability within the data. These SPD matrices are then mapped onto simpler, flat spaces (tangent spaces) using a mathematical tool called the logarithm operator, which helps to reduce their complexity and dimensionality. Subsequently, regularized local Fisher discriminant analysis (RLFDA) is employed to refine these simplified data points on the tangent plane, focusing on local data structures to optimize the distances between the points and prevent overfitting. The optimized points are then transformed back into a complex, curved space (SPD manifold) using the exponential operator to enhance robustness. Finally, classification is performed using the minimum Riemannian mean distance (MRMD) algorithm, which assigns each data point to the class with the closest mean in the Riemannian space. Through experiments on the ETH-80 (Eidgenössische Technische Hochschule Zürich-80 object category), AFEW (acted facial expressions in the wild), and FPHA (first-person hand action) datasets, the proposed method demonstrates superior performance, with accuracy scores of 97.50%, 37.27%, and 88.47%, respectively. It outperforms all the comparison methods, effectively preserving the unique topological structure of the SPD matrices and significantly boosting image set classification accuracy. Full article

As industrial processes grow increasingly complex, fault identification becomes challenging, and even minor errors can significantly impact both productivity and system safety. Fault detection and diagnosis (FDD) has emerged as a crucial strategy for maintaining system reliability and safety through condition monitoring and abnormality recovery to manage this challenge. Statistical-based FDD methods that rely on large-scale process data and their features have been developed for detecting faults. This paper overviews recent investigations and developments in statistical-based FDD methods, focusing on probabilistic models. The theoretical background of these models is presented, including Bayesian learning and maximum likelihood. We then discuss various techniques and methodologies, e.g., probabilistic principal component analysis (PPCA), probabilistic partial least squares (PPLS), probabilistic independent component analysis (PICA), probabilistic canonical correlation analysis (PCCA), and probabilistic Fisher discriminant analysis (PFDA). Several test statistics are analyzed to evaluate the discussed methods. In industrial processes, these methods require complex matrix operation and cost computational load. Finally, we discuss the current challenges and future trends in FDD. Full article

Xylella Fastidiosa has been recently detected for the first time in southern Italy, representing a very dangerous phytobacterium capable of inducing severe diseases in many plants. In particular, the disease induced in olive trees is called olive quick decline syndrome (OQDS), which provokes the rapid desiccation and, ultimately, death of the infected plants. In this paper, we analyse about two thousands pixels of MODIS satellite evapotranspiration time series, covering infected and uninfected olive groves in southern Italy. Our aim is the identification of Xylella Fastidiosa-linked patterns in the statistical features of evapotranspiration data. The adopted methodology is the well-known Fisher–Shannon analysis that allows one to characterize the time dynamics of complex time series by means of two informational quantities, the Fisher information measure (FIM) and the Shannon entropy power (SEP). On average, the evapotranspiration of Xylella Fastidiosa-infected sites is characterized by a larger SEP and lower FIM compared to uninfected sites. The analysis of the receiver operating characteristic curve suggests that SEP and FIM can be considered binary classifiers with good discrimination performance that, moreover, improves if the yearly cycle, very likely linked with the meteo-climatic variability of the investigated areas, is removed from the data. Furthermore, it indicated that FIM exhibits superior effectiveness compared to SEP in discerning healthy and infected pixels. Full article