An Improved k-Nearest Neighbor Algorithm for Recognition and Classification of Thyroid Nodules.

OBJECTIVES: To complete the task of automatic recognition and classification of thyroid nodules and solve the problem of high classification error rates when the samples are imbalanced. METHODS: An improved k-nearest neighbor (KNN) algorithm is proposed and a method for automatic thyroid nodule classification based on the improved KNN algorithm is established. In the improved KNN algorithm, we consider not only the number of class labels for various classes of data in KNNs, but also the corresponding weights. And we use the Minkowski distance measure instead of the Euclidean distance measure. RESULTS: A total of 508 ultrasound images of thyroid nodules, including 415 benign nodules and 93 malignant nodules, were used in the paper. Experimental results show the improved KNN has 0.872549 accuracy, 0.867347 precision, 1 recall, and 0.928962 F1-score. At the same time, we also considered the influence of different distance weights, the value of k, different distance measures on the classification results. CONCLUSIONS: A comparison result shows that our method has a better performance than the traditional KNN and other classical machine learning methods.

Accurate modeling of DNA conformational flexibility by a multivariate Ising model.

The sequence-dependent structure and deformability of DNA play a major role for binding of proteins and regulation of gene expression. So far, most efforts to model DNA flexibility are based on unimodal harmonic stiffness models at base-pair resolution. However, multimodal behavior due to distinct conformational substates also contributes significantly to the conformational flexibility of DNA. Moreover, these local substates are correlated to their nearest-neighbor substates. A description for DNA elasticity which includes both multimodality and nearest-neighbor coupling has remained a challenge, which we solve by combining our multivariate harmonic approximation with an Ising model for the substates. In a series of applications to DNA fluctuations and protein-DNA complexes, we demonstrate substantial improvements over the unimodal stiffness model. Furthermore, our multivariate Ising model reveals a mechanical destabilization for adenine (A)-tracts to undergo nucleosome formation. Our approach offers a wide range of applications to determine sequence-dependent deformation energies of DNA and to investigate indirect readout contributions to protein-DNA recognition.

Spatial Pattern Analysis using Closest Events (SPACE)-A Nearest Neighbor Point Pattern Analysis Framework for Assessing Spatial Relationships from Digital Images.

The quantitative description of biological structures is a valuable yet difficult task in the life sciences. This is commonly accomplished by imaging samples using fluorescence microscopy and analyzing resulting images using Pearson's correlation or Manders' co-occurrence intensity-based colocalization paradigms. Though conceptually and computationally simple, these approaches are critically flawed due to their reliance on signal overlap, sensitivity to cursory signal qualities, and inability to differentiate true and incidental colocalization. Point pattern analysis provides a framework for quantitative characterization of spatial relationships between spatial patterns using the distances between observations rather than their overlap, thus overcoming these issues. Here we introduce an image analysis tool called Spatial Pattern Analysis using Closest Events (SPACE) that leverages nearest neighbor-based point pattern analysis to characterize the spatial relationship of fluorescence microscopy signals from image data. The utility of SPACE is demonstrated by assessing the spatial association between mRNA and cell nuclei from confocal images of cardiac myocytes. Additionally, we use synthetic and empirical images to characterize the sensitivity of SPACE to image segmentation parameters and cursory image qualities such as signal abundance and image resolution. Ultimately, SPACE delivers performance superior to traditional colocalization methods and offers a valuable addition to the microscopist's toolbox.

Machine learning-driven SERS analysis platform for rapid and accurate detection of precancerous lesions of gastric cancer.

A novel approach is proposed leveraging surface-enhanced Raman spectroscopy (SERS) combined with machine learning (ML) techniques, principal component analysis (PCA)-centroid displacement-based nearest neighbor (CDNN). This label-free approach can identify slight abnormalities between SERS spectra of gastric lesions at different stages, offering a promising avenue for detection and prevention of precancerous lesion of gastric cancer (PLGC). The agaric-shaped nanoarray substrate was prepared using gas-liquid interface self-assembly and reactive ion etching (RIE) technology to measure SERS spectra of serum from mice model with gastric lesions at different stages, and then a SERS spectral recognition model was trained and constructed using the PCA-CDNN algorithm. The results showed that the agaric-shaped nanoarray substrate has good uniformity, stability, cleanliness, and SERS enhancement effect. The trained PCA-CDNN model not only found the most important features of PLGC, but also achieved satisfactory classification results with accuracy, area under curve (AUC), sensitivity, and specificity up to 100%. This demonstrated the enormous potential of this analysis platform in the diagnosis of PLGC.

Fault Diagnosis of a Multistage Centrifugal Pump Using Explanatory Ratio Linear Discriminant Analysis.

This study introduces an innovative approach for fault diagnosis of a multistage centrifugal pump (MCP) using explanatory ratio (ER) linear discriminant analysis (LDA). Initially, the method addresses the challenge of background noise and interference in vibration signals by identifying a fault-sensitive frequency band (FSFB). From the FSFB, raw hybrid statistical features are extracted in time, frequency, and time-frequency domains, forming a comprehensive feature pool. Recognizing that not all features adequately represent MCP conditions and can reduce classification accuracy, we propose a novel ER-LDA method. ER-LDA evaluates feature importance by calculating the explanatory ratio between interclass distance and intraclass scatteredness, facilitating the selection of discriminative features through LDA. This fusion of ER-based feature assessment and LDA yields the novel ER-LDA technique. The resulting selective feature set is then passed into a k-nearest neighbor (K-NN) algorithm for condition classification, distinguishing between normal, mechanical seal hole, mechanical seal scratch, and impeller defect states of the MCP. The proposed technique surpasses current cutting-edge techniques in fault classification.

Underwater Single-Photon 3D Reconstruction Algorithm Based on K-Nearest Neighbor.

The high sensitivity and picosecond time resolution of single-photon avalanche diodes (SPADs) can improve the operational range and imaging accuracy of underwater detection systems. When an underwater SPAD imaging system is used to detect targets, backward-scattering caused by particles in water often results in the poor quality of the reconstructed underwater image. Although methods such as simple pixel accumulation have been proven to be effective for time-photon histogram reconstruction, they perform unsatisfactorily in a highly scattering environment. Therefore, new reconstruction methods are necessary for underwater SPAD detection to obtain high-resolution images. In this paper, we propose an algorithm that reconstructs high-resolution depth profiles of underwater targets from a time-photon histogram by employing the K-nearest neighbor (KNN) to classify multiple targets and the background. The results contribute to the performance of pixel accumulation and depth estimation algorithms such as pixel cross-correlation and ManiPoP. We use public experimental data sets and underwater simulation data to verify the effectiveness of the proposed algorithm. The results of our algorithm show that the root mean square errors (RMSEs) of land targets and simulated underwater targets are reduced by 57.12% and 23.45%, respectively, achieving high-resolution single-photon depth profile reconstruction.

Rolling Bearing Fault Diagnosis Based on SABO-VMD and WMH-KNN.

To improve the performance of roller bearing fault diagnosis, this paper proposes an algorithm based on subtraction average-based optimizer (SABO), variational mode decomposition (VMD), and weighted Manhattan-K nearest neighbor (WMH-KNN). Initially, the SABO algorithm uses a composite objective function, including permutation entropy and mutual information entropy, to optimize the input parameters of VMD. Subsequently, the optimized VMD is used to decompose the signal to obtain the optimal decomposition characteristics and the corresponding intrinsic mode function (IMF). Finally, the weighted Manhattan function (WMH) is used to enhance the classification distance of the KNN algorithm, and WMH-KNN is used for fault diagnosis based on the optimized IMF features. The performance of the SABO-VMD and WMH-KNN models is verified through two experimental cases and compared with traditional methods. The results show that the accuracy of motor-bearing fault diagnosis is significantly improved, reaching 97.22% in Dataset 1, 98.33% in Dataset 2, and 99.2% in Dataset 3. Compared with traditional methods, the proposed method significantly reduces the false positive rate.

Vehicle Position Detection Based on Machine Learning Algorithms in Dynamic Wireless Charging.

Dynamic wireless charging (DWC) has emerged as a viable approach to mitigate range anxiety by ensuring continuous and uninterrupted charging for electric vehicles in motion. DWC systems rely on the length of the transmitter, which can be categorized into long-track transmitters and segmented coil arrays. The segmented coil array, favored for its heightened efficiency and reduced electromagnetic interference, stands out as the preferred option. However, in such DWC systems, the need arises to detect the vehicle's position, specifically to activate the transmitter coils aligned with the receiver pad and de-energize uncoupled transmitter coils. This paper introduces various machine learning algorithms for precise vehicle position determination, accommodating diverse ground clearances of electric vehicles and various speeds. Through testing eight different machine learning algorithms and comparing the results, the random forest algorithm emerged as superior, displaying the lowest error in predicting the actual position.

Radio Signal Modulation Recognition Method Based on Hybrid Feature and Ensemble Learning: For Radar and Jamming Signals.

The detection performance of radar is significantly impaired by active jamming and mutual interference from other radars. This paper proposes a radio signal modulation recognition method to accurately recognize these signals, which helps in the jamming cancellation decisions. Based on the ensemble learning stacking algorithm improved by meta-feature enhancement, the proposed method adopts random forests, K-nearest neighbors, and Gaussian naive Bayes as the base-learners, with logistic regression serving as the meta-learner. It takes the multi-domain features of signals as input, which include time-domain features including fuzzy entropy, slope entropy, and Hjorth parameters; frequency-domain features, including spectral entropy; and fractal-domain features, including fractal dimension. The simulation experiment, including seven common signal types of radar and active jamming, was performed for the effectiveness validation and performance evaluation. Results proved the proposed method's performance superiority to other classification methods, as well as its ability to meet the requirements of low signal-to-noise ratio and few-shot learning.

Automatic classification of seizure and seizure-free EEG signals based on phase space reconstruction features.

Epilepsy is a type of brain disorder triggered by an abrupt electrical imbalance of neuronal networks. An electroencephalogram (EEG) is a diagnostic tool to capture the underlying brain mechanisms and detect seizure onset in epileptic patients. To detect seizures, neurologists need to manually monitor EEG recordings for long periods, which is challenging and susceptible to errors depending on expertise and experience. Therefore, automatic identification of seizure and seizure-free EEG signals becomes essential. This study introduces a method based on the features extracted from the phase space reconstruction for classifying seizure and seizure-free EEG signals. The computed features are derived from the elliptical area and interquartile range of the Euclidean distance by varying percentage values of data points ranging from 50 to 100%. We consider two public datasets and evaluate these features in each EEG epoch that includes the healthy, interictal, preictal, and ictal stages of epileptic subjects, utilizing the K-nearest neighbor classifier for classification. Results show that the features have higher values during the seizure than the seizure-free EEG signals and healthy subjects. Furthermore, the proposed features can effectively discriminate seizure EEG signals from the seizure-free and normal subjects with 100% accuracy, sensitivity, and specificity in both datasets. Likewise, the classification between the preictal stage and seizure EEG signals attains 98% accuracy. Overall, the reconstructed phase space features significantly enhance the accuracy of detecting epileptic EEG signals compared with existing methods. This advancement holds great potential in assisting neurologists in swiftly and accurately diagnosing epileptic seizures from EEG signals.

Streamflow Prediction Using Complex Networks.

The reliable prediction of streamflow is crucial for various water resources, environmental, and ecosystem applications. The current study employs a complex networks-based approach for the prediction of streamflow. The approach consists of three major steps: (1) the formation of a network using streamflow time series; (2) the calculation of the clustering coefficient (CC) as a network measure; and (3) the use of a clustering coefficient-based nearest neighbor search procedure for streamflow prediction. For network construction, each timestep is considered as a node and the existence of link between any node pair is identified based on the difference (distance) between the streamflow values of the nodes. Different distance threshold values are used to identify the critical distance threshold to form the network. The complex networks-based approach is implemented for the prediction of daily streamflow at 142 stations in the contiguous United States. The prediction accuracy is quantified using three statistical measures: correlation coefficient (R), normalized root mean square error (NRMSE), and Nash-Sutcliffe efficiency (NSE). The influence of the number of neighbors on the prediction accuracy is also investigated. The results, obtained with the critical distance threshold, reveal that the clustering coefficients for the 142 stations range from 0.799 to 0.999. Overall, the prediction approach yields reasonably good results for all 142 stations, with R values ranging from 0.05 to 0.99, NRMSE values ranging from 0.1 to 12.3, and the NSE values ranging from -0.89 to 0.99. An attempt is also made to examine the relationship between prediction accuracy and the catchment characteristics/streamflow statistical properties (drainage area, mean flow, coefficient of variation of flow). The results suggest that the prediction accuracy does not have much of a relationship with the drainage area and the mean streamflow values, but with the coefficient of variation of flow. The outcomes from this study are certainly promising regarding the application of complex networks-based concepts for the prediction of streamflow (and other hydrologic) time series.

Gene regulation network inference using k-nearest neighbor-based mutual information estimation: revisiting an old DREAM.

BACKGROUND: A cell exhibits a variety of responses to internal and external cues. These responses are possible, in part, due to the presence of an elaborate gene regulatory network (GRN) in every single cell. In the past 20 years, many groups worked on reconstructing the topological structure of GRNs from large-scale gene expression data using a variety of inference algorithms. Insights gained about participating players in GRNs may ultimately lead to therapeutic benefits. Mutual information (MI) is a widely used metric within this inference/reconstruction pipeline as it can detect any correlation (linear and non-linear) between any number of variables (n-dimensions). However, the use of MI with continuous data (for example, normalized fluorescence intensity measurement of gene expression levels) is sensitive to data size, correlation strength and underlying distributions, and often requires laborious and, at times, ad hoc optimization. RESULTS: In this work, we first show that estimating MI of a bi- and tri-variate Gaussian distribution using k-nearest neighbor (kNN) MI estimation results in significant error reduction as compared to commonly used methods based on fixed binning. Second, we demonstrate that implementing the MI-based kNN Kraskov-Stoögbauer-Grassberger (KSG) algorithm leads to a significant improvement in GRN reconstruction for popular inference algorithms, such as Context Likelihood of Relatedness (CLR). Finally, through extensive in-silico benchmarking we show that a new inference algorithm CMIA (Conditional Mutual Information Augmentation), inspired by CLR, in combination with the KSG-MI estimator, outperforms commonly used methods. CONCLUSIONS: Using three canonical datasets containing 15 synthetic networks, the newly developed method for GRN reconstruction-which combines CMIA, and the KSG-MI estimator-achieves an improvement of 20-35% in precision-recall measures over the current gold standard in the field. This new method will enable researchers to discover new gene interactions or better choose gene candidates for experimental validations.

Novel HLA associations with outcomes of Mycobacterium tuberculosis exposure and sarcoidosis in individuals of African ancestry using nearest-neighbor feature selection.

Tuberculosis and sarcoidosis are inflammatory diseases characterized by granulomas that may occur in any organ but are often found in the lung. The panoply of classical human leukocyte antigen (HLA) alleles associated with occurrence and/or severity of both diseases varies considerably across studies. This heterogeneity of results, due to variation in factors like ancestry and disease subphenotype, as well as the use of simple modeling strategies to elucidate likely complex relationships, has made conclusions about underlying commonalities difficult. Here we perform HLA association analyses in individuals of African ancestry, using a greater resolution to include subphenotypes of disease and employing more comprehensive analytical techniques. Using a novel application of nearest-neighbor feature selection to score allelic importance, we investigated HLA allele association with Mycobacterium tuberculosis exposure outcomes in the first analysis of both latent Mycobacterium tuberculosis infection and active disease compared with those who, despite long-term exposure to active index cases, have neither positive diagnostic tests nor display clinical symptoms. We also compared persistent to resolved sarcoidosis. This led to the identification of novel HLA associations and evidence of main effects and interaction effects. We found strikingly similar main effects and interaction effects at HLA-DRB1, -DQB1, and -DPB1 in those resistant to tuberculosis (either latent or active) and persistent sarcoidosis.

Machine learning classification reveals robust morphometric biomarker of glial and neuronal arbors.

Neurons and glia are the two main cell classes in the nervous systems of most animals. Although functionally distinct, neurons and glia are both characterized by multiple branching arbors stemming from the cell bodies. Glial processes are generally known to form smaller trees than neuronal dendrites. However, the full extent of morphological differences between neurons and glia in multiple species and brain regions has not yet been characterized, nor is it known whether these cells can be reliably distinguished based on geometric features alone. Here, we show that multiple supervised learning algorithms deployed on a large database of morphological reconstructions can systematically classify neuronal and glial arbors with nearly perfect accuracy and precision. Moreover, we report multiple morphometric properties, both size related and size independent, that differ substantially between these cell types. In particular, we newly identify an individual morphometric measurement, Average Branch Euclidean Length that can robustly separate neurons from glia across multiple animal models, a broad diversity of experimental conditions, and anatomical areas, with the notable exception of the cerebellum. We discuss the practical utility and physiological interpretation of this discovery.

iSMNN: batch effect correction for single-cell RNA-seq data via iterative supervised mutual nearest neighbor refinement.

Batch effect correction is an essential step in the integrative analysis of multiple single-cell RNA-sequencing (scRNA-seq) data. One state-of-the-art strategy for batch effect correction is via unsupervised or supervised detection of mutual nearest neighbors (MNNs). However, both types of methods only detect MNNs across batches of uncorrected data, where the large batch effects may affect the MNN search. To address this issue, we presented a batch effect correction approach via iterative supervised MNN (iSMNN) refinement across data after correction. Our benchmarking on both simulation and real datasets showed the advantages of the iterative refinement of MNNs on the performance of correction. Compared to popular alternative methods, our iSMNN is able to better mix the cells of the same cell type across batches. In addition, iSMNN can also facilitate the identification of differentially expressed genes (DEGs) that are relevant to the biological function of certain cell types. These results indicated that iSMNN will be a valuable method for integrating multiple scRNA-seq datasets that can facilitate biological and medical studies at single-cell level.

SMNN: batch effect correction for single-cell RNA-seq data via supervised mutual nearest neighbor detection.

Batch effect correction has been recognized to be indispensable when integrating single-cell RNA sequencing (scRNA-seq) data from multiple batches. State-of-the-art methods ignore single-cell cluster label information, but such information can improve the effectiveness of batch effect correction, particularly under realistic scenarios where biological differences are not orthogonal to batch effects. To address this issue, we propose SMNN for batch effect correction of scRNA-seq data via supervised mutual nearest neighbor detection. Our extensive evaluations in simulated and real datasets show that SMNN provides improved merging within the corresponding cell types across batches, leading to reduced differentiation across batches over MNN, Seurat v3 and LIGER. Furthermore, SMNN retains more cell-type-specific features, partially manifested by differentially expressed genes identified between cell types after SMNN correction being biologically more relevant, with precision improving by up to 841.0%.

Multivariate functional data clustering using adaptive density peak detection.

Clustering for multivariate functional data is a challenging problem since the data are represented by a set of curves and functions belonging to an infinite-dimensional space. In this article, we propose a novel clustering method for multivariate functional data using an adaptive density peak detection technique. It is a quick cluster center identification algorithm based on the two measures of each functional data observation: the functional density estimate and the distance to the closest observation with a higher functional density. We suggest two types of functional density estimators for multivariate functional data. The first one is a functional k $$ k $$ -nearest neighbor density estimator based on (a) an L2 distance between raw functional curves, or (b) a semimetric of multivariate functional principal components. The second one is a k $$ k $$ -nearest neighbor density estimator based on multivariate functional principal scores. Our clustering method is computationally fast since it does not need an iterative process. The flexibility and advantages of the method are examined by comparing it with other existing clustering methods in simulation studies. A user-friendly R package FADPclust is developed for public use. Finally, our method is applied to a real case study in lung cancer research.

Nearest-neighbor parameters for predicting DNA duplex stability in diverse molecular crowding conditions.

The intracellular environment is crowded and heterogeneous. Although the thermodynamic stability of nucleic acid duplexes is predictable in dilute solutions, methods of predicting such stability under specific intracellular conditions are not yet available. We recently showed that the nearest-neighbor model for self-complementary DNA is valid under molecular crowding condition of 40% polyethylene glycol with an average molecular weight of 200 (PEG 200) in 100 mM NaCl. Here, we determined nearest-neighbor parameters for DNA duplex formation under the same crowding condition to predict the thermodynamics of DNA duplexes in the intracellular environment. Preferential hydration of the nucleotides was found to be the key factor for nearest-neighbor parameters in the crowding condition. The determined parameters were shown to predict the thermodynamic parameters (∆H°, ∆S°, and ∆G°37) and melting temperatures (Tm) of the DNA duplexes in the crowding condition with significant accuracy. Moreover, we proposed a general method for predicting the stability of short DNA duplexes in different cosolutes based on the relationship between duplex stability and the water activity of the cosolute solution. The method described herein would be valuable for investigating biological processes that occur under specific intracellular crowded conditions and for the application of DNA-based biotechnologies in crowded environments.

An Advanced Rider-Cornering-Assistance System for PTW Vehicles Developed Using ML KNN Method.

The dynamic behavior of a Powered Two-Wheeler (PTW) is much more complicated than that of a car, which is due to the strong coupling between the longitudinal and lateral dynamics produced by the large roll angles. This makes the analysis of the dynamics, and therefore the design and synthesis of the controller, particularly complex and difficult. In relation to assistance in dangerous situations, several recent manuscripts have suggested devices with limitations of cornering velocity by proposing restrictive models. However, these models can lead to repulsion by the users of PTW vehicles, significantly limiting vehicle performance. In the present work, the authors developed an Advanced Rider-cornering Assistance System (ARAS) based on the skills learned by riders running across curvilinear trajectories using Artificial Intelligence (AI) and Neural Network (NN) techniques. New algorithms that allow the value of velocity to be estimated by prediction accuracy of up to 99.06% were developed using the K-Nearest Neighbor (KNN) Machine Learning (ML) technique.

A Second-Order Method for Removing Mixed Noise from Remote Sensing Images.

Remote sensing image denoising is of great significance for the subsequent use and research of images. Gaussian noise and salt-and-pepper noise are prevalent noises in images. Contemporary denoising algorithms often exhibit limitations when addressing such mixed noise scenarios, manifesting in suboptimal denoising outcomes and the potential blurring of image edges subsequent to the denoising process. To address the above problems, a second-order removal method for mixed noise in remote sensing images was proposed. In the first stage of the method, dilated convolution was introduced into the DnCNN (denoising convolutional neural network) network framework to increase the receptive field of the network, so that more feature information could be extracted from remote sensing images. Meanwhile, a DropoutLayer was introduced after the deep convolution layer to build the noise reduction model to prevent the network from overfitting and to simplify the training difficulty, and then the model was used to perform the preliminary noise reduction on the images. To further improve the image quality of the preliminary denoising results, effectively remove the salt-and-pepper noise in the mixed noise, and preserve more image edge details and texture features, the proposed method employed a second stage on the basis of adaptive median filtering. In this second stage, the median value in the original filter window median was replaced by the nearest neighbor pixel weighted median, so that the preliminary noise reduction result was subjected to secondary processing, and the final denoising result of the mixed noise of the remote sensing image was obtained. In order to verify the feasibility and effectiveness of the algorithm, the remote sensing image denoising experiments and denoised image edge detection experiments were carried out in this paper. When the experimental results are analyzed through subjective visual assessment, images denoised using the proposed method exhibit clearer and more natural details, and they effectively retain edge and texture features. In terms of objective evaluation, the performance of different denoising algorithms is compared using metrics such as mean square error (MSE), peak signal-to-noise ratio (PSNR), and mean structural similarity index (MSSIM). The experimental outcomes indicate that the proposed method for denoising mixed noise in remote sensing images outperforms traditional denoising techniques, achieving a clearer image restoration effect.