Small Sample Size Datasets Research Articles

TabDEG: Classifying differentially expressed genes from RNA-seq data based on feature extraction and deep learning framework.

Traditional differential expression genes (DEGs) identification models have limitations in small sample size datasets because they require meeting distribution assumptions, otherwise resulting high false positive/negative rates due to sample variation. In contrast, tabular data model based on deep learning (DL) frameworks do not need to consider the data distribution types and sample variation. However, applying DL to RNA-Seq data is still a challenge due to the lack of proper labeling and the small sample size compared to the number of genes. Data augmentation (DA) extracts data features using different methods and procedures, which can significantly increase complementary pseudo-values from limited data without significant additional cost. Based on this, we combine DA and DL framework-based tabular data model, propose a model TabDEG, to predict DEGs and their up-regulation/down-regulation directions from gene expression data obtained from the Cancer Genome Atlas database. Compared to five counterpart methods, TabDEG has high sensitivity and low misclassification rates. Experiment shows that TabDEG is robust and effective in enhancing data features to facilitate classification of high-dimensional small sample size datasets and validates that TabDEG-predicted DEGs are mapped to important gene ontology terms and pathways associated with cancer.

Sparse L0-norm least squares support vector machine with feature selection

Least squares support vector machine (LSSVM) is a powerful classification tool based on hyperplanes. But the classical LSSVM does not perform well on small sample size data sets (SSS) because it lacks feature selection capability. To address this issue, we introduce an L0-norm regularization term into the objective function of LSSVM, and propose a new sparse LSSVM model called L0-LSSVM. However, the introduction of the L0-norm makes the resulting optimization problem nonconvex and nonsmooth. To overcome these challenges, instead of using the traditional convex or non-convex approximation of the L0-norm, we design an efficient algorithm for L0-LSSVM based on alternating direction method of multipliers (ADMM) where the L0-norm is handled by employing the proximity operator. Numerical results demonstrate that L0-LSSVM not only exhibits excellent generalization performance and feature selection capability but also offers a substantial improvement in computational speed.

ADscreen: A speech processing-based screening system for automatic identification of patients with Alzheimer's disease and related dementia

Alzheimer's disease and related dementias (ADRD) present a looming public health crisis, affecting roughly 5 million people and 11 % of older adults in the United States. Despite nationwide efforts for timely diagnosis of patients with ADRD, >50 % of them are not diagnosed and unaware of their disease. To address this challenge, we developed ADscreen, an innovative speech-processing based ADRD screening algorithm for the protective identification of patients with ADRD. ADscreen consists of five major components: (i) noise reduction for reducing background noises from the audio-recorded patient speech, (ii) modeling the patient's ability in phonetic motor planning using acoustic parameters of the patient's voice, (iii) modeling the patient's ability in semantic and syntactic levels of language organization using linguistic parameters of the patient speech, (iv) extracting vocal and semantic psycholinguistic cues from the patient speech, and (v) building and evaluating the screening algorithm. To identify important speech parameters (features) associated with ADRD, we used the Joint Mutual Information Maximization (JMIM), an effective feature selection method for high dimensional, small sample size datasets. Modeling the relationship between speech parameters and the outcome variable (presence/absence of ADRD) was conducted using three different machine learning (ML) architectures with the capability of joining informative acoustic and linguistic with contextual word embedding vectors obtained from the DistilBERT (Bidirectional Encoder Representations from Transformers). We evaluated the performance of the ADscreen on an audio-recorded patients' speech (verbal description) for the Cookie-Theft picture description task, which is publicly available in the dementia databank. The joint fusion of acoustic and linguistic parameters with contextual word embedding vectors of DistilBERT achieved F1-score = 84.64 (standard deviation [std] = ±3.58) and AUC-ROC = 92.53 (std = ±3.34) for training dataset, and F1-score = 89.55 and AUC-ROC = 93.89 for the test dataset. In summary, ADscreen has a strong potential to be integrated with clinical workflow to address the need for an ADRD screening tool so that patients with cognitive impairment can receive appropriate and timely care.

Research on Fire Detection in Laboratories Based on CNN and Transfer Learning

To solve the problem that model may not be convergent owing to datasets with small sample sizes results from the secrecy of labs, a fire detection method for lab fire detection based on CNN (Convolutional Neural Network) and transfer leaning is proposed. Experiments of transfer learning were conducted with configuration of super parameters and a small lab fire dataset, whose pre-trained models are GoogLeNet, ResNet and MobileNet trained with ImageNet dataset. We analyze and assess the loss function curve and the performances of the above models. Result shows that MoblieNet model is chosen to be the optimum model for fire detection, providing a reference to lab fire detection with small sample size datasets. Different from artificial fire surveillance, the proposed method based on CNN and transfer learning realize real-time, autonomy and efficient fire detection.

Real-time classification for Φ-OTDR vibration events in the case of small sample size datasets

Efficient classification of vibration signals detected by phase-sensitive optical time domain reflectometer (Φ-OTDR) based on small samples is an effective method to reduce the false alarm rate without GPU or large data sets. This paper proposes a fiber optic system vibration event recognition method based on a combination of image segmentation pre-processing, texture, statistical, morphological feature extraction, and weighted support vector machine (WSVM), which can effectively classify-five types of vibration events in high-speed railway perimeter intrusion detection with small sample data and no parallel processing units. Erosion and dilation operations are applied to vibration signal image feature enhancement in image pre-processing. The vibration signal region and background are separated by the maximum inter-class variance method, then 35 features of the vibration signal region are calculated and finally employed to construct a WSVM. Experiments show that the method achieves 99 FPS and 98.8% accuracy on the test set with 330 vibration images as the training set to build the model without GPU and in the presence of interference signals. It provides a generalized Φ-OTDR vibration event recognition method for small samples.

Feature fusion Siamese network for breast cancer detection comparing current and prior mammograms.

Automatic detection of very small and nonmass abnormalities from mammogram images has remained challenging. In clinical practice for each patient, radiologists commonly not only screen the mammogram images obtained during the examination, but also compare them with previous mammogram images to make a clinical decision. To design an artificial intelligence (AI) system to mimic radiologists for better cancer detection, in this work we proposed an end-to-end enhanced Siamese convolutional neural network to detect breast cancer using previous year and current year mammogramimages. The proposed Siamese-based network uses high-resolution mammogram images and fuses features of pairs of previous year and current year mammogram images to predict cancer probabilities. The proposed approach is developed based on the concept of one-shot learning that learns the abnormal differences between current and prior images instead of abnormal objects, and as a result can perform better with small sample size data sets. We developed two variants of the proposed network. In the first model, to fuse the features of current and previous images, we designed an enhanced distance learning network that considers not only the overall distance, but also the pixel-wise distances between the features. In the other model, we concatenated the features of current and previous images to fusethem. We compared the performance of the proposed models with those of some baseline models that use current images only (ResNet and VGG) and also use current and prior images (long short-term memory [LSTM] and vanilla Siamese) in terms of accuracy, sensitivity, precision, F1 score, and area under the curve (AUC). Results show that the proposed models outperform the baseline models and the proposed model with the distance learning network performs the best (accuracy: 0.92, sensitivity: 0.93, precision: 0.91, specificity: 0.91, F1: 0.92 and AUC: 0.95). Integrating prior mammogram images improves automatic cancer classification, specially for very small and nonmass abnormalities. For classification models that integrate current and prior mammogram images, using an enhanced and effective distance learning network can advance the performance of themodels.

Estimating Rubber Covered Conveyor Belting Cure Times Using Multiple Simultaneous Optimizations Ensemble

Multiple response surface methodology (MRSM) has been the favorite method for optimizing multiple response processes though it has two weaknesses which challenge the credibility of its solutions. The first weakness is the use of experimentally generated small sample size datasets, and the second is the selection, using classical model selection criteria, of single best models for each response for use in simultaneous optimization to obtain the optimum or desired solution. Classical model selection criteria do not always agree on the best model resulting in model uncertainty. The selection of single best models for each response for simultaneous optimization loses information in rejected models. This work proposes the use of multiple simultaneous optimizations to estimate multiple solutions that are ensembled in solving a conveyor belting cure time problem. The solution is compared with one obtained by simultaneous optimization of single best models for each response. The two results were different. However, results show that it is possible to obtain a more credible solution through ensembling of solutions from multiple simultaneous optimizations.

Reconstructing codependent cellular cross-talk in lung adenocarcinoma using REMI.

Cellular cross-talk in tissue microenvironments is fundamental to normal and pathological biological processes. Global assessment of cell-cell interactions (CCIs) is not yet technically feasible, but computational efforts to reconstruct these interactions have been proposed. Current computational approaches that identify CCI often make the simplifying assumption that pairwise interactions are independent of one another, which can lead to reduced accuracy. We present REMI (REgularized Microenvironment Interactome), a graph-based algorithm that predicts ligand-receptor (LR) interactions by accounting for LR dependencies on high-dimensional, small–sample size datasets. We apply REMI to reconstruct the human lung adenocarcinoma (LUAD) interactome from a bulk flow-sorted RNA sequencing dataset, then leverage single-cell transcriptomics data to increase the cell type resolution and identify LR prognostic signatures among tumor-stroma-immune subpopulations. We experimentally confirmed colocalization of CTGF:LRP6 among malignant cell subtypes as an interaction predicted to be associated with LUAD progression. Our work presents a computational approach to reconstruct interactomes and identify clinically relevant CCIs.

The prediction power of machine learning on estimating the sepsis mortality in the intensive care unit

BackgroundThe prediction of sepsis mortality of intensive care unit (ICU) observations using machine learning (ML) methods is hypothesized to yield better or as good as performance compared to the prognostic scores. This paper aims to show that the accuracy of ML in sepsis mortality estimation can be superior and supportive knowledge to SAPS II, APACHE II, and SOFA (traditional) scores even under small sample restrictions. MethodsThe retrospective collection of data from the patients (n = 200) admitted to ICU of Acibadem Hospital, Istanbul-Turkey, between 2015 and 2020 is utilized to detect the sepsis mortality risk using eight ML methods and a generated ensemble model along with the traditional prognostic scores. The mortality as a decisive indicator is evaluated according to the explanatory variables included quantifying the traditional scores. In the calibration of the data, five different predetermined splits of the random samples are used for the training and the validation of the ML methods. The efficiency of the prediction results of ML methods and the traditional scoring methods are investigated by AUC-ROC curves and other accuracy indicators. Consecutive processes of Box-Cox and Min-Max transformations on data and parameter optimization are performed to increase the efficiency of algorithms. ResultsThe accuracy in the mortality prediction is achieved the best by the Multi-Layer Perceptron algorithm compared to SAPS II and APACHE II methods and is as good as the one with what SOFA predicts. The prediction power of the best performing ML methods for APACHE II, SAPS II, and SOFA are found to be 84.45%, 85.25% and 73.47%, respectively. The ensemble of eight ML methods is found to increase the performance around 2% in APACHE II score. ConclusionsThe outcomes of this study have clinical merits in evaluating the potential use of ML methods in predicting ICU mortality superior to traditional scores APACHE II, SAPS II, and as good as SOFA. Additionally, it explores which of the variables contributing to sepsis mortality risk should be taken as apriori information in treating the patients and requires fewer number of explanatory variables, with reliable prediction powers even for considerably small sample size data sets.

Multi-task manifold learning for small sample size datasets

In this study, we develop a method for multi-task manifold learning. The method aims to improve the performance of manifold learning for multiple tasks, particularly when each task has a small number of samples. Furthermore, the method also aims to generate new samples for new tasks, in addition to new samples for existing tasks. In the proposed method, we use two different types of information transfer: instance transfer and model transfer. For instance transfer, datasets are merged among similar tasks, whereas for model transfer, the manifold models are averaged among similar tasks. For this purpose, the proposed method consists of a set of generative manifold models corresponding to the tasks, which are integrated into a general model of a fiber bundle. We applied the proposed method to artificial datasets and face image sets, and the results showed that the method was able to estimate the manifolds, even for a tiny number of samples.

Deep belief networks with self-adaptive sparsity

To have the sparsity of deep neural networks is crucial, which can improve the learning ability of them, especially for application to high-dimensional data with small sample size. Commonly used regularization terms for keeping the sparsity of deep neural networks are based on L1-norm or L2-norm; however, they are not the most reasonable substitutes of L0-norm. In this paper, based on the fact that the minimization of a log-sum function is one effective approximation to that of L0-norm, the sparse penalty term on the connection weights with the log-sum function is introduced. By embedding the corresponding iterative re-weighted-L1 minimization algorithm with k-step contrastive divergence, the connections of deep belief networks can be updated in a way of sparse self-adaption. Experiments on two kinds of biomedical datasets which are two typical small sample size datasets with a large number of variables, i.e., brain functional magnetic resonance imaging data and single nucleotide polymorphism data, show that the proposed deep belief networks with self-adaptive sparsity can learn the layer-wise sparse features effectively. And results demonstrate better performances including the identification accuracy and sparsity capability than several typical learning machines.

A Bonferroni Mean Based Fuzzy K Nearest Centroid Neighbor Classifier

K-nearest neighbor (KNN) is an effective nonparametric classifier that determines the neighbors of a point based only on distance proximity. The classification performance of KNN is disadvantaged by the presence of outliers in small sample size datasets and its performance deteriorates on datasets with class imbalance. We propose a local Bonferroni Mean based Fuzzy K-Nearest Centroid Neighbor (BM-FKNCN) classifier that assigns class label of a query sample dependent on the nearest local centroid mean vector to better represent the underlying statistic of the dataset. The proposed classifier is robust towards outliers because the Nearest Centroid Neighborhood (NCN) concept also considers spatial distribution and symmetrical placement of the neighbors. Also, the proposed classifier can overcome class domination of its neighbors in datasets with class imbalance because it averages all the centroid vectors from each class to adequately interpret the distribution of the classes. The BM-FKNCN classifier is tested on datasets from the Knowledge Extraction based on Evolutionary Learning (KEEL) repository and benchmarked with classification results from the KNN, Fuzzy-KNN (FKNN), BM-FKNN and FKNCN classifiers. The experimental results show that the BM-FKNCN achieves the highest overall average classification accuracy of 89.86% compared to the other four classifiers.

A Cascade Flexible Neural Forest Model for Cancer Subtypes Classification on Gene Expression Data.

The correct classification of cancer subtypes is of great significance for the in-depth study of cancer pathogenesis and the realization of accurate treatment for cancer patients. In recent years, the classification of cancer subtypes using deep neural networks and gene expression data has become a hot topic. However, most classifiers may face the challenges of overfitting and low classification accuracy when dealing with small sample size and high-dimensional biological data. In this paper, the Cascade Flexible Neural Forest (CFNForest) Model was proposed to accomplish cancer subtype classification. CFNForest extended the traditional flexible neural tree structure to FNT Group Forest exploiting a bagging ensemble strategy and could automatically generate the model's structure and parameters. In order to deepen the FNT Group Forest without introducing new hyperparameters, the multilayer cascade framework was exploited to design the FNT Group Forest model, which transformed features between levels and improved the performance of the model. The proposed CFNForest model also improved the operational efficiency and the robustness of the model by sample selection mechanism between layers and setting different weights for the output of each layer. To accomplish cancer subtype classification, FNT Group Forest with different feature sets was used to enrich the structural diversity of the model, which make it more suitable for processing small sample size datasets. The experiments on RNA-seq gene expression data showed that CFNForest effectively improves the accuracy of cancer subtype classification. The classification results have good robustness.

A Robust Least Squares Support Vector Machine Based on L∞-norm

A family of norm-based least squares support vector machines (LSSVMs) is the effective statistical learning tool and has attracted considerable attentions. However, there are two critical problems: (1) LSSVM is ill-conditioned or singular when the sample size is much less than feature number and thus causes over-fitting for small sample size (SSS) problem, while other norm-based LSSVMs own slower training speed and cannot deal with large scale data. (2) Norm-based LSSVMs pay less attention to the edge points that are important for learning the final classifier. To overcome the above drawbacks, by replacing $$ L_{2}\text{-}norm $$ with $$ L_{\infty }\text{-}norm $$ in empirical loss, we construct a robust classifier, named $$ L_{\infty }\text{-}LSSVM $$ in short. After adjustment, $$ L_{\infty }\text{-}LSSVM $$ can detect edge points effectively and enhances the capability of the robustness and the generalization. Also, inspired by the idea of sequential minimal optimization (SMO), we design a novel SMO-typed iterative algorithm. The new algorithm not only guarantees the convergence of optimum solution in theory, but also owns the lower computational time and storage space when data size is large. Finally, extensive numerical experiments validate the above opinions again on four groups of artificial data, Non-i.i.d data: fault detection of railway turnout, four SSS datasets, two types of massive datasets and six benchmark datasets.

Efficient Feature Selection via $\ell _{2,0}$ ℓ2,0-norm Constrained Sparse Regression

Sparse regression based feature selection method has been extensively investigated these years. However, because it has a non-convex constraint, i.e., $\ell _{2,0}$ℓ2,0-norm constraint, this problem is very hard to solve. In this paper, unlike most of the other methods which only solve its slack version by introducing sparsity regularization into objective function forcibly, a novel framework is proposed by us to solve the original $\ell _{2,0}$ℓ2,0-norm constrained sparse regression based feature selection problem. We transform our objective function into Linear Discriminant Analysis (LDA) by using a new label coding method, thus enabling our model to calculate the ratio of inter-class scatter to intra-class scatter of features which is the most widely used feature discrimination evaluation metric. According to that ratio, features can be selected by a simple sorting method. The projection gradient descent algorithm is introduced to further improve the performance of our algorithm by using the solution obtained before as its initial solution. This ensures the stability of this iterative algorithm. We prove that the proposed method can get the global optimal solution of this non-convex problem when all features are statistically independent. For the general case where features are statistically dependent, extensive experiments on six small sample size datasets and one large-scale dataset show that our algorithm has comparable or better classification capability comparing with other eight state-of-the-art feature selection methods by the SVM classifier. We also show that our algorithm can obtain a low loss value, which means the solution of our algorithm can get very close to this NP-hard problem’s real solution. What is more, because we solve the original $\ell _{2,0}$ℓ2,0-norm constrained problem, we avoid the heavy work of tuning the regularization parameter because its meaning is explicit in our method, i.e., the number of selected features. At last, we evaluate the stability of our algorithm from two perspectives, i.e., the objective function values and the selected features, by experiments. From both perspectives, our algorithm shows satisfactory stability performance.

A Practical Solution to the Small Sample Size Bias and Uncertainty Problems of Model Selection Criteria in Two-Input Process Multiple Response Surface Methodology Datasets

Multiple response surface methodology (MRSM) most often involves the analysis of small sample size datasets which have associated inherent statistical modeling problems. Firstly, classical model selection criteria in use are very inefficient with small sample size datasets. Secondly, classical model selection criteria have an acknowledged selection uncertainty problem. Finally, there is a credibility problem associated with modeling small sample sizes of the order of most MRSM datasets. This work focuses on determination of a solution to these identified problems. The small sample model selection uncertainty problem is analysed using sixteen model selection criteria and a typical two-input MRSM dataset. Selection of candidate models, for the responses in consideration, is done based on response surface conformity to expectation to deliberately avoid selection of models using the problematic classical model selection criteria. A set of permutations of combinations of response models with conforming response surfaces is determined. Each combination is optimised and results are obtained using overlaying of data matrices. The permutation of results is then averaged to obtain credible results. Thus, a transparent multiple model approach is used to obtain the solution which gives some credibility to the small sample size results of the typical MRSM dataset. The conclusion is that, for a two-input process MRSM problem, conformity of response surfaces can be effectively used to select candidate models and thus the use of the problematic model selection criteria is avoidable.

A semi-supervised learning approach for model selection based on class-hypothesis testing

Abstract This paper deals with the topic of learning from unlabeled or noisy-labeled data in the context of a classification problem. In the classification problem the outcome yields one of a discrete set of values thus, assumptions on them could be established to obtain the most likely prediction model at the training stage . In this paper, a novel case-based model selection method is proposed, which combines hypothesis testing from a discrete set of expected outcomes and feature extraction within a cross-validated classification stage. This wrapper-type procedure acts on fully-observable variables under hypothesis-testing and improves the classification accuracy on the test set, or keeps its performance at least at the level of the statistical classifier. The model selection strategy in the cross validation loop allows building an ensemble classifier that could improve the performance of any expert and intelligence system, particularly on small sample-size datasets. Experiments were carried out on several databases yielding a clear improvement on the baseline, i.e., SPECT dataset A c c = 86.35 ± 1.51 , with S e n = 91.10 ± 2.77 , and S p e = 81.11 ± 1.61 . In addition, the CV error estimate for the classifier under our approach was found to be an almost unbiased estimate (as the baseline approach) of the true error that the classifier would incur on independent data.

Bayesian nonparametric estimation of hazard rate in monotone Aalen model

This text describes a method of estimating the hazard rate of survival data following monotone Aalen regression model. The proposed approach is based on techniques which were introduced by Arjas and Gasbarra [4]. The unknown functional parameters are assumed to be a priori piecewise constant on intervals of varying count and size. The estimates are obtained with the aid of the Gibbs sampler and its variants. The performance of the method is explored by simulations. The results indicate that the method is applicable on small sample size datasets.

Two-way analysis of high-dimensional collinear data

We present a Bayesian model for two-way ANOVA-type analysis of high-dimensional, small sample-size datasets with highly correlated groups of variables. Modern cellular measurement methods are a main application area; typically the task is differential analysis between diseased and healthy samples, complicated by additional covariates requiring a multi-way analysis. The main complication is the combination of high dimensionality and low sample size, which renders classical multivariate techniques useless. We introduce a hierarchical model which does dimensionality reduction by assuming that the input variables come in similarly-behaving groups, and performs an ANOVA-type decomposition for the set of reduced-dimensional latent variables. We apply the methods to study lipidomic profiles of a recent large-cohort human diabetes study.

Integration of prior knowledge of measurement noise in kernel density classification

Samples can be measured with different precisions and reliabilities in different experiments, or even within the same experiment. These varying levels of measurement noise may deteriorate the performance of a pattern recognition system, if not treated with care. Here we seek to investigate the benefit of incorporating prior knowledge about measurement noise into system construction. We propose a kernel density classifier which integrates such prior knowledge. Instead of using an identical kernel for each sample, we transform the prior knowledge into a distinct kernel for each sample. The integration procedure is straightforward and easy to interpret. In addition, we show how to estimate the diverse measurement noise levels in a real world dataset. Compared to the basic methods, the new kernel density classifier can give a significantly better classification performance. As expected, this improvement is more obvious for small sample size datasets and large number of features.