Automatic Classification Method Research Articles

Automatic classification of arrhythmias using multi-branch convolutional neural networks based on channel-based attention and bidirectional LSTM

Cardiac arrhythmia is an abnormal rhythm of the heartbeat and can be life-threatening Electrocardiogram (ECG) is a technology that uses an electrocardiograph machine to record a graph of the changes in electrical activity produced by the heart at each cardiac cycle. ECG can generally be used to check whether the examinee has arrhythmia, ion channel disease, cardiomyopathy, electrolyte disorder and other diseases. To reduce the workload of doctors and improve the accuracy of ECG signal recognition, a novel and lightweight automatic ECG classification method based on Convolutional Neural Network (CNN) is proposed. The multi-branch network with different receptive fields is used to extract the multi-spatial deep features of heartbeats. The Channel Attention Module (CAM) and Bidirectional Long Short-Term Memory neural network (BLSTM) module are used to filter redundant ECG features. CAM and BLSTM are beneficial for distinguishing different categories of heartbeats. In the experiments, a four-fold cross-validation technique is used to improve the generalization capability of the network, and it shows good performance on the testing set. This method divides heartbeats into five categories according to the American Advancement of Medical Instrumentation (AAMI) criteria, which is validated in the MIT-BIH arrhythmia database. The sensitivity of this method to Ventricular Ectopic Beat (VEB) is 98.5% and the F1 score is 98.2%. The precision of the Supraventricular Ectopic Beat (SVEB) is 91.1%, and the corresponding F1 score is 90.8%. The proposed method has high classification performance and a lightweight feature. In a word, it has broad application prospects in clinical medicine and health testing.

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet

Automatic defect depth estimation for ultrasonic testing in carbon fiber reinforced composites using deep learning

Ultrasonic testing (UT) is commonly used to inspect the geometric shape of internal damage in composite materials and the test results need to be interpreted by trained experts. In this work, an automatic signal classification method based on deep learning is proposed for depth estimation of the detects introduced by low-velocity impact (LVI) in carbon fiber reinforced plastics (CFRPs). Three kinds of neural networks, LSTM, CNN, and CNN-LSTM are used to analyze the attributes with different depths. Then, trained models are applied to identify the depth information of impact damage. The results show that the CNN-LSTM model is a more accurate in-depth classification for LVI defects in CFRP based on A-scan signals than the other two structures.

A deep feature fusion and selection‐based retinal eye disease detection from OCT images

AbstractOptical coherence tomography (OCT) is one of the principal imaging modalities for retinal eye disease detection and classification. Different retinal eye diseases are the leading cause of blindness that can be overcome by early detection. However, ophthalmologists are currently carrying out retinal eye disease detection manually with the help of OCT images that may be erroneous and subjective. Different methods have been presented to automate the manual retinal eye disease detection process that needs further improvement in detection accuracy. This research proposed an automatic method for retinal eye disease detection and classification from OCT images using fusion and selection techniques. First, the modified‐Alexnet and ResNet‐50 are utilized for deep feature vector extraction. In the next step, these vectors are fused serially and rectified by the proposed feature selection framework and passed as input to different machine learning classifiers for retinal disease diagnosis. For this purpose, a publicly available dataset of retinal eye diseases with four classes is utilized. The proposed retinal eye disease detection method achieved an overall average accuracy index of greater than 99.95%, higher than the top one in the literature, that is, 99.39%. Experimental results authenticated that the proposed retinal eye disease detection methodology can reliably be used for automatic eye disease detection from OCT images. Furthermore, the proposed deep feature and selection‐based retinal eye disease detection methodology achieved state‐of‐the‐art performance.

Classifying the surrounding rock of tunnel face using machine learning

Accurately classifying the surrounding rock of tunnel face is essential. In this paper, we propose a machine learning-based automatic classification and dynamic prediction method of the surrounding rocks of tunnel face using the data monitored by a computerized rock drilling trolley based on the intelligent mechanized construction process for drilling and blasting tunnels. This method provides auxiliary support for the intelligent decision of dynamic support at the construction site. First, this method solves the imbalance in the classification of the surrounding rock samples by constructing the Synthetic Minority Oversampling Technique (SMOTE) algorithm using 500 samples of drilling parameters covering different levels and lithologies of a tunnel. Second, it filters the importance of the characteristic samples based on the random forest method. Third, it uses the XGBoost algorithm to model the processed data and compare it with AdaBoost and BP neural network models. The results show that the XGBoost model achieves a higher accuracy of 87.5% when the sample size is small. Finally, we validate the application scenarios of the above algorithm/model regarding the key aspects of the tunnel construction process, such as surrounding rock identification, design interaction, construction supervision, and quality evaluation, which facilitates the upgrading of intelligent tunnel construction.

Automatic land-sea classification in a nearshore environment using satellite-based photon-counting LiDAR data.

The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) can measure the global surface with unprecedented resolution. Accurate classification of land and sea data is the prerequisite for generating high-quality data products. Current land-sea classification methods rely on assisted data or manual participation, and the automation degree cannot meet the needs of massive data processing. Therefore, using the land-sea difference of photon-counting LiDAR data, an index called normalized photon rate-elevation ratio (NPRER) is designed. Inspired by this, an automatic land-sea classification method is proposed, and the results are obtained through preliminary classification, reclassification, and post-processing enhancement. The results in Cook Inlet, Alaska, show that NPRER can measure the probability of sea appearance in the nearshore environment. At the same time, the automatic classification method can achieve an overall accuracy of 97.98%. The changes in the coastal type, data collection time, and classification feature sets have little influence on this method. Therefore, the method provides a reliable technical scheme for improving the automation of land-sea classification of satellite-based photon-counting LiDAR data.

Evaluation of automatic discrimination between benign and malignant prostate tissue in the era of high precision digital pathology

BackgroundProstate cancer is a major health concern in aging men. Paralleling an aging society, prostate cancer prevalence increases emphasizing the need for efficient diagnostic algorithms.MethodsRetrospectively, 106 prostate tissue samples from 48 patients (mean age, 66pm 6.6 years) were included in the study. Patients suffered from prostate cancer (n = 38) or benign prostatic hyperplasia (n = 10) and were treated with radical prostatectomy or Holmium laser enucleation of the prostate, respectively. We constructed tissue microarrays (TMAs) comprising representative malignant (n = 38) and benign (n = 68) tissue cores. TMAs were processed to histological slides, stained, digitized and assessed for the applicability of machine learning strategies and open–source tools in diagnosis of prostate cancer. We applied the software QuPath to extract features for shape, stain intensity, and texture of TMA cores for three stainings, H&E, ERG, and PIN-4. Three machine learning algorithms, neural network (NN), support vector machines (SVM), and random forest (RF), were trained and cross-validated with 100 Monte Carlo random splits into 70% training set and 30% test set. We determined AUC values for single color channels, with and without optimization of hyperparameters by exhaustive grid search. We applied recursive feature elimination to feature sets of multiple color transforms.ResultsMean AUC was above 0.80. PIN-4 stainings yielded higher AUC than H&E and ERG. For PIN-4 with the color transform saturation, NN, RF, and SVM revealed AUC of 0.93pm 0.04, 0.91pm 0.06, and 0.92pm 0.05, respectively. Optimization of hyperparameters improved the AUC only slightly by 0.01. For H&E, feature selection resulted in no increase of AUC but to an increase of 0.02–0.06 for ERG and PIN-4.ConclusionsAutomated pipelines may be able to discriminate with high accuracy between malignant and benign tissue. We found PIN-4 staining best suited for classification. Further bioinformatic analysis of larger data sets would be crucial to evaluate the reliability of automated classification methods for clinical practice and to evaluate potential discrimination of aggressiveness of cancer to pave the way to automatic precision medicine.

Automatic Segmentation and Classification for Antinuclear Antibody Images Based on Deep Learning.

Antinuclear antibodies (ANAs) testing is the main serological diagnosis screening test for autoimmune diseases. ANAs testing is conducted principally by the indirect immunofluorescence (IIF) on human epithelial cell-substrate (HEp-2) protocol. However, due to its high variability and human subjectivity, there is an insistent need to develop an efficient method for automatic image segmentation and classification. This article develops an automatic segmentation and classification framework based on artificial intelligence (AI) on the ANA images. The Otsu thresholding method and watershed segmentation algorithm are adopted to segment IIF images of cells. Moreover, multiple texture features such as scale-invariant feature transform (SIFT), local binary pattern (LBP), cooccurrence among adjacent LBPs (CoALBP), and rotation invariant cooccurrence among adjacent LBPs (RIC-LBP) are utilized. Firstly, this article adopts traditional machine learning methods such as support vector machine (SVM), k-nearest neighbor algorithm (KNN), and random forest (RF) and then uses ensemble classifier (ECLF) combined with soft voting rules to merge these machine learning methods for classification. The deep learning method InceptionResNetV2 is also utilized to train on the classification of cell images. Eventually, the best accuracy of 0.9269 on the Changsha dataset and 0.9635 on the ICPR 2016 dataset for the traditional methods is obtained by a combination of SIFT and RIC-LBP with the ECLF classifier, and the best accuracy obtained by the InceptionResNetV2 is 0.9465 and 0.9836 separately, which outperforms other schemes.

Methods for Automatic Cyst Detection and Classification in Ultrasound Images of the Female Genitalia Using Image Processing

Methods for Automatic Cyst Detection and Classification in Ultrasound Images of the Female Genitalia Using Image Processing

Interpretable vertebral fracture quantification via anchor-free landmarks localization.

Vertebral body compression fractures are early signs of osteoporosis. Though these fractures are visible on Computed Tomography (CT) images, they are frequently missed by radiologists in clinical settings. Prior research on automatic methods of vertebral fracture classification proves its reliable quality; however, existing methods provide hard-to-interpret outputs and sometimes fail to process cases with severe abnormalities such as highly pathological vertebrae or scoliosis. We propose a new two-step algorithm to localize the vertebral column in 3D CT images and then detect individual vertebrae and quantify fractures in 2D simultaneously. We train neural networks for both steps using a simple 6-keypoints based annotation scheme, which corresponds precisely to the current clinical recommendation. Our algorithm has no exclusion criteria, processes 3D CT in 2 seconds on a single GPU, and provides an interpretable and verifiable output. The method approaches expert-level performance and demonstrates state-of-the-art results in vertebrae 3D localization (the average error is 1mm), vertebrae 2D detection (precision and recall are 0.99), and fracture identification (ROC AUC at the patient level is up to 0.96). Our anchor-free vertebra detection network shows excellent generalizability on a new domain by achieving ROC AUC 0.95, sensitivity 0.85, specificity 0.9 on a challenging VerSe dataset with many unseen vertebra types.

Domain Knowledge Informed Multitask Learning for Landslide-Induced Seismic Classification

Automatic seismic signal classification methods are extensively investigated to reduce or replace manual interpretation, with great potential in previous research. Discriminative seismic wave propagation physical characteristics, such as velocities and accelerations, are rarely considered for classification. A multitask learning scheme is proposed that utilises the seismic wave equation and three-dimensional (3D) P-wave velocity <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Vp</i> model for signal representation learning. The classifier uses the obtained latent feature maps on a convolutional neural network architecture for classification of rockfall, slide quake, earthquake, and natural/anthropogenic noise events, recorded at an ongoing landslide. Our experimental results show that our approach outperforms state-of-the-art methods.

A novel ultrasonic inspection method of the heat exchangers based on circumferential waves and deep neural networks.

The heat exchanger (HE) is an important component of almost every energy generation system. Periodic inspection of the HEs is particularly important to keep high efficiency of the entire system. In this paper, a novel ultrasonic water immersion inspection method is presented based on circumferential wave (CW) propagation to detect defective HE. Thin patch-type piezoelectric elements with multiple resonance frequencies were adopted for the ultrasonic inspection of narrow-spaced HE in an immersion test. Water-filled HE was used to simulate defective HE because water is the most reliable indicator of the defect. The HE will leak water no matter what the defect pattern is. Furthermore, continuous wavelet transform (CWT) was used to investigate the received CW, and inverse CWT was applied to separate frequency bands corresponding to the thickness and lateral resonance modes of the piezoelectric element. Different arrangements of intact and leaky HE were tested with several pairs of thin piezoelectric patch probes in various instrumental setups. Also, direct waveforms in the water without HE were used as reference signals, to indicate instrumental gain and probe sensitivity. Moreover, all filtered CW corresponding to resonance modes together with the direct waveforms in the water were used to train the deep neural networks (DNNs). As a result, an automatic HE state classification method was obtained, and the accuracy of the applied DNN was estimated as 99.99%.

Breast cancer disease prediction using ensemble techniques

Breast Cancer is a highly lethal reproductive cancer that disproportionately affects women and is a leading cause of death worldwide. Cancer is characterized by the uncontrolled division and invasion of abnormal cells into the surrounding tissues. Early detection is crucial in the diagnosis of Breast Cancer, as it accounts for a significant percentage of cancer diagnoses and deaths among women. To prevent unnecessary tests, accurate classification of malignant and benign tumors is necessary. Researchers have developed numerous automated classification methods for Breast Cancer, with soft computing techniques being widely used due to their high performance in classification. Machine learning algorithms, known for their ability to identify critical features from medical datasets, are also extensively utilized in Breast Cancer prediction. Therefore, this study seeks to employ Boosting algorithms in machine learning to predict Breast Cancer accurately. Over the years, the mortality rate in Breast Cancer diagnosis has decreased due to research efforts.

Classification tree algorithm and its application in general archives management system

With the rapid development of higher education in China, more and more archives are managed by the archives of colleges and universities. Therefore, many colleges and universities have equipped archives management software to manage archives by computer. However, at present, the mainstream archives management software for colleges and universities does not have the function of automatic classification of archives. In order to reduce the workload of college archives staff, this paper explores a text automatic classification method suitable for college archives. Classification tree algorithm is a method of organizing and storing information. The classification tree is used to classify data into various categories, such as customers and products. It also provides an effective way to search for data using keywords or phrases. Classification tree is very useful in document processing, because it can help us find documents with similar characteristics, such as customers with similar requirements, products with similar functions, etc. The working principle of classification tree algorithm is to decompose the problem into smaller subproblems, and then solve one subproblem at a time until all problems are completely solved.

Electric Power Audit Text Classification With Multi-Grained Pre-Trained Language Model

Electric power audit text classification is one of the important research problem in electric power systems. Recently, kinds of automatic classification methods for these texts based on machine learning or deep learning models have been applied. At present, the development of computing technology makes “pre-training and fine-tuning” the newest paradigm of text classification, which achieves better results than previous fully-supervised models. Based on pre-training theory, domain-related pre-training tasks can enhance the performance of downstream tasks in the specific domain. However, existing pre-training models usually use general corpus for pre-training, and do not use texts related to the field of electric power, especially electric power audit texts. This results in that the model does not learn too much electric-power-related morphology or semantics in the pre-training stage, so that less information can be used in the fine-tuning stage. Based on the research status, in this paper, we propose EPAT-BERT, a BERT-based model pre-trained by two-granularity pre-training tasks: word-level masked language model and entity-level masked language model. These two tasks predict word and entity in electric-power-related texts to learn abundant morphology and semantics about electric power. We then fine-tune EPAT-BERT for electric power audit text classification task. The experimental results show that, compared with fully supervised machine learning models, neural network models, and general pre-trained language models, EPAT-BERT can significantly outperform existing models in a variety of evaluation metrics. Therefore, EPAT-BERT can be further applied to electric power audit text classification. We also conduct ablation studies to prove the effectiveness of each component in EPAT-BERT to further illustrate our motivations.

Feed Forward Neural Network based Brain Tumor Diagnosis in Magnetic Resonance Images

In the realm of medicine, value, resource use and final care are determined by good technological advancement. However, there are crucial components that must be present for a disease to be diagnosed. The monitoring of illness progression traditionally relies primarily on a subjective human judgment and is neither precise nor timely. One important aspect that utilizes data at various disease progression phases is to maintain routine disease surveillance. The Feed Forward Neural Network based Brain Tumor Diagnosis in Magnetic Resonance Images is provided in this paper as an automatic brain cancer diagnosis and grade classification method. It is highly helpful to have accurate information about the disease in order to classify it and make decisions. The suggested brain tumor diagnosis system can diagnose the condition and provide a reliable foundation for appropriate regulation, as opposed to manual approaches. Finally, the evaluated outcomes of the suggested model investigate numerous Magnetic Resonance Images of healthy and disease and demonstrate that, the proposed method has the highest accuracy.

Exhaled Air Metabolome Analysis for Pulmonary Arterial Hypertension Fingerprints Identification—The Preliminary Study

Pulmonary arterial hypertension (PAH) is a rare disease with a serious prognosis. The aim of this study was to identify biomarkers for PAH in the breath phase and to prepare an automatic classification method to determine the changing metabolome trends and molecular mapping. A group of 37 patients (F/M: 8/29 women, mean age 60.4 ± 10.9 years, BMI 27.6 ± 6.0 kg/m2) with diagnosed PAH were enrolled in the study. The breath phase of all the patients was collected on a highly porous septic material using a special patented holder PL230578, OHIM 002890789-0001. The collected air was then examined with gas chromatography coupled with mass spectrometry (GC/MS). The algorithms of Spectral Clustering, KMeans, DBSCAN, and hierarchical clustering methods were used to perform the cluster analysis. The identification of the changes in the ratio of the whole spectra of biomarkers allowed us to obtain a multidimensional pathway for PAH characteristics and showed the metabolome differences in the four subgroups divided by the cluster analysis. The use of GC/MS, supported with novel porous polymeric materials, for the breath phase analysis seems to be a useful tool in selecting bio-fingerprints in patients with PAH. The four metabolome classes which were obtained constitute novel data in the PAH population.

A Lightweight Decentralized-Learning-Based Automatic Modulation Classification Method for Resource-Constrained Edge Devices

Due to the computing capability and memory limitations, it is difficult to apply the traditional deep learning (DL) models to the edge devices (EDs) for realizing lightweight automatic modulation classification (AMC). Recently, many works attempt to use different ways to realize lightweight AMC methods for EDs. However, the lightweight seems to be a contradiction with the classification performance in these lightweight networks. In this article, we propose an efficient lightweight decentralized-learning-based AMC (DecentAMC) method using spatiotemporal hybrid deep neural network based on multichannels and multifunction blocks (MCMBNN). Specifically, the lightweight network is designed from the perspectives of comprehensive consideration of lightweight and classification performance, which is composed of three parts to extract different features for realizing high classification performance and they are phase estimator and transformer (PET) block, spatial feature extraction block and temporal feature extraction & Softmax block. In addition, we use a multichannel input to extract complementary features of different channels for a better classification performance. The proposed DecentAMC method is an efficient training method, which is achieved by the cooperation in which multiple EDs update and upload the model weight to a central device (CD) for model aggregation to avoid the data privacy disclosure and reduce the computing power and storage pressure of CD. Experimental results show that the proposed MCMBNN can obtain an improved classification accuracy while reducing model complexity with the contributions of three blocks. Moreover, the proposed DecentAMC method can be deployed on EDs efficiently. Thus, the method has the advantages of avoiding data leakage on EDs and relieving the computing pressure of CD with relatively lower communication overhead. The simulation code and datasets are shared on GitHub.

Soybean Pod Morphology Classification Based on Machine Vision and Convolutional Neural Network

Pod morphology is one of the most crucial indicators for breeding high-quality soybeans. Therefore, the accurate classification of pod morphology is pivotal. Aiming at the problem that the morphology of soybean pods cannot be accurately classified, we proposed an automatic classification method based on machine vision and convolutional neural networks in this paper. First, we designed an image acquisition system to obtain soybean pod images. Then, a convolutional neural network model was designed to extract the deep features of soybean pods in the images. Finally, the deep feature training support vector machine (SVM) classifier was used to classify soybean pods of different forms, and the traditional features are designed to compare the morphological classification of soybean pods. The classification accuracies of this method were 98.8%, 99.2%, and 98.1% for straight, curved scythe, and bowhead bean pods, respectively, with an average classification accuracy of 98.7%, and the classification accuracy of deep features is 5.5% higher than that of traditional feature extraction. Experimental results prove that the proposed automatic classification method is feasible for soybean pod morphology classification, and it also has potential in classification of other characteristics of soybean, such as leaves, grains, and diseases.

Cross-scenario automatic sleep stage classification using transfer learning and single-channel EEG

Sleep staging via manual inspection is time-consuming and inefficient, and as such, automatic sleep stage classification methods have been proposed and successfully implemented. However, these conventional supervised learning models are not suitable for cross-scenario sleep staging tasks due to the shift data distribution problem. This study aims to explore the use of transfer learning to transfer knowledge from a labeled source dataset to a new target domain where labels are difficult to obtain, so as to help solve the classification problem in a new domain in practical sleep staging scenarios. A novel end-to-end deep learning network that includes a feature extractor, a sleep stage classifier, and a domain adaptation network was proposed. The proposed domain adaptation network can accomplish distribution alignment between source and target domains, through the application of both joint distribution loss and stage transition loss. Cross-channel, cross-subject and channel, and cross-dataset experiments were designed to verify the feasibility of the proposed network using three publicly available datasets. The results indicate an average accuracy improvement ranging from 0.1% to 6.1% (average of 2.9%), a macro-averaging F1-score improvement ranging from 0.1% to 5.1% (average of 2.6%), and a Cohen’s kappa coefficient improvement ranging from 0.016 to 0.083 (average of 0.045) after using transfer learning, when compared with a non-transfer model. Relative to similar existing transfer learning methods for sleep staging, the proposed method was implemented in an unsupervised manner and achieved success in cross-scenario sleep staging tasks, which is more suitable for practical applications in daily life.