Learning High-level Features Research Articles

Hybrid model-based early diagnosis of esophageal disorders using convolutional neural network and refined logistic regression

Accurate diagnosis of the stage of esophageal disorders is crucial in the treatment planning for patients with esophageal cancer and in improving the 5-year survival rate. The progression of esophageal cancer typically begins with precancerous esophageal disorders such as gastroesophageal reflux disease (GERD), esophagitis, and non-dysplasia Barrett’s esophagus, eventually advancing to low- and high-dysplasia Barrett’s esophagus and ultimately to esophageal adenocarcinoma (EAC). The majority of prior research efforts have primarily focused on the identification of general gastrointestinal (GI) tract diseases and the detection of esophageal cancer, with limited attention to the diverse spectrum of esophageal disorders. To address this gap, an innovative framework called Hybrid Model-Based Esophageal Disorder Diagnosis (HMEDD) is developed in this work. The primary goal of HMEDD is to enable early diagnosis of various esophageal disorders using gastroscopic images. HMEDD combines the feature extraction capabilities of an Esophageal Convolutional Neural Network (EsoNet) with the high classification accuracy of a Refined Logistic Regression (RLR) model. EsoNet comprises 14 weight layers and kernels 3×3\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\left(3\ imes 3\\right)$$\\end{document} used for high-level deep feature learning. Esophageal disorders are classified using the RLR model, which is developed by fine-tuning hyperparameters in the traditional Logistic Regression (LR) model using Random Search Cross-Validation (RandomizedSearchCV). HMEDD is extensively validated using a data set containing numerous esophageal abnormalities captured through gastroscopic images. The results of this work demonstrate the effectiveness of HMEDD in accurately classifying different esophageal disorders, with an impressive accuracy of 92.15%. These findings will assist physicians in the accurate early diagnosis of esophageal disorders, ultimately preventing their progression to cancer.

CDAM-Net: Channel shuffle dual attention based multi-scale CNN for efficient glaucoma detection using fundus images

Glaucoma is a typical eye disorder that induces damage to the optic nerve due to increased intraocular pressure. It ultimately leads to partial or complete blindness without clinical reversal. Hence, it is of utmost importance to screen and detect glaucoma at an early stage. Most of the earlier glaucoma diagnosis methods rely on manual feature engineering, which is time-consuming and requires domain experts. Although recent methods, particularly, convolutional neural networks (CNNs) facilitate learning high-level feature representations from fundus images, they need extensive parameters and pose overfitting issues due to insufficient training samples. Further, conventional CNNs often ignore the minute changes in the lesion region. To overcome these issues, we propose a lightweight multi-scale CNN architecture called as CDAM-Net for effective glaucoma identification from retinal fundus images. Additionally, we introduce an attention module called channel shuffle dual attention (CSDA), comprising of a channel attention block, a spatial attention block and a channel shuffle unit, to focus on important regions in the fundus images, thereby extracting class-specific features. The CDAM-Net mainly consists of multi-scale feature representation (MFR) blocks that enable the extraction of multi-scale features from fundus images. Each MFR block is followed by a CSDA module, which further helps enrich the feature representation. The CDAM-Net is evaluated on a retinal fundus image (RFI) dataset containing 1426 fundus images (837 glaucoma and 589 normal), and the results indicate that CDAM-Net yields promising classification performance compared to existing techniques. Also, ablation studies are carried out to test the effectiveness of each component of the CDAM-Net.

Multi-perspective interpretation for one-dimensional conventional neural network model to identify iron-bearing waste material

Identification of iron-bearing waste material (IWM) is challenging, as the physical and chemical properties of IWMs vary greatly with sources and lacking of common markers. Here, we proposed an interpretable method based on a one-dimensional convolutional neural network (1D CNN) and laser-induced breakdown spectroscopy (LIBS) that can recognize IWM. The proposed 1D CNN model obtained a test accuracy of 99.08%±1.37% with five-time repetition and showed sensitivities and specificities per class within the range of 98%-99%. To shed light into the black-box nature of neural network, a multi-perspective interpretation of CNN algorithm towards the analysis of spectroscopy was performed to provide a more complete view of feature importance, intermediate outputs and decision process. As for identifying IWM, the most critical elements captured by the CNN using a visualization algorithm (class activation map) were Na, Mn, Al, Ca, Fe, Mg, K, Cr, Zn, F and Cl. In the intermediate outputs, it was found that the kernels in the same convolution extract different local informative variables and contribute to learning high-level features. In the final decision process, the CNN was proved to accumulate more valuable information over the increased convolutions using another visualization algorithm (t-SNE) and the output was calculated through mathematical inference. This is the first study in which CNN-assisted LIBS spectroscopy was used to identify IWM while providing a multi-perspective interpretable method for CNN algorithm, which can lay a foundation for widening the application of CNN-based spectroscopic techniques in waste identification.

Predicting circRNA-RBP Binding Sites Using a Hybrid Deep Neural Network.

Circular RNAs (circRNAs) are non-coding RNAs generated by reverse splicing. They are involved in biological process and human diseases by interacting with specific RNA-binding proteins (RBPs). Due to traditional biological experiments being costly, computational methods have been proposed to predict the circRNA-RBP interaction. However, these methods have problems of single feature extraction. Therefore, we propose a novel model called circ-FHN, which utilizes only circRNA sequences to predict circRNA-RBP interactions. The circ-FHN approach involves feature coding and a hybrid deep learning model. Feature coding takes into account the physicochemical properties of circRNA sequences and employs four coding methods to extract sequence features. The hybrid deep structure comprises a convolutional neural network (CNN) and a bidirectional gated recurrent unit (BiGRU). The CNN learns high-level abstract features, while the BiGRU captures long-term dependencies in the sequence. To assess the effectiveness of circ-FHN, we compared it to other computational methods on 16 datasets and conducted ablation experiments. Additionally, we conducted motif analysis. The results demonstrate that circ-FHN exhibits exceptional performance and surpasses other methods. circ-FHN is freely available at https://github.com/zhaoqi106/circ-FHN .

Optimizing Speech Emotion Recognition with Deep Learning and Grey Wolf Optimization: A Multi-Dataset Approach

Machine learning and speech emotion recognition are rapidly evolving fields, significantly impacting human-centered computing. Machine learning enables computers to learn from data and make predictions, while speech emotion recognition allows computers to identify and understand human emotions from speech. These technologies contribute to the creation of innovative human–computer interaction (HCI) applications. Deep learning algorithms, capable of learning high-level features directly from raw data, have given rise to new emotion recognition approaches employing models trained on advanced speech representations like spectrograms and time–frequency representations. This study introduces CNN and LSTM models with GWO optimization, aiming to determine optimal parameters for achieving enhanced accuracy within a specified parameter set. The proposed CNN and LSTM models with GWO optimization underwent performance testing on four diverse datasets—RAVDESS, SAVEE, TESS, and EMODB. The results indicated superior performance of the models compared to linear and kernelized SVM, with or without GWO optimizers.

A learning and evolution-based intelligence algorithm for multi-objective heterogeneous cloud scheduling optimization

The multi-objective directed acyclic graph scheduling problem (MDAGSP) is prevalent in cloud scheduling systems, involving the selection, assignment, and execution of multiple tasks/jobs with complex coupling interdependencies. High-quality solutions can yield substantial economic benefits. However, prevailing methods face challenges in obtaining a set of superior solutions for MDAGSP, due to the multifaceted nature of its variables, objectives, constraints, and heterogeneity. Firstly, this paper formulates a three-objective MDAGSP that includes makespan, energy costs and revenue, to model cloud scheduling systems. Subsequently, we propose a composite algorithm consisting of a selection phase and an assignment phase to automatically generate an efficient scheduling policy for this model. During the selection phase, a graph convolutional neural network learns high-level features to extract complex dependencies between tasks. During the assignment phase, an adaptive evolutionary algorithm assigns tasks to the appropriate executors. Finally, a series of experiments are conducted to validate the model’s accuracy and assess the algorithm’s efficiency. Compared to heuristic approaches, the algorithm achieves at least a 20.1% makespan reduction and 3.17% revenue increase. Remarkably, the algorithm obtains at least an 8.86% reduction in energy costs over the eleven baselines. In conclusion, the proposed algorithm provides decision-makers with a global view of scheduling plan.

Forecasting credit default risk with graph attention networks

The importance of credit default risk management has risen that companies can utilize it to identify and forecast future credit default risk. Several approaches have been proposed, however, they paid little attention on the various underlying relationships between users, which can provide significant improvement. In this paper, we propose a Graph Attention Network (GAT)-based model for predicting credit default risk, leveraging various types of data, including credit default history, credit status and personal profile. These data provide a comprehensive representation of users’ overall status, including historical financial credit, recent financial credit and wealth status. Different graphs are constructed based on the similarities between users using these data, respectively. Then, for graphs, GAT modules are used to capture both the relationships with adjacent and high-order neighbors, as well as the linear and non-linear relationships. After fusing learned high-level features from GAT modules, final predictive results, whether users will default or not, are predicted. The effectiveness of our prediction model is validated using real-world datasets, and experimental results depict that our model can accurately predict credit default risks, outperforming several baseline methods. The codes and datasets are freely available at https://github.com/ZJUDataIntelligence/Foreknow.

Vehicular Network Intrusion Detection Using a Cascaded Deep Learning Approach with Multi-Variant Metaheuristic.

Vehicle malfunctions have a direct impact on both human and road safety, making vehicle network security an important and critical challenge. Vehicular ad hoc networks (VANETs) have grown to be indispensable in recent years for enabling intelligent transport systems, guaranteeing traffic safety, and averting collisions. However, because of numerous types of assaults, such as Distributed Denial of Service (DDoS) and Denial of Service (DoS), VANETs have significant difficulties. A powerful Network Intrusion Detection System (NIDS) powered by Artificial Intelligence (AI) is required to overcome these security issues. This research presents an innovative method for creating an AI-based NIDS that uses Deep Learning methods. The suggested model specifically incorporates the Self Attention-Based Bidirectional Long Short-Term Memory (SA-BiLSTM) for classification and the Cascaded Convolution Neural Network (CCNN) for learning high-level features. The Multi-variant Gradient-Based Optimization algorithm (MV-GBO) is applied to improve CCNN and SA-BiLSTM further to enhance the model's performance. Additionally, information gained using MV-GBO-based feature extraction is employed to enhance feature learning. The effectiveness of the proposed model is evaluated on reliable datasets such as KDD-CUP99, ToN-IoT, and VeReMi, which are utilized on the MATLAB platform. The proposed model achieved 99% accuracy on all the datasets.

Temporal segment dropout for human action video recognition

Temporal information is important for human action video recognition. With the widely used spatio-temporal neural networks, researchers have found that the learned high-level features preserve overfitted spatial information and limited temporal information, leading to inferior performance. This is because existing networks lack efficient regularization for the temporal structure. To learn more robust temporal features, we propose a temporal regularization method named Temporal Segment Dropout (TSD). TSD drops the most salient spatial features in order to enhance the temporal features in a clip of temporal segments. Without learning from complex examples, TSD can be easily deployed in existing networks. In the experiment, TSD is extensively evaluated on benchmark action recognition datasets, which brings consistent improvements over the baselines, especially for the action-centric classes.

Feature relocation network for fine-grained image classification

In fine-grained image classification, there are only very subtle differences between classes. It is challenging to learn local discriminative features and remove distractive features in fine-grained image classification. Existing fine-grained image classification methods learn discriminative feature mainly via manual part annotation or attention mechanisms. However, due to the large intraclass variance and interclass similarity, the discriminative information and distractive information still are not distinguished effectively. To address this problem, we propose a feature relocation network (FRe-Net) which takes advantage of the different natures of features learned from different stages of the network. Our network consists of a distractive feature learning module and a relocated high-level feature learning module. In the distractive feature learning module, we propose to exploit the difference between low-level features and high-level features to design a distractive loss Ldistractive, which guides the attention to locate distractive regions more accurately. In the relocated high-level feature learning module, we enhance the representing capacity of the middle-level feature via the attention module and subtract the distractive feature learned from the distractive feature learning module in order to learn more local discriminative features. In end-to-end model training, the distractive feature learning module and the relocated high-level feature learning module are beneficial to each other via joint optimization. We conducted comprehensive experiments on three benchmark datasets widely used in fine-grained image classification. The experimental results show that FRe-Net achieves state-of-the-art performance, which validates the effectiveness of FRe-Net.

Dynamic Representation Learning via Recurrent Graph Neural Networks

A large number of real-world systems generate graphs that are structured data aligned with nodes and edges. Graphs are usually dynamic in many scenarios, where nodes or edges keep evolving over time. Recently, graph representation learning (GRL) has received great success in network analysis, which aims to produce informative and representative features or low-dimensional embeddings by exploring node attributes and network topology. Most state-of-the-art models for dynamic GRL are composed of a static representation learning model and a recurrent neural network (RNN). The former generates the representations of a graph or nodes from one static graph at a discrete time step, while the latter captures the temporal correlation between adjacent graphs. However, the two-stage design ignores the temporal dynamics between contiguous graphs during the learning processing of graph representations. To alleviate this problem, this article proposes a representation learning model for dynamic graphs, called DynGNN. Differently, it is a single-stage model that embeds an RNN into a graph neural network to produce better representations in a compact form. This takes the fusion of temporal and topology correlations into account from low-level to high-level feature learning, enabling the model to capture more fine-grained evolving patterns. From the experimental results on both synthetic and real-world networks, the proposed DynGNN yields significant improvements in multiple tasks compared to the state-of-the-art counterparts.

Masked Autoencoders for 3D Point Cloud Self-supervised Learning

Self-supervised learning methods are able to learn latent features from unlabeled data samples by designing pre-text tasks, and hence have attracted a great deal of interest in terms of sample-efficient learning. As a promising scheme of self-supervised learning, masked autoencoding has significantly advanced natural language processing and computer vision, but has not been introduced to point cloud yet. In this paper, we propose a novel scheme of masked autoencoders for 3D point cloud self-supervised learning, addressing the special challenges posed by point cloud, including leakage of location information and uneven information density. Concretely, we divide the input point cloud into irregular point patches and randomly mask them at a high ratio. Then, a standard Transformer-based autoencoder, with an asymmetric design and a shifting mask tokens operation, learns high-level latent features from unmasked point patches, aiming to reconstruct the masked point patches. Extensive experiments show that our approach is efficient during pre-training and generalizes well on various downstream tasks. Apart from the proposed method, we will also introduce potential directions for 3D point cloud self-supervised learning, including improvements for masked autoencoding, developments for point cloud scene understanding, etc.

Adaptive Mask Sampling and Manifold to Euclidean Subspace Learning With Distance Covariance Representation for Hyperspectral Image Classification

For the abundant spectral and spatial information recorded in hyperspectral images (HSIs), fully exploring spectral-spatial relationships has attracted widespread attention in hyperspectral image classification (HSIC) community. However, there are still some intractable obstructs. For one thing, in the patch based processing pattern, some spatial neighbor pixels are often inconsistent with the central pixel in land-cover class. For another thing, linear and nonlinear correlations between different spectral bands are vital yet tough for representing and excavating. To overcome these mentioned issues, an adaptive mask sampling and manifold to Euclidean subspace learning (AMS-M2ESL) framework is proposed for HSIC. Specifically, an adaptive mask based intra-patch sampling (AMIPS) module is firstly formulated for intra-patch sampling in an adaptive mask manner based on central spectral vector oriented spatial relationships. Then, based on distance covariance descriptor, a dual channel distance covariance representation (DC-DCR) module is proposed for modeling unified spectral-spatial feature representations and exploring spectral-spatial relationships, especially linear and nonlinear interdependence in spectral domain. Furthermore, considering that distance covariance matrix lies on the symmetric positive definite (SPD) manifold, we implement a manifold to Euclidean subspace learning (M2ESL) module respecting Riemannian geometry of SPD manifold for high-level spectral-spatial feature learning. Additionally, we introduce an approximate matrix square-root (ASQRT) layer for efficient Euclidean subspace projection. Extensive experimental results on three popular HSI data sets with limited training samples demonstrate the superior performance of the proposed method compared with other state-of-the-art methods. The source code is available at https://github.com/lms-07/AMS-M2ESL.

A fatigue assessment method based on attention mechanism and surface electromyography

Surface electromyography (sEMG) signals can be used to quantitatively assess muscle fatigue, thereby directly and objectively reflecting the functional state of neuromuscular activity. Effective fatigue diagnosis can prevent muscle damage, thereby improving the safety of rehabilitation exercise. Traditional fatigue diagnosis has certain limitations, including strong subjectivity and poor accuracy. This paper designs a sEMG signals acquisition circuit and collects the sEMG signals of the upper limb biceps brachii and triceps brachii in the force-relaxation state in a dual-channel form. Muscle fatigue classification assessment using Dynamic Time Warping-K Nearest Neighbor (DTW-KNN) and three deep learning algorithms. The experimental results show that compared with traditional machine learning algorithms, deep learning algorithm can achieve higher accuracy and time efficiency. In addition, this study introduces an attention mechanism to dynamically and reasonably assign network weights to achieve high level feature learning. The Attention-Long Short-Term Memory (Attention Based LSTM) neural network achieves 93.5% assessment accuracy with a time overhead of only 3.73s, allowing for real-time assessment of muscle fatigue.

Genetic Programming for High-Level Feature Learning in Crop Classification

Information on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are critical to the accuracy of crop maps. Although several approaches are available for extracting spectral, temporal, and phenological features for crop identification, these methods depend heavily on domain knowledge and human experiences, adding uncertainty to the final crop classification. This study proposed a novel Genetic Programming (GP) approach to learning high-level features from time series images for crop classification to address this issue. We developed a new representation of GP to extend the GP tree’s width and depth to dynamically generate either fixed or flexible informative features without requiring domain knowledge. This new GP approach was wrapped with four classifiers, i.e., K-Nearest Neighbor (KNN), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM), and was then used for crop classification based on MODIS time series data in Heilongjiang Province, China. The performance of the GP features was compared with the traditional features of vegetation indices (VIs) and the advanced feature learning method Multilayer Perceptron (MLP) to show GP effectiveness. The experiments indicated that high-level features learned by GP improved the classification accuracies, and the accuracies were higher than those using VIs and MLP. GP was more robust and stable for diverse classifiers, different feature numbers, and various training sample sets compared with classification using VI features and the classifier MLP. The proposed GP approach automatically selects valuable features from the original data and uses them to construct high-level features simultaneously. The learned features are explainable, unlike those of a black-box deep learning model. This study demonstrated the outstanding performance of GP for feature learning in crop classification. GP has the potential of becoming a mainstream method to solve complex remote sensing tasks, such as feature transfer learning, image classification, and change detection.

OneSketch: learning high-level shape features from simple sketches

OneSketch: learning high-level shape features from simple sketches

A Novel Approach for UAV Image Crack Detection.

Cracks are the most significant pre-disaster of a road, and are also important indicators for evaluating the damage level of a road. At present, road crack detection mainly depends on manual detection and road detection vehicles, with which the safety of detection workers is not guaranteed and the detection efficiency is low. A road detection vehicle can speed up the efficiency to a certain extent, but the automation level is low and it is easy to block the traffic. Unmanned Aerial Vehicles (UAV) have the characteristics of low energy consumption and easy control. If UAV technology can be applied to road crack detection, it will greatly improve the detection efficiency and produce huge economic benefits. In order to find a way to apply UAV to road crack detection, we developed a new technique for road crack detection based on UAV pictures, called DenxiDeepCrack, which is a trainable deep convolutional neural network for automatic crack detection which utilises learning high-level features for crack representation. In addition, we create a new dataset based on drone images called UCrack 11 to enrich the crack database of drone images for future crack detection research.

KiU-Net: Overcomplete Convolutional Architectures for Biomedical Image and Volumetric Segmentation.

Most methods for medical image segmentation use U-Net or its variants as they have been successful in most of the applications. After a detailed analysis of these "traditional" encoder-decoder based approaches, we observed that they perform poorly in detecting smaller structures and are unable to segment boundary regions precisely. This issue can be attributed to the increase in receptive field size as we go deeper into the encoder. The extra focus on learning high level features causes U-Net based approaches to learn less information about low-level features which are crucial for detecting small structures. To overcome this issue, we propose using an overcomplete convolutional architecture where we project the input image into a higher dimension such that we constrain the receptive field from increasing in the deep layers of the network. We design a new architecture for im- age segmentation- KiU-Net which has two branches: (1) an overcomplete convolutional network Kite-Net which learns to capture fine details and accurate edges of the input, and (2) U-Net which learns high level features. Furthermore, we also propose KiU-Net 3D which is a 3D convolutional architecture for volumetric segmentation. We perform a detailed study of KiU-Net by performing experiments on five different datasets covering various image modalities. We achieve a good performance with an additional benefit of fewer parameters and faster convergence. We also demonstrate that the extensions of KiU-Net based on residual blocks and dense blocks result in further performance improvements. Code: https://github.com/jeya-maria-jose/KiU-Net-pytorch.

A joint learning Im-BiLSTM model for incomplete time-series Sentinel-2A data imputation and crop classification

Multi-temporal deep learning approaches can make full use of crop growth patterns and phenological characteristics, resulting in excellent crop classification performance in large areas. However, obtaining complete time-series remote sensing images during the growing season is challenging due to cloud contamination. Hence, given time-series multispectral data, it is important to impute missing data and accurately classify crops. A novel Imputation-BiLSTM model (Im-BiLSTM) was developed based on Bidirectional Long Short-term Memory network (BiLSTM) to jointly perform missing data imputation and crop classification. The Im-BiLSTM model regards missing data as network variables, which are efficiently updated during backpropagation. Im-BiLSTM treats the interaction between imputation and classification tasks, reducing the error and uncertainty caused by the separation operation of imputation to classification. Furthermore, we improved the interpretability of the Im-BiLSTM model by evaluating the importance of input features and visualizing hidden state units. In Shawan County, Xinjiang, China, we acquired a total of 10 Sentinel-2A images from April to October 2016, of which 3 images lost partial data due to cloud cover. The Im-BiLSTM model was applied to incomplete time-series data containing 10 time-steps for pixel-level crop classification, and the BiLSTM model was constructed based on cloud-free images for comparison. The performance of the proposed model was tested in four different cases of images missing and missing rates. The results showed that the classification of the Im-BiLSTM model outperformed the BiLSTM model, the overall accuracy was improved by a maximum of 4.2%, and the F1-scores of spring corn and tomato was improved by a maximum of 16.1% and 21.4%, respectively. Therefore, the Im-BiLSTM model can effectively improve classification performance by jointly imputing missing data. The imputation results (the coefficient of determination values range 0.4 ∼ 0.9) indicated that the bands with the larger contribution to the classification had higher imputation accuracy. Feature importance evaluation showed that the Im-BiLSTM model captured the key phenological periods and features from time-series input. The visualization of hidden units demonstrated that the Im-BiLSTM model accumulated useful information over time, and the learned high-level features made the crops more separable than the original inputs.

SVseg: Stacked Sparse Autoencoder-Based Patch Classification Modeling for Vertebrae Segmentation

Precise vertebrae segmentation is essential for the image-related analysis of spine pathologies such as vertebral compression fractures and other abnormalities, as well as for clinical diagnostic treatment and surgical planning. An automatic and objective system for vertebra segmentation is required, but its development is likely to run into difficulties such as low segmentation accuracy and the requirement of prior knowledge or human intervention. Recently, vertebral segmentation methods have focused on deep learning-based techniques. To mitigate the challenges involved, we propose deep learning primitives and stacked Sparse autoencoder-based patch classification modeling for Vertebrae segmentation (SVseg) from Computed Tomography (CT) images. After data preprocessing, we extract overlapping patches from CT images as input to train the model. The stacked sparse autoencoder learns high-level features from unlabeled image patches in an unsupervised way. Furthermore, we employ supervised learning to refine the feature representation to improve the discriminability of learned features. These high-level features are fed into a logistic regression classifier to fine-tune the model. A sigmoid classifier is added to the network to discriminate the vertebrae patches from non-vertebrae patches by selecting the class with the highest probabilities. We validated our proposed SVseg model on the publicly available MICCAI Computational Spine Imaging (CSI) dataset. After configuration optimization, our proposed SVseg model achieved impressive performance, with 87.39% in Dice Similarity Coefficient (DSC), 77.60% in Jaccard Similarity Coefficient (JSC), 91.53% in precision (PRE), and 90.88% in sensitivity (SEN). The experimental results demonstrated the method’s efficiency and significant potential for diagnosing and treating clinical spinal diseases.