Multi-class Identification Research Articles

Knowledge-informed FIR-based cross-category filtering framework for interpretable machinery fault diagnosis under small samples

Relying on sufficient training data, the existing fault diagnosis methods rarely focus on the methodological interpretability and the data scarcity in real industrial scenarios simultaneously. Motivated by this issue, we deeply reexamined the intrinsic characteristics of fault signals and the guiding significance of classical signal-processing methods for feature enhancement. From the perspective of multiscale modes, this study tailors multiple learnable knowledge-informed finite impulse response (FIR) filtering kernels to extract sensitive modes for explainable feature enhancement. On this foundation, a knowledge-informed FIR-based cross-category filtering (FIR-CCF) framework is further proposed for interpretable small-sample fault diagnosis. With the consideration of the mode complexity, a cross-category filtering strategy is explored to further enhance feature expressions for identifying single state. To be special, this strategy divides a multi-class recognition process into multiple two-class recognition task. A multi-task learning is then presented where multiple binary-class base learners (BCBLearners) that consists of a feature extractor and a two-class classifier is established to seek discriminate mode features for each type of state. Eventually, all feature extractors are fixed and a multi-class classifier is established and to fuse all mode features for high-precision multi-class identification via ensemble learning. As a variant of signal-processing-collaborated deep learning frameworks, the FIR-CCF method fully exploits the strengths of signal-processing methods in interpretability and feature extraction. Three experimental cases highlight the superiority and significant improvement of the FIR-CCF framework over other five state-of-the-art diagnosis methods and three ablation models. Specially, extensive visualization is implemented to place in-depth insight into how the FIR-CCF framework works. It can be also foreseen that the signal-processing-collaborated deep learning framework shows enormous potential in interpretable fault diagnosis for knowledge-informed artificial intelligence. Related source codes will be available at: https://github.com/BITS/FIR-CCF-main.

MetaNet: Interpretable unknown mobile malware identification with a novel meta-features mining algorithm

The continuous emergence of malware has threatened to the Android platform and user privacy. With the evolution of the Android system and malware, it is challenging to design a method that can accurately identify the categories of sophisticated malware, including known and unknown families, as well as their obfuscated variants, given that they may be newly emerging and lack available detection knowledge. Although some methods try to use anomaly detection and zero-shot technology to identify unseen applications, they are limited to binary classification or lack the robustness, stability, universality, and interpretability in multi-class identification. To this end, we first propose a generic meta-features mining algorithm, which can discover the potential relationships between samples belonging to the same category. Then we present metaNet, a novel method leveraging meta-features to identify sophisticated Android malware. Specifically, metaNet is mainly powered by four components: (i) mExtractor is a feature collector to obtain the static and dynamic features. (ii) mProcessor is taking unique meta-features of each category from extracted features. (iii) mLearner is a machine learning suite that leverages features and meta-features to design and train a classifier called HSU-Net. (iv) mEnforcer is a flexible deployer that identifies categories of malware families in the real world. We implement a prototype of metaNet with 15K lines of Python code and compare it with state-of-the-art (SOTA) methods. The results show that it can not only achieve superior performance in terms of known families (99.52% of accuracy) and unknown families (99.31% of accuracy trained with 80% known families) for binary classification, but also perform well in multi-class identification, i.e., 99.05% and 93.45% of accuracy for known and unknown families, respectively. Furthermore, we deploy and evaluate metaNet in the real world. It can identify applications over an acceptable time and memory overheads, i.e., average of 11.8 s and 56 MB per sample with a size of 8 MB. Also, the few-shot detection and feature perturbation experiments reflect its robustness and stability benefiting from meta-features. Finally, we collect the traffic of 112 decentralized applications (DApps) belonging to 16 categories, such as finance and health, and evaluated metaNet in DApp identification. The results illustrate its applicability across various tasks. That is, it can accurately classify 94.6% and 81.36% of DApp flows in all-known and 80%-known DApp scenarios, respectively, outperforming the SOTA methods.

Unveiling AI-Generated Financial Text: A Computational Approach Using Natural Language Processing and Generative Artificial Intelligence

This study is an in-depth exploration of the nascent field of Natural Language Processing (NLP) and generative Artificial Intelligence (AI), and it concentrates on the vital task of distinguishing between human-generated text and content that has been produced by AI models. Particularly, this research pioneers the identification of financial text derived from AI models such as ChatGPT and paraphrasing tools like QuillBot. While our primary focus is on financial content, we have also pinpointed texts generated by paragraph rewriting tools and utilized ChatGPT for various contexts this multiclass identification was missing in previous studies. In this paper, we use a comprehensive feature extraction methodology that combines TF–IDF with Word2Vec, along with individual feature extraction methods. Importantly, combining a Random Forest model with Word2Vec results in impressive outcomes. Moreover, this study investigates the significance of the window size parameters in the Word2Vec approach, revealing that a window size of one produces outstanding scores across various metrics, including accuracy, precision, recall and the F1 measure, all reaching a notable value of 0.74. In addition to this, our developed model performs well in classification, attaining AUC values of 0.94 for the ‘GPT’ class; 0.77 for the ‘Quil’ class; and 0.89 for the ‘Real’ class. We also achieved an accuracy of 0.72, precision of 0.71, recall of 0.72, and F1 of 0.71 for our extended prepared dataset. This study contributes significantly to the evolving landscape of AI text identification, providing valuable insights and promising directions for future research.

Multi-class identification of tonal contrasts in Chokri using supervised machine learning algorithms

This study examines and explores the effectiveness of various Machine Learning Algorithms (MLAs) in identifying intricate tonal contrasts in Chokri (ISO 639-3), an under-documented and endangered Tibeto-Burman language of the Sino-Tibetan language family spoken in Nagaland, India. Seven different supervised MLAs, viz., [Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB)], and one neural network (NN)-based algorithms [Artificial Neural Network (ANN)] are implemented to explore five-way tonal contrasts in Chokri. Acoustic correlates of tonal contrasts, encompassing fundamental frequency fluctuations, viz., f0 height and f0 direction, are examined. Contrary to the prevailing notion of NN supremacy, this study underscores the impressive accuracy achieved by the RF. Additionally, it reveals that combining f0 height and directionality enhances tonal contrast recognition for female speakers, while f0 directionality alone suffices for male speakers. The findings demonstrate MLAs’ potential to attain accuracy rates of 84–87% for females and 95–97% for males, showcasing their applicability in deciphering the intricate tonal systems of Chokri. The proposed methodology can be extended to predict multi-class problems in diverse fields such as image processing, speech classification, medical diagnosis, computer vision, and social network analysis.

Deep transfer CNNs models performance evaluation using unbalanced histopathological breast cancer dataset

Cancer is one of the top deadly diseases. Of this disease, around about 9.8 million death cause annually. It has been recorded by the American Cancer Society that every eight women die due to breast cancer in the USA. In this paper, we have identified eight different lesion categories: Benign Tumor: Adenosis-Adenoma, Fibro-Adenoma, Phyllodes-Tumor, Tubular-Adenoma, and Malignant Tumor; Ductal-Carcinoma, Lobular-Carcinoma, Mucinous-Carcinoma, Papillary-Carcinoma. The main contribution of this paper is to examine the performance of five pre-trained CNN models on an unbalanced cancer dataset for cancer prediction. The identification of different cancer tumors has been recognized by using transfer learning models namely ResNet50, ResNet101, ResNet152, VGG16, and VGG19. BreakHis dataset has four different magnifications (40x-100x-200x-400x), and used for experiments setup in this study. The total number of images for all magnification levels is 7909. The experimental results state that the pre-trained model Residual Net with different variations has worked well 91%~96% as compared to other pre-trained models. The ResNet101 architecture model has gained a multiclass identification higher than 95%. In this paper, the proposed methodology has different evaluation parameters such as accuracy, recall, and f1-score of all pre-trained models that will help to build optimal, and automated breast lesion multiclass identification.

Explainable equivariant neural networks for particle physics: PELICAN

PELICAN is a novel permutation equivariant and Lorentz invariant or covariant aggregator network designed to overcome common limitations found in architectures applied to particle physics problems. Compared to many approaches that use non-specialized architectures that neglect underlying physics principles and require very large numbers of parameters, PELICAN employs a fundamentally symmetry group-based architecture that demonstrates benefits in terms of reduced complexity, increased interpretability, and raw performance. We present a comprehensive study of the PELICAN algorithm architecture in the context of both tagging (classification) and reconstructing (regression) Lorentz-boosted top quarks, including the difficult task of specifically identifying and measuring the W-boson inside the dense environment of the Lorentz-boosted top-quark hadronic final state. We also extend the application of PELICAN to the tasks of identifying quark-initiated vs. gluon-initiated jets, and a multi-class identification across five separate target categories of jets. When tested on the standard task of Lorentz-boosted top-quark tagging, PELICAN outperforms existing competitors with much lower model complexity and high sample efficiency. On the less common and more complex task of 4-momentum regression, PELICAN also outperforms hand-crafted, non-machine learning algorithms. We discuss the implications of symmetry-restricted architectures for the wider field of machine learning for physics.

Remote Emotion Recognition Using Continuous-Wave Bio-Radar System.

The Bio-Radar is herein presented as a non-contact radar system able to capture vital signs remotely without requiring any physical contact with the subject. In this work, the ability to use the proposed system for emotion recognition is verified by comparing its performance on identifying fear, happiness and a neutral condition, with certified measuring equipment. For this purpose, machine learning algorithms were applied to the respiratory and cardiac signals captured simultaneously by the radar and the referenced contact-based system. Following a multiclass identification strategy, one could conclude that both systems present a comparable performance, where the radar might even outperform under specific conditions. Emotion recognition is possible using a radar system, with an accuracy equal to 99.7% and an F1-score of 99.9%. Thus, we demonstrated that it is perfectly possible to use the Bio-Radar system for this purpose, which is able to be operated remotely, avoiding the subject awareness of being monitored and thus providing more authentic reactions.

Fault detection and identification in induction motor using weightless neural network.

This paper presents a novel multiclass, multiparameter real-valued weightless neural network-based classifier for fault detection and identification. In contrast to primitive variants of weightless neural nets, the proposed method is capable of multiclass identification with real-valued input features and has improved recognition and generalization capabilities. The major accomplishments of the proposed method include an improved input-to-address mapping strategy that is suitable for address assignment within discriminator units and an effective memory expansion scheme based on similarity metric-based membership value. The developed classifier is utilized for fault detection and identification in single-phase and three-phase induction motors. The sensitivity analysis of the method is investigated for design parameter variation and the fault diagnosis is performed for multiple faults of the motor. The proposed method achieves the highest accuracy of 99.6% and 89.25% for single-phase and three-phase induction motor fault identification, respectively.

A rapid LC-MS/MS method for multi-class identification and quantification of cyanotoxins

Cyanobacteria can form harmful blooms in specific environmental conditions due to certain species producing toxic metabolites known as cyanotoxins. These toxins pose significant risks to public health and the environment, making it critical to identify and quantify them in food and water sources to avoid contamination. However, current screening methods only focus on a single class of cyanotoxins, limiting their effectiveness. Thus, fast and sensitive liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) method was developed to analyze eighteen cyanotoxins simultaneously. A simplified extraction procedure using lyophilized samples of cyanobacterial biomass was also used, eliminating the need for traditional solid-phase extraction methods. This method uses multiple reaction monitoring and allows accurate determination and quantification of eighteen cyanotoxins, including anatoxin-a, homoanatoxin-a, cylindrospermopsin, deoxy-cylindrospermopsin, nodularin, guanitoxin, seven microcystins (RR, [D-Asp3] RR, LA, LR, LY, LW, and YR), and five saxitoxins (gonyautoxins - GTX-1&4, GTX-2&3, GTX-5), decarbamoylgonyautoxin (dcGTX-2&3), and N-Sulfocarbamoylgonyautoxin (C1&C2), all in a short acquisition time of 8 min. Therefore, this method provides a simple and efficient approach to identify and quantify harmful compounds produced by cyanobacteria. Hence, this represents the first method to detecting guanitoxin among cyanotoxins. By expanding the range of toxins analyzed, this method can help ensure high-quality food and drinking water and protect recreational users from exposure to cyanotoxins.

Identifying Alzheimer's disease and mild cognitive impairment with atlas-based multi-modal metrics.

Multi-modal neuroimaging metrics in combination with advanced machine learning techniques have attracted more and more attention for an effective multi-class identification of Alzheimer's disease (AD), mild cognitive impairment (MCI) and health controls (HC) recently. In this paper, a total of 180 subjects consisting of 44 AD, 66 MCI and 58 HC subjects were enrolled, and the multi-modalities of the resting-state functional magnetic resonance imaging (rs-fMRI) and the structural MRI (sMRI) for all participants were obtained. Then, four kinds of metrics including the Hurst exponent (HE) metric and bilateral hippocampus seed independently based connectivity metrics generated from fMRI data, and the gray matter volume (GMV) metric obtained from sMRI data, were calculated and extracted in each region of interest (ROI) based on a newly proposed automated anatomical Labeling (AAL3) atlas after data pre-processing. Next, these metrics were selected with a minimal redundancy maximal relevance (MRMR) method and a sequential feature collection (SFC) algorithm, and only a subset of optimal features were retained after this step. Finally, the support vector machine (SVM) based classification methods and artificial neural network (ANN) algorithm were utilized to identify the multi-class of AD, MCI and HC subjects in single modal and multi-modal metrics respectively, and a nested ten-fold cross-validation was utilized to estimate the final classification performance. The results of the SVM and ANN based methods indicated the best accuracies of 80.36 and 74.40%, respectively, by utilizing all the multi-modal metrics, and the optimal accuracies for AD, MCI and HC were 79.55, 78.79 and 82.76%, respectively, in the SVM based method. In contrast, when using single modal metric, the SVM based method obtained a best accuracy of 72.62% with the HE metric, and the accuracies for AD, MCI and HC subjects were just 56.82, 80.30 and 75.86%, respectively. Moreover, the overlapping abnormal brain regions detected by multi-modal metrics were mainly located at posterior cingulate gyrus, superior frontal gyrus and cuneus. Taken together, the SVM based method with multi-modal metrics could provide effective diagnostic information for identifying AD, MCI and HC subjects.

A New Deep Learning Framework for Imbalance Detection of a Rotating Shaft.

Rotor unbalance is the most common cause of vibration in industrial machines. The unbalance can result in efficiency losses and decreased lifetime of bearings and other components, leading to system failure and significant safety risk. Many complex analytical techniques and specific classifiers algorithms have been developed to study rotor imbalance. The classifier algorithms, though simple to use, lack the flexibility to be used efficiently for both low and high numbers of classes. Therefore, a robust multiclass prediction algorithm is needed to efficiently classify the rotor imbalance problem during runtime and avoid the problem's escalation to failure. In this work, a new deep learning (DL) algorithm was developed for detecting the unbalance of a rotating shaft for both binary and multiclass identification. The model was developed by utilizing the depth and efficacy of ResNet and the feature extraction property of Convolutional Neural Network (CNN). The new algorithm outperforms both ResNet and CNN. Accelerometer data collected by a vibration sensor were used to train the algorithm. This time series data were preprocessed to extract important vibration signatures such as Fast Fourier Transform (FFT) and Short-Time Fourier Transform (STFT). STFT, being a feature-rich characteristic, performs better on our model. Two types of analyses were carried out: (i) balanced vs. unbalanced case detection (two output classes) and (ii) the level of unbalance detection (five output classes). The developed model gave a testing accuracy of 99.23% for the two-class classification and 95.15% for the multilevel unbalance classification. The results suggest that the proposed deep learning framework is robust for both binary and multiclass classification problems. This study provides a robust framework for detecting shaft unbalance of rotating machinery and can serve as a real-time fault detection mechanism in industrial applications.

Improved Object Detection Artificial Intelligence Using the Revised RetinaNet Model for the Automatic Detection of Ulcerations, Vascular Lesions, and Tumors in Wireless Capsule Endoscopy.

The use of computer-aided detection models to diagnose lesions in images from wireless capsule endoscopy (WCE) is a topical endoscopic diagnostic solution. We revised our artificial intelligence (AI) model, RetinaNet, to better diagnose multiple types of lesions, including erosions and ulcers, vascular lesions, and tumors. RetinaNet was trained using the data of 1234 patients, consisting of images of 6476 erosions and ulcers, 1916 vascular lesions, 7127 tumors, and 14,014,149 normal tissues. The mean area under the receiver operating characteristic curve (AUC), sensitivity, and specificity for each lesion were evaluated using five-fold stratified cross-validation. Each cross-validation set consisted of between 6,647,148 and 7,267,813 images from 217 patients. The mean AUC values were 0.997 for erosions and ulcers, 0.998 for vascular lesions, and 0.998 for tumors. The mean sensitivities were 0.919, 0.878, and 0.876, respectively. The mean specificities were 0.936, 0.969, and 0.937, and the mean accuracies were 0.930, 0.962, and 0.924, respectively. We developed a new version of an AI-based diagnostic model for the multiclass identification of small bowel lesions in WCE images to help endoscopists appropriately diagnose small intestine diseases in daily clinical practice.

A hybrid explainable ensemble transformer encoder for pneumonia identification from chest X-ray images

IntroductionPneumonia is a microorganism infection that causes chronic inflammation of the human lung cells. Chest X-ray imaging is the most well-known screening approach used for detecting pneumonia in the early stages. While chest-Xray images are mostly blurry with low illumination, a strong feature extraction approach is required for promising identification performance. ObjectivesA new hybrid explainable deep learning framework is proposed for accurate pneumonia disease identification using chest X-ray images. MethodsThe proposed hybrid workflow is developed by fusing the capabilities of both ensemble convolutional networks and the Transformer Encoder mechanism. The ensemble learning backbone is used to extract strong features from the raw input X-ray images in two different scenarios: ensemble A (i.e., DenseNet201, VGG16, and GoogleNet) and ensemble B (i.e., DenseNet201, InceptionResNetV2, and Xception). Whereas, the Transformer Encoder is built based on the self-attention mechanism with multilayer perceptron (MLP) for accurate disease identification. The visual explainable saliency maps are derived to emphasize the crucial predicted regions on the input X-ray images. The end-to-end training process of the proposed deep learning models over all scenarios is performed for binary and multi-class classification scenarios. ResultsThe proposed hybrid deep learning model recorded 99.21% classification performance in terms of overall accuracy and F1-score for the binary classification task, while it achieved 98.19% accuracy and 97.29% F1-score for multi-classification task. For the ensemble binary identification scenario, ensemble A recorded 97.22% accuracy and 97.14% F1-score, while ensemble B achieved 96.44% for both accuracy and F1-score. For the ensemble multiclass identification scenario, ensemble A recorded 97.2% accuracy and 95.8% F1-score, while ensemble B recorded 96.4% accuracy and 94.9% F1-score. ConclusionThe proposed hybrid deep learning framework could provide promising and encouraging explainable identification performance comparing with the individual, ensemble models, or even the latest AI models in the literature. The code is available here: https://github.com/chiagoziemchima/Pneumonia_Identificaton.

Volumetric macromolecule identification in cryo-electron tomograms using capsule networks

BackgroundDespite recent advances in cellular cryo-electron tomography (CET), developing automated tools for macromolecule identification in submolecular resolution remains challenging due to the lack of annotated data and high structural complexities. To date, the extent of the deep learning methods constructed for this problem is limited to conventional Convolutional Neural Networks (CNNs). Identifying macromolecules of different types and sizes is a tedious and time-consuming task. In this paper, we employ a capsule-based architecture to automate the task of macromolecule identification, that we refer to as 3D-UCaps. In particular, the architecture is composed of three components: feature extractor, capsule encoder, and CNN decoder. The feature extractor converts voxel intensities of input sub-tomograms to activities of local features. The encoder is a 3D Capsule Network (CapsNet) that takes local features to generate a low-dimensional representation of the input. Then, a 3D CNN decoder reconstructs the sub-tomograms from the given representation by upsampling.ResultsWe performed binary and multi-class localization and identification tasks on synthetic and experimental data. We observed that the 3D-UNet and the 3D-UCaps had an F_1-score mostly above 60% and 70%, respectively, on the test data. In both network architectures, we observed degradation of at least 40% in the F_1-score when identifying very small particles (PDB entry 3GL1) compared to a large particle (PDB entry 4D8Q). In the multi-class identification task of experimental data, 3D-UCaps had an F_1-score of 91% on the test data in contrast to 64% of the 3D-UNet. The better F_1-score of 3D-UCaps compared to 3D-UNet is obtained by a higher precision score. We speculate this to be due to the capsule network employed in the encoder. To study the effect of the CapsNet-based encoder architecture further, we performed an ablation study and perceived that the F_1-score is boosted as network depth is increased which is in contrast to the previously reported results for the 3D-UNet. To present a reproducible work, source code, trained models, data as well as visualization results are made publicly available.ConclusionQuantitative and qualitative results show that 3D-UCaps successfully perform various downstream tasks including identification and localization of macromolecules and can at least compete with CNN architectures for this task. Given that the capsule layers extract both the existence probability and the orientation of the molecules, this architecture has the potential to lead to representations of the data that are better interpretable than those of 3D-UNet.

COLET: A dataset for COgnitive workLoad estimation based on eye-tracking

Background and Objective: The cognitive workload is an important component in performance psychology, ergonomics, and human factors. Publicly available datasets are scarce, making it difficult to establish new approaches and comparative studies. In this work, COLET-COgnitive workLoad estimation based on Eye-Tracking dataset is presented. Methods: Forty-seven (47) individuals’ eye movements were monitored as they solved puzzles involving visual search activities of varying complexity and duration. The participants’ cognitive workload level was evaluated with the subjective test of NASA-TLX and this score is used as an annotation of the activity. Extensive data analysis was performed in order to derive eye and gaze features from low-level eye recorded metrics, and a range of machine learning models were evaluated and tested regarding the estimation of the cognitive workload level. Results: The activities induced four different levels of cognitive workload. Multi tasking and time pressure have induced a higher level of cognitive workload than the one induced by single tasking and absence of time pressure. Multi tasking had a significant effect on 17 eye features while time pressure had a significant effect on 7 eye features. Both binary and multi-class identification attempts were performed by testing a variety of well-known classifiers, resulting in encouraging results towards cognitive workload levels estimation, with up to 88% correct predictions between low and high cognitive workload. Conclusions: Machine learning analysis demonstrated potential in discriminating cognitive workload levels using only eye-tracking characteristics. The proposed dataset includes a much higher sample size and a wider spectrum of eye and gaze metrics than other similar datasets, allowing for the examination of their relations with various cognitive states.

Performance evaluation of deep transfer learning on multi-class identification of common weed species in cotton production systems

Precision weed management offers a promising solution for sustainable cropping systems through the use of chemical-reduced/non-chemical robotic weeding techniques, which apply suitable control tactics to individual weeds or small clusters. Therefore, accurate identification of weed species plays a crucial role in such systems to enable precise, individualized weed treatment. Despite recent progress, the development of a robust weed identification and localization system in the presence of unstructured field environments remains a serious challenge, requiring supervised modeling using large volumes of annotated data. This paper makes a first comprehensive evaluation of deep transfer learning (DTL) for identifying common weed species specific to cotton (Gossypium hirsutum L.) production systems in southern United States (U.S.). A new dataset for weed identification was created, consisting of 5187 color images of 15 weed classes collected under natural light conditions and at varied weed growth stages, in cotton fields (primarily in Mississippi and North Carolina) during the 2020 and 2021 growth seasons. We evaluated 35 state-of-the-art deep learning models through transfer learning with repeated holdout validations and established an extensive benchmark for the considered weed identification task. DTL achieved high classification accuracy of F1 scores exceeding 95%, requiring reasonably short training time (less than 2.5 h) across models. ResNeXt101 achieved the best overall F1-score of 98.93 ± 0.34%, whereas 10 out of the 35 models achieved F1 scores near or above 98.0%. However, the performance on minority weed classes with few training samples was less satisfactory for models trained with a conventional, unweighted cross entropy loss function. To address this issue, a weighted cross entropy loss function was adopted, which achieved substantially improved accuracies for minority weed classes (e.g., the F1-scores for Xception and MnasNet on the Spurred Anoda weed increased from 48% to 90% and 50% to 82%, respectively). Furthermore, a deep learning-based cosine similarity metric was employed to analyze the similarity among weed classes, assisting in the interpretation of classifications. Both the codes (https://github.com/Derekabc/CottonWeeds) for model benchmarking and the weed dataset (https://www.kaggle.com/yuzhenlu/cottonweedid15) of this study are made publicly available, which expect to be a valuable resource for future research on weed identification and beyond.

Intertwining Binary Decision Trees and Probabilistic Neural Networks for Maximising Accuracy and Efficiency in Classification Tasks A Pilot /Proof–of–Concept Demonstration on the Iris Benchmark Classification Dataset

Intertwining binary decision trees (BDTs) and probabilistic neural networks (PNNs) is put forward as a powerful custom–made methodology for simultaneously maximising accuracy and efficiency in classification tasks. The proposed methodology brings together the complementary strengths of (I) the parametric, global, recursive, efficient as well as maximal dichotomisation of the BDT, and (II) the non–parametric, accurate–to–local–detail, multi–class identification of the PNN. The BDT/PNN combination evolves level–by–level, with each level comprising two steps: (step 1) the optimal BDT (hyper–)linear bi–partitioning of the entire dataset for the first level, and of every non–terminal partition thereof for the next levels, is determined, with each resulting partition being assigned to a new node of the next BDT level; (step 2) as many PNNs are created as there are  multi–class partitions resulting from (step 1), implementing the non–parametric creation of the hyperplane that maximises class separability over each corresponding partition of the problem–space. BDT/PNN expansion is applied recursively – and independently – to each non–terminal partition of the dataset, concluding once the cumulative classification accuracy (CCA) of the terminal – BDT and PNN – nodes of the BDT/PNN is maximised. A proof–of–concept demonstration on the iris benchmark classification dataset (IBCD) illustrates the details of the BDT/PNN combination, attesting to its simplicity, effectiveness, and efficiency of operation. Follow–up research shall focus upon the classification of additional benchmark datasets, as well as of “hard” and “Big” real–world problems, for further evaluating and validating the proposed methodology.

DeepMC-iNABP: Deep learning for multiclass identification and classification of nucleic acid-binding proteins

Nucleic acid-binding proteins (NABPs), including DNA-binding proteins (DBPs) and RNA-binding proteins (RBPs), play vital roles in gene expression. Accurate identification of these proteins is crucial. However, there are two existing challenges: one is the problem of ignoring DNA- and RNA-binding proteins (DRBPs), and the other is a cross-predicting problem referring to DBP predictors predicting DBPs as RBPs, and vice versa. In this study, we proposed a computational predictor, called DeepMC-iNABP, with the goal of solving these difficulties by utilizing a multiclass classification strategy and deep learning approaches. DBPs, RBPs, DRBPs and non-NABPs as separate classes of data were used for training the DeepMC-iNABP model. The results on test data collected in this study and two independent test datasets showed that DeepMC-iNABP has a strong advantage in identifying the DRBPs and has the ability to alleviate the cross-prediction problem to a certain extent. The web-server of DeepMC-iNABP is freely available at http://www.deepmc-inabp.net/. The datasets used in this research can also be downloaded from the website.

BCI‐control and monitoring system for smart home automation using wavelet classifiers

Brain Computer Interface (BCI) is a major research field that is based upon Electroencephalography (EEG) brain signals, which are captured using EEG electrodes, amplified and filtered before being converted to the digital form in order to perform thorough pre-processing and machine-learning. In this study, the design and implementation of the BCI control and monitoring system for smart home automation using wavelet features, which is based upon a dual-channel analogue EEG signal acquisition module is reported. The designed analogue EEG module performs EEG signal acquisition, signal amplification and filtering. Although the EEG data set contains thousands of samples and more than 15 different classes, we limit our study on 226 samples grouped into seven classes with 8-second time duration per sample. With careful settings of deep-learning classifier model parameters, the training and testing were successful with high accuracy results. The designed BCI system has several advantages including a large bandwidth of 400 Hz, low number of EEG electrodes, easy setup, simple user interface, pre-processing and digital filtering, fast machine learning, multi-class identification, monitoring and control models, high classification accuracy and low cost. This research work provided several contributions including the creation of recent and original EEG data set using well-labelled recordings at an adequate sampling rate of 2 kHz. The EEG signal acquisition module with 400-Hz bandwidth provides precise and rich EEG signal information needed for feature extraction. Our results are reproducible and have been tested and deployed on Raspberry pi 4 with Python. The designed wavelet-based BCI system consists of analogue EEG signal acquisition and machine-learning modules, which consist of deep-learning Multi-layer perceptron (MLP) classifiers and linear discriminant analysis (LDA) as well as other classifier models for comparison including convolutional neural networks (CNN). The deep learning and LDA classifiers models produced the best performance with average accuracy of 95.6% and 96% for both training and testing data sets.

Identification of Autism Subtypes Based on Wavelet Coherence of BOLD FMRI Signals Using Convolutional Neural Network.

The functional connectivity (FC) patterns of resting-state functional magnetic resonance imaging (rs-fMRI) play an essential role in the development of autism spectrum disorders (ASD) classification models. There are available methods in literature that have used FC patterns as inputs for binary classification models, but the results barely reach an accuracy of 80%. Additionally, the generalizability across multiple sites of the models has not been investigated. Due to the lack of ASD subtypes identification model, the multi-class classification is proposed in the present study. This study aims to develop automated identification of autism spectrum disorder (ASD) subtypes using convolutional neural networks (CNN) using dynamic FC as its inputs. The rs-fMRI dataset used in this study consists of 144 individuals from 8 independent sites, labeled based on three ASD subtypes, namely autistic disorder (ASD), Asperger’s disorder (APD), and pervasive developmental disorder not otherwise specified (PDD-NOS). The blood-oxygen-level-dependent (BOLD) signals from 116 brain nodes of automated anatomical labeling (AAL) atlas are used, where the top-ranked node is determined based on one-way analysis of variance (ANOVA) of the power spectral density (PSD) values. Based on the statistical analysis of the PSD values of 3-level ASD and normal control (NC), putamen_R is obtained as the top-ranked node and used for the wavelet coherence computation. With good resolution in time and frequency domain, scalograms of wavelet coherence between the top-ranked node and the rest of the nodes are used as dynamic FC feature input to the convolutional neural networks (CNN). The dynamic FC patterns of wavelet coherence scalogram represent phase synchronization between the pairs of BOLD signals. Classification algorithms are developed using CNN and the wavelet coherence scalograms for binary and multi-class identification were trained and tested using cross-validation and leave-one-out techniques. Results of binary classification (ASD vs. NC) and multi-class classification (ASD vs. APD vs. PDD-NOS vs. NC) yielded, respectively, 89.8% accuracy and 82.1% macro-average accuracy, respectively. Findings from this study have illustrated the good potential of wavelet coherence technique in representing dynamic FC between brain nodes and open possibilities for its application in computer aided diagnosis of other neuropsychiatric disorders, such as depression or schizophrenia.