Machine Learning Dimensionality Reduction Research Articles

E.4 Machine learning based patient classification to predict neurological deterioration in mild Degenerative Cervical Myelopathy

Background: Degenerative Cervical Myelopathy (DCM) is the functional derangement of the spinal cord because of compression from degenerate tissues. Typical neurological symptoms of DCM include gait imbalance and upper extremity paresthesia. While it is thought that greater spinal cord compression leads to increased neurological deterioration, our clinical experience suggests a more complex mechanism involving spinal canal diameter (SCD). Methods: 124 MRI scans from 59 non-operative DCM patients underwent manual scoring of cord compression and SCD measurements. Unsupervised machine learning dimensionality reduction techniques and k-means clustering were used to establish patient groups. These patient groups underwent manual inspection of common compression patterns and SCD similarities to define their unique risk criteria. Results: We found that compression pattern is unimportant at SCD extremes (≤14.5 mm or >15.75 mm). Otherwise, stenosis with clear signs of cord compression at two disc levels and stenosis without clear signs of cord compression at two disc levels result in a relatively higher and lower likelihood of deterioration, respectively. We elucidated five patient groups with unique associated risks for neurological deterioration, according to both SCD range and their cord compression pattern. Conclusions: The specific combination of narrow SCD with focal cord compression increases the likelihood of neurological deterioration in non-operative patients with DCM.

GBRUN: A Gradient Search-based Binary Runge Kutta Optimizer for Feature Selection

<p>Feature selection (FS) is a pre-processing technique for data dimensionality reduction in machine learning and data mining algorithms. FS technique reduces the number of features and improves the model generalization ability. This study presents a Gradient Search-based Binary Runge Kutta Optimizer (GBRUN) for solving the FS problem of high-dimensional. First, the proposed method converts the continuous Runge Kutta optimizer (RUN) into a binary version through S-, V-, and U-shaped transfer functions. Second, a gradient search method is introduced to improve the exploration capability of the algorithm. Five standard datasets provided by Arizona State University’s Data Mining and Machine Learning Lab were selected to verify the performance of the GBRUN algorithm. The experimental results show that GBRUN has better performance than other advanced algorithms regarding classification accuracy and the number of selected features. Moreover, the GBRUN algorithm is also combined with EfficientNet in this manuscript, using the GBRUN algorithm to select the features extracted by EfficientNet. The results show that the V-shaped (GBRUN-V) and U-shaped (GBRUN-U) algorithms have better performance than other algorithms.</p> <p> </p>

Enhancing Outlier Detection and Dimensionality Reduction in Machine Learning for Extreme Value Analysis

An outlier is generally defined as an observation that significantly deviates from the rest of the considered compliance. In the realm of ultramodern machine literacy, where analysis of multidimensional datasets is current, improving data quality is imperative for economists aiming to gain robust results. Numerous machine learning algorithms are sensitive to both the range and distribution of trait values within the input data. The presence of outliers in the input data can distort and misguide the training process of machine literacy algorithms, leading to prolonged training times, less precise models, and eventually inferior issues. Prior to the construction of prophetic models using training data, outliers can introduce incorrect representations, thereby impacting interpretations of the collected data. Traditional outlier discovery methods frequently concentrate on banning the tails of distributions and overlooking the data generation processes specific to individual datasets. Colorful styles are useful for detecting different types of outliers in high-dimensional datasets from two distinct perspectives: relating the devious aspects of a data object and setting devious data objects within a dataset. N-dimensional data poses a significant challenge within the realm of machine literacy, contributing to the diversity of challenges faced in the field.

5G Enabled Moving Robot Captured Image Encryption with Principal Component Analysis Method

Estimating the captured image of moving robots is very difficult. These images are vital in analyzing earth's surface objects for many applications like studying environmental conditions, Land use and Land Cover changes, and change detection studies of worldwide change. Multispectral robot-captured images have a massive amount of low-resolution data, which is lost due to a lack of capture efficiency due to artificial and atmospheric reasons. The image transformation is required in a 5G network with effective transmission by reducing noise, inconsistent lighting, and low resolution, degrading image quality. In this paper, the authors proposed the machine learning dimensionality reduction technique i.e. Principle Component Analysis (PCA) and which is used for metastasizing the 5 G-enabled moving robot captured image to enrich the image's visual perception to analyze the exact information of global or local data. The encryption algorithm implanted for data reduction and transmission over the 5G network gives sophisticated results compared with other standard methods. This proposed algorithm gives better performance in developing data reduction, network convergence speed, reduces the training time for object classification, and improves accuracy for multispectral moving robot-captured images by the support of 5G network.

Unipolar and Bipolar Depression Detection and Classification Based on Actigraphic Registration of Motor Activity Using Machine Learning and Uniform Manifold Approximation and Projection Methods.

Modern technology frequently uses wearable sensors to monitor many aspects of human behavior. Since continuous records of heart rate and activity levels are typically gathered, the data generated by these devices have a lot of promise beyond counting the number of daily steps or calories expended. Due to the patient's inability to obtain the necessary information to understand their conditions and detect illness, such as depression, objectively, methods for evaluating various mental disorders, such as the Montgomery-Asberg depression rating scale (MADRS) and observations, currently require a significant amount of effort on the part of specialists. In this study, a novel dataset was provided, comprising sensor data gathered from depressed patients. The dataset included 32 healthy controls and 23 unipolar and bipolar depressive patients with motor activity recordings. Along with the sensor data collected over several days of continuous measurement for each patient, some demographic information was also offered. The result of the experiment showed that less than 70 of the 100 epochs of the model's training were completed. The Cohen Kappa score did not even pass 0.1 in the validation set, due to an imbalance in the class distribution, whereas in the second experiment, the majority of scores peaked in about 20 epochs, but because training continued during each epoch, it took much longer for the loss to decline before it fell below 0.1. In the second experiment, the model soon reached an accuracy of 0.991, which is as expected given the outcome of the UMAP dimensionality reduction. In the last experiment, UMAP and neural networks worked together to produce the best outcomes. They used a variety of machine learning classification algorithms, including the nearest neighbors, linear kernel SVM, Gaussian process, and random forest. This paper used the UMAP unsupervised machine learning dimensionality reduction without the neural network and showed a slightly lower score (QDA). By considering the ratings of the patient's depressive symptoms that were completed by medical specialists, it is possible to better understand the relationship between depression and motor activity.

A Tutorial on Principal Component Analysis for Dimensionality Reduction in Machine Learning

A Tutorial on Principal Component Analysis for Dimensionality Reduction in Machine Learning

A Tutorial on Principal Component Analysis for Dimensionality Reduction in Machine Learning

A Tutorial on Principal Component Analysis for Dimensionality Reduction in Machine Learning

Searching for the extra-tidal stars of globular clusters using high-dimensional analysis and a core particle spray code

ABSTRACT Three-body interactions can eject stars from the core of a globular cluster, causing them to enter the Galactic halo as extra-tidal stars. While finding extra-tidal stars is imperative for understanding cluster evolution, connecting isolated extra-tidal field stars back to their birth cluster is extremely difficult. In this work, we present a new methodology consisting of high-dimensional data analysis and a particle spray code to identify extra-tidal stars of any Galactic globular cluster using M3 as a case study. Using the t-Stochastic Neighbour Embedding and Uniform Manifold Approximation and Projection machine learning dimensionality reduction algorithms, we first identify a set of 103 extra-tidal candidates in the APOGEE DR17 data catalogue with chemical abundances similar to M3 stars. To confirm each candidate’s extra-tidal nature, we introduce corespray – a new python-based three-body particle spray code that simulates extra-tidal stars for any Galactic globular cluster. Using Gaia EDR3 proper motions and APOGEE DR17 radial velocities, we apply multivariate Gaussian modelling and an extreme deconvolution to identify the extra-tidal candidates that are more likely to be associated with a distribution of corespray-simulated M3 extra-tidal stars than the field. Through these methods, we identify 10 new high-probability extra-tidal stars produced via three-body interactions in M3. We also explore whether any of our extra-tidal candidates are consistent with being ejected from M3 through different dynamical processes. Future applications of corespray will yield better understandings of core dynamics, star formation histories, and binary fractions in globular clusters.

Dimensionality reduction in machine learning for nonadiabatic molecular dynamics: Effectiveness of elemental sublattices in lead halide perovskites.

Supervised machine learning (ML) and unsupervised ML have been performed on descriptors generated from nonadiabatic (NA) molecular dynamics (MD) trajectories representing non-radiative charge recombination in CsPbI3, a promising solar cell and optoelectronic material. Descriptors generated from every third atom of the iodine sublattice alone are sufficient for a satisfactory prediction of the bandgap and NA coupling for the use in the NA-MD simulation of nonradiative charge recombination, which has a strong influence on material performance. Surprisingly, descriptors based on the cesium sublattice perform better than those of the lead sublattice, even though Cs does not contribute to the relevant wavefunctions, while Pb forms the conduction band and contributes to the valence band. Simplification of the ML models of the NA-MD Hamiltonian achieved by the present analysis helps to overcome the high computational cost of NA-MD through ML and increase the applicability of NA-MD simulations.

Eye movement anomalies as a source of diagnostic information in decision process analysis.

As deviations from what is expected, anomalies are typically seen as an obstruction to making good predictions or an impulse to revise the predictive framework. Here, we consider a different possibility-that anomalies, particularly those related to cognitive processing, may be a valuable source of diagnostic information. More specifically, we hypothesize that the extent to which the prechoice information search has been atypical (anomalous) can be used to reverse-infer important latent features of the decision process. For instance, based on atypicalities in how juries examine courtroom evidence, can we infer if they were biased by media reports; or if financial traders browsed public stock market data in a way sufficiently unusual to indicate access to insider information? In our preregistered experiment, participants viewed expert opinions about financial stocks, before deciding whether to invest and get paid according to the subsequent market return, adjusted by an independent random amount. We used eye-tracking and machine-learning dimensionality reduction and anomaly detection techniques to measure the extent to which eye-movements while viewing opinions were idiosyncratic/anomalous. We found that nudging participants by disclosing the return adjustment value beforehand (thus giving them "privileged information") made their patterns of subsequently viewing opinions more idiosyncratic. With those idiosyncrasies as potential markers of top-down attentional control, we demonstrated a reverse-inference of motivation and prior knowledge from attention, predicting if people were nudged in a particular direction based on how idiosyncratically they then searched for information. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Unique Clusters of Sudden Unexpected Infant Death (SUID) Descendants Revealed Through Machine Learning Epidemiological Analyses

Sudden unexpected infant death (SUID) rates have substantially declined due to public health prevention efforts as well as mechanistic research over the past thirty years. Following an extensive medico-legal investigation, the decedent under one year of age is assigned as the cause of death: sudden infant death syndrome (SIDS), accidental suffocation and strangulation, or other ill-defined or unspecified cause of mortality. However, the extent to which these diseases manifest with an underlying neuropathological mechanism is highly controversial due to the heterogeneity of findings. In addition, substantial disparities are present when comparing SUID rates amongst race/ethnicity in the United States (US). The intersection of an underlying unknown biological abnormality and health disparities in the role of SUID is not well understood. Cluster analysis within heterogenous diseases has been highly instructive in neuropathology by identifying unique patient subtypes. Using maternal and infantile characteristics, we sought to identify unique groups of SUID decedents indicating distinct risk factors and neurodevelopmental causes. Using the US Period Linked Birth/Infant Mortality Files from 1990-2017 (excluding 1992-1994), we performed unsupervised machine learning dimensionality reduction techniques. Following identification of unique groups, we analyzed SUID rates in these clusters at the state level. We were able to identify three groups of SIDS decedents, each with a unique mean age at death when adjusting for length of gestation and performed similar analysis with the remaining SUID causes. We found that the implementation of clinical guidelines has had a distinct impact in decreasing the SUID rate in the unique clusters across geographical regions, however not all are experiencing a rapid reduction. The unexpected death of a previously seemingly healthy infant places a hard emotional burden on families. Through identifying unique risk factors and developmental timepoints of distinct SUID clusters, our findings can assist with public health infant mortality prevention policies and aid basic mechanistic research.

Stochastic Mutual Information Gradient Estimation for Dimensionality Reduction Networks.

Feature ranking and selection is a widely used approach in various applications of supervised dimensionality reduction in discriminative machine learning. Nevertheless there exists significant evidence on feature ranking and selection algorithms based on any criterion leading to potentially sub-optimal solutions for class separability. In that regard, we introduce emerging information theoretic feature transformation protocols as an end-to-end neural network training approach. We present a dimensionality reduction network (MMINet) training procedure based on the stochastic estimate of the mutual information gradient. The network projects high-dimensional features onto an output feature space where lower dimensional representations of features carry maximum mutual information with their associated class labels. Furthermore, we formulate the training objective to be estimated non-parametrically with no distributional assumptions. We experimentally evaluate our method with applications to high-dimensional biological data sets, and relate it to conventional feature selection algorithms to form a special case of our approach.

Utilizing machine learning dimensionality reduction for risk stratification of chest pain patients in the emergency department

BackgroundChest pain is among the most common presenting complaints in the emergency department (ED). Swift and accurate risk stratification of chest pain patients in the ED may improve patient outcomes and reduce unnecessary costs. Traditional logistic regression with stepwise variable selection has been used to build risk prediction models for ED chest pain patients. In this study, we aimed to investigate if machine learning dimensionality reduction methods can improve performance in deriving risk stratification models.MethodsA retrospective analysis was conducted on the data of patients > 20 years old who presented to the ED of Singapore General Hospital with chest pain between September 2010 and July 2015. Variables used included demographics, medical history, laboratory findings, heart rate variability (HRV), and heart rate n-variability (HRnV) parameters calculated from five to six-minute electrocardiograms (ECGs). The primary outcome was 30-day major adverse cardiac events (MACE), which included death, acute myocardial infarction, and revascularization within 30 days of ED presentation. We used eight machine learning dimensionality reduction methods and logistic regression to create different prediction models. We further excluded cardiac troponin from candidate variables and derived a separate set of models to evaluate the performance of models without using laboratory tests. Receiver operating characteristic (ROC) and calibration analysis was used to compare model performance.ResultsSeven hundred ninety-five patients were included in the analysis, of which 247 (31%) met the primary outcome of 30-day MACE. Patients with MACE were older and more likely to be male. All eight dimensionality reduction methods achieved comparable performance with the traditional stepwise variable selection; The multidimensional scaling algorithm performed the best with an area under the curve of 0.901. All prediction models generated in this study outperformed several existing clinical scores in ROC analysis.ConclusionsDimensionality reduction models showed marginal value in improving the prediction of 30-day MACE for ED chest pain patients. Moreover, they are black box models, making them difficult to explain and interpret in clinical practice.

Sleep time and efficiency in patients undergoing laboratory-based polysomnography.

Sleep quality in patients studied with laboratory-based polysomnography may differ from sleep quality in patients studied at home but remains clinically relevant and important to describe. We assessed objective sleep quality and explored factors associated with poor sleep in patients undergoing laboratory-based polysomnography. We reviewed diagnostic polysomnography studies from a 10-year period at a single sleep center. Total sleep time (TST) and sleep efficiency (SE) were assessed as markers of sleep quality. Poor sleep was defined as TST ≤ 4 hours or SE ≤ 50%. Multivariable analysis was performed to determine associations between objective sleep quality as an outcome and multiple candidate predictors including age, sex, race, body mass index, comorbidities, severity of obstructive sleep apnea, and central nervous system medications. Among 4957 patients (age 53 ± 15 years), average TST and median SE were 5.8 hours and 79%, respectively. There were 556 (11%) and 406 (8%) patients who had poor sleep based on TST and SE, respectively. In multivariable analysis, those who were older (per 10 years: 1.48 [1.34, 1.63]), male (1.38 [1.14,1.68]), and had severe obstructive sleep apnea (1.76 [1.28, 2.43]) were more likely to have short sleep. Antidepressant use was associated with lower odds of short sleep (0.77 [0.59,1.00]). Older age (per 10 years: 1.48 [1.34, 1.62]), male sex (1.34 [1.07,1.68]), and severe obstructive sleep apnea (2.16 [1.47, 3.21]) were associated with higher odds of poor SE. We describe TST and SE from a single sleep center cohort. Multiple demographic characteristics were associated with poor objective sleep in patients during laboratory-based polysomnography. Harrison EI, Roth RH, Lobo JM, et al. Sleep time and efficiency in patients undergoing laboratory-based polysomnography. J Clin Sleep Med. 2021;17(8):1591-1598.

Analysis of bath motion in MM-SQC dynamics via dimensionality reduction approach: Principal component analysis.

The system-plus-bath model is an important tool to understand the nonadiabatic dynamics of large molecular systems. Understanding the collective motion of a large number of bath modes is essential for revealing their key roles in the overall dynamics. Here, we applied principal component analysis (PCA) to investigate the bath motion in the basis of a large dataset generated from the symmetrical quasi-classical dynamics method based on the Meyer-Miller mapping Hamiltonian nonadiabatic dynamics for the excited-state energy transfer in the Frenkel-exciton model. The PCA method clearly elucidated that two types of bath modes, which either display strong vibronic coupling or have frequencies close to that of the electronic transition, are important to the nonadiabatic dynamics. These observations were fully consistent with the physical insights. The conclusions were based on the PCA of the trajectory data and did not involve significant pre-defined physical knowledge. The results show that the PCA approach, which is one of the simplest unsupervised machine learning dimensionality reduction methods, is a powerful one for analyzing complicated nonadiabatic dynamics in the condensed phase with many degrees of freedom.

Distributed Principal Subspace Analysis for Partitioned Big Data: Algorithms, Analysis, and Implementation

Principal Subspace Analysis (PSA) -- and its sibling, Principal Component Analysis (PCA) -- is one of the most popular approaches for dimensionality reduction in signal processing and machine learning. But centralized PSA/PCA solutions are fast becoming irrelevant in the modern era of big data, in which the number of samples and/or the dimensionality of samples often exceed the storage and/or computational capabilities of individual machines. This has led to the study of distributed PSA/PCA solutions, in which the data are partitioned across multiple machines and an estimate of the principal subspace is obtained through collaboration among the machines. It is in this vein that this paper revisits the problem of distributed PSA/PCA under the general framework of an arbitrarily connected network of machines that lacks a central server. The main contributions of the paper in this regard are threefold. First, two algorithms are proposed in the paper that can be used for distributed PSA/PCA, with one in the case of data partitioned across samples and the other in the case of data partitioned across (raw) features. Second, in the case of sample-wise partitioned data, the proposed algorithm and a variant of it are analyzed, and their convergence to the true subspace at linear rates is established. Third, extensive experiments on both synthetic and real-world data are carried out to validate the usefulness of the proposed algorithms. In particular, in the case of sample-wise partitioned data, an MPI-based distributed implementation is carried out to study the interplay between network topology and communications cost as well as to study the effects of straggler machines on the proposed algorithms.

Network Intrusion Detection Systems Using Supervised Machine Learning Classification and Dimensionality Reduction Techniques: A Systematic Review

Network Intrusion Detection Systems Using Supervised Machine Learning Classification and Dimensionality Reduction Techniques: A Systematic Review

Revisiting the Neuropathology of Sudden Infant Death Syndrome (SIDS).

Background: Sudden infant death syndrome (SIDS) is one of the leading causes of infant mortality in the United States (US). The extent to which SIDS manifests with an underlying neuropathological mechanism is highly controversial. SIDS correlates with markers of poor prenatal and postnatal care, generally rooted in the lack of access and quality of healthcare endemic to select racial and ethnic groups, and thus can be viewed in the context of health disparities. However, some evidence suggests that at least a subset of SIDS cases may result from a neuropathological mechanism. To explain these issues, a triple-risk hypothesis has been proposed, whereby an underlying biological abnormality in an infant facing an extrinsic risk during a critical developmental period SIDS is hypothesized to occur. Each SIDS decedent is thus thought to have a unique combination of these risk factors leading to their death. This article reviews the neuropathological literature of SIDS and uses machine learning tools to identify distinct subtypes of SIDS decedents based on epidemiological data.Methods: We analyzed US Period Linked Birth/Infant Mortality Files from 1990 to 2017 (excluding 1992–1994). Using t-SNE, an unsupervised machine learning dimensionality reduction algorithm, we identified clusters of SIDS decedents. Following identification of these groups, we identified changes in the rates of SIDS at the state level and across three countries.Results: Through t-SNE and distance based statistical analysis, we identified three groups of SIDS decedents, each with a unique peak age of death. Within the US, SIDS is geographically heterogeneous. Following this, we found low birth weight and normal birth weight SIDS rates have not been equally impacted by implementation of clinical guidelines. We show that across countries with different levels of cultural heterogeneity, reduction in SIDS rates has also been distinct between decedents with low vs. normal birth weight.Conclusions: Different epidemiological and extrinsic risk factors exist based on the three unique SIDS groups we identified with t-SNE and distance based statistical measurements. Clinical guidelines have not equally impacted the groups, and normal birth weight infants comprise more of the cases of SIDS even though low birth weight infants have a higher SIDS rate.

Self-generated persistent random forces drive phase separation in growing tumors.

A single solid tumor, composed of nearly identical cells, exhibits heterogeneous dynamics. Dynamics of cells in the core is glass-like, whereas those in the periphery undergoes diffusive or super-diffusive behavior. Quantification of heterogeneity using the mean square displacement or the self-intermediate scattering function, which involves averaging over the cell population, hides the complexity of the collective movement. Using the t-distributed stochastic neighbor embedding (t-SNE), a popular unsupervised machine learning dimensionality reduction technique, we show that the phase space structure of an evolving colony of cells, driven by cell division and apoptosis, partitions into nearly disjoint sets composed principally of the core and periphery cells. The non-equilibrium phase separation is driven by the differences in the persistence of self-generated active forces induced by cell division. Extensive heterogeneity revealed by t-SNE paves the way toward understanding the origins of intratumor heterogeneity using experimental imaging data.

Design of Compact and Efficient Silicon Photonic Micro Antennas With Perfectly Vertical Emission

Compact and efficient optical antennas are fundamental components for many applications, including high-density fiber-chip coupling and optical phased arrays. Here we present the design of grating-based micro-antennas with perfectly vertical emission in the 300-nm silicon-on-insulator platform. We leverage a methodology combining adjoint optimization and machine learning dimensionality reduction to efficiently map the multiparameter design space of the antennas, analyse a large number of relevant performance metrics, carry out the required multi-objective optimization, and discover high performance designs. Using a one-step apodized grating we achieve a vertical upward diffraction efficiency of almost 92% with a 3.6 {\mu}m-long antenna. When coupled with an ultra-high numerical aperture fiber, the antenna exhibits a coupling efficiency of more than 81% (-0.9 dB) and a 1-dB bandwidth of almost 158 nm. The reflection generated by the perfectly vertical antenna is smaller than -20 dB on a 200-nm bandwidth centered at {\lambda} = 1550 nm.