A Comprehensive Review on Analysing of Brain Signals Using Different Clustering Methods DOI
Palanichamy Naveen,

T. Akilan,

P. Manikandan

et al.

Published: Sept. 20, 2023

Classification and clustering are crucial tasks in analyzing brain signals, which can be broadly categorized into two main methods: invasive non-invasive. Invasive techniques involve placing electrodes directly inside or on the surface of to measure activity, whereas non-invasive methods activity without need for procedures. The latter includes EEG, MEG, fMRI, PET, NIRS. Brain signals classified based type being measured, such as brainwaves, evoked potentials, event-related functional imaging. This classification help researchers better understand underlying mechanisms function develop new diagnosing treating neurological disorders. major divisions include hard clustering, soft density-based model-based hierarchical subspace clustering. In each signal is assigned a single cluster similarity centroid cluster, while assigns probability belonging degree centroids clusters. Density-based regions high density feature space, probabilistic model that describes distribution data, done manner. Subspace subspaces space. These different approaches used combination achieve results depending characteristics data research question at hand. Overall, essential advancing our understanding developing

Language: Английский

Efficient artificial intelligence approaches for medical image processing in healthcare: comprehensive review, taxonomy, and analysis DOI Creative Commons
Omar Abdullah Murshed Farhan Alnaggar, Basavaraj N Jagadale, Mufeed Ahmed Naji Saif

et al.

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(8)

Published: July 29, 2024

Abstract In healthcare, medical practitioners employ various imaging techniques such as CT, X-ray, PET, and MRI to diagnose patients, emphasizing the crucial need for early disease detection enhance survival rates. Medical Image Analysis (MIA) has undergone a transformative shift with integration of Artificial Intelligence (AI) Machine Learning (ML) Deep (DL), promising advanced diagnostics improved healthcare outcomes. Despite these advancements, comprehensive understanding efficiency metrics, computational complexities, interpretability, scalability AI based approaches in MIA is essential practical feasibility real-world environments. Existing studies exploring applications lack consolidated review covering major stages specifically focused on evaluating approaches. The absence structured framework limits decision-making researchers, practitioners, policymakers selecting implementing optimal healthcare. Furthermore, standardized evaluation metrics complicates methodology comparison, hindering development efficient This article addresses challenges through review, taxonomy, analysis existing AI-based taxonomy covers image processing stages, classifying each stage method further analyzing them origin, objective, method, dataset, reveal their strengths weaknesses. Additionally, comparative conducted evaluate over five publically available datasets: ISIC 2018, CVC-Clinic, 2018 DSB, DRIVE, EM terms accuracy, precision, Recall, F-measure, mIoU, specificity. popular public datasets are briefly described analyzed. resulting provides landscape facilitating evidence-based guiding future research efforts toward scalable meet current needs.

Language: Английский

Citations

11
Comprehensive Review on MRI-Based Brain Tumor Segmentation: A Comparative Study from 2017 Onwards DOI
Amit Verma, Shiv Naresh Shivhare, Shailendra Pratap Singh

et al.

Archives of Computational Methods in Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: May 20, 2024

Language: Английский

Citations

7
Al-enabled properties distribution prediction for high-pressure die casting Al-Si alloy DOI

Yu-Tong Yang,

Zhong-Yuan Qiu,

Zhen Zheng

et al.

Advances in Manufacturing, Journal Year: 2024, Volume and Issue: 12(3), P. 591 - 602

Published: March 17, 2024

Language: Английский

Citations

3
Lesion Classification of Coronary Artery CTA Images Based on CBAM and Transfer Learning DOI
Yongze Jin, Xin Ye, Nan Feng

et al.

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 14

Published: Jan. 1, 2024

Classification of coronary artery stenosis is essential in assisting physicians diagnosing cardiovascular diseases. However, due to the complexity medical diagnosis and confidentiality images, it difficult obtain many image samples for scientific research general. In addition, degree, location, morphology different patients, as well noise CT angiography (CTA) make challenging extract typing features effectively. To address above problems, firstly, a joint segmentation method proposed based on maximum between-class variance region growing key regions from CTA images facilitate further feature extraction. Then, classification model Convolutional Block Attention Module (CBAM) transfer learning constructed, which can effectively improve training effect under insufficient samples. Finally, dataset actual patients applied experimental verification. Experiment results show that accuracy up 98.99%, greatly improved compared with several machine algorithms neural network methods. It be concluded considerably improved, reasonable basis provided clinical diagnosis.

Language: Английский

Citations

3
Pneumonia Screening From Radiology Images Using Homomorphic Transformation Filter‐Based FAWT and Customized VGG‐16 DOI
Rajneesh Kumar Patel,

Ankit Choudhary,

Nancy Kumari

et al.

International Journal of Imaging Systems and Technology, Journal Year: 2025, Volume and Issue: 35(3)

Published: May 1, 2025

ABSTRACT Pneumonia, attributable to pathogens and autoimmune disorders, accounts for approximately 450 million cases annually. Chest x‐ray analysis remains the gold standard pneumonia detection, DL has revolutionized study of high‐dimensional data, including images, audio, video. This research enhances validates a CAD system distinguishing from normal health states using imaging. paper presents novel methodology that integrates CLHAE Homographic Transformation Filter‐based Flexible Analytical Wavelet Transform (HTF‐FAWT) image decomposition, enabling systematic decomposition pre‐processed input images into four distinct sub‐band across six hierarchical levels. Feature extraction employs VGG‐16 Deep Learning techniques, with extracted features subsequently classified by support vector machine Morlet, Mexican‐hat wavelet, radial basis function kernels. Employing tenfold cross‐validation, our model exhibited remarkable classification performance, achieving an accuracy 97.51%, specificity 97.77%, sensitivity 96.5% in spotting via x‐rays. The utility feature maps Grad‐CAM highlighting critical regions accurate prediction was confirmed, offering visual validation model's efficacy. Statistical examinations validate superior performance proposed framework, demonstrating its potential as expedient diagnostic tool medical imaging specialists rapidly detecting pneumonia. It demonstrates effectiveness various classifiers classification, method outperforming state‐of‐the‐art approaches. diagnosis high (97.51%), visualization, automated interpretation, faster, reliable screening clinical integration reducing reliance on manual assessment radiology.

Language: Английский

Citations

0
PIF-Net: A parallel interweave fusion network for knee joint segmentation DOI
Xiwang Xie,

Lijie Xie,

Xipeng Pan

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 109, P. 107967 - 107967

Published: May 9, 2025

Language: Английский

Citations

0
A Review of Brain Tumor Segmentation Using MRIs from 2019 to 2023 (Statistical Information, Key Achievements, and Limitations) DOI

Yasaman Zakeri,

Babak Karasfi,

Afsaneh Jalalian

et al.

Journal of Medical and Biological Engineering, Journal Year: 2024, Volume and Issue: 44(2), P. 155 - 180

Published: April 1, 2024

Language: Английский

Citations

3
Robust Brain Tumor Detection and Classification via Multi-Technique Image Analysis DOI Creative Commons

N Salma,

G R Madhuri,

Basavaraj N Jagadale

et al.

Physica Scripta, Journal Year: 2024, Volume and Issue: 99(7), P. 076020 - 076020

Published: June 17, 2024

Abstract Accurate detection and classification of brain tumors play a critical role in neurological diagnosis treatment.Proposed work developed sophisticated technique to precisely identify classify neoplasms medical imaging. Our approach integrates various techniques, including Otsu’s thresholding, anisotropic diffusion, modified 3-category Fuzzy C-Means (FCM) for segmentation after skull stripping wavelet transformation post-processing segmentation, Convolution neural networks classification. This not only recognizes that discriminating healthy tissue from tumor-affected areas is challenging, yet it also focuses on finding abnormalities inside early tiny tumor structures. Initial preprocessing stages improve the visibility images identification regions while accurately classifying locations into core, edema, enhancing by as well. Ultimately, these segmented zones are refined using transforms, which remove noise feature extraction. CNN architecture uses learned abstractions distinguish between malignant regions, ensuring robust It particularly good at identifying detecting anomalies provides substantial advances accurate detection. Comprehensive hypothetical evaluations validate its efficacy, could clinical diagnostics perhaps influence research treatment approaches.

Language: Английский

Citations

1
Advancing image segmentation with DBO-Otsu: Addressing rubber tree diseases through enhanced threshold techniques DOI Creative Commons
Zhenjing Xie, Jinran Wu,

Weirui Tang

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(3), P. e0297284 - e0297284

Published: March 21, 2024

Addressing the profound impact of Tapping Panel Dryness (TPD) on yield and quality in global rubber industry, this study introduces a cutting-edge Otsu threshold segmentation technique, enhanced by Dung Beetle Optimization (DBO-Otsu). This innovative approach optimizes combination accelerating convergence diversifying search methodologies. Following initial segmentation, TPD severity levels are meticulously assessed using morphological characteristics, enabling precise determination optimal thresholds for final segmentation. The efficacy DBO-Otsu is rigorously evaluated against mainstream benchmarks like Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Feature (FSIM), compared with six contemporary swarm intelligence algorithms. findings reveal that substantially surpasses its counterparts image processing speed. Further empirical analysis dataset comprising cases from level 1 to 5 underscores algorithm’s practical utility, achieving an impressive 80% accuracy identification underscoring potential recognition tasks.

Language: Английский

Citations

1
A New Approach In Metaheuristic Clustering: Coot Clustering DOI Open Access
Gökhan Kayhan, İsmail İşeri

Academic Platform Journal of Engineering and Smart Systems, Journal Year: 2024, Volume and Issue: 12(2), P. 59 - 67

Published: May 28, 2024

As a result of technological advancements, the increase in vast amounts data today's world has made artificial intelligence and mining significantly crucial. In this context, clustering process, which aims to explore hidden patterns meaningful relationships within complex datasets by grouping similar features conduct more effective analyses, holds vital importance. an alternative classical methods that face challenges such as large volumes computational complexities, metaheuristic method utilizing Coot Optimization (COOT), swarm intelligence-based algorithm, been proposed. COOT, inspired hunting stages eagles recently introduced into literature, is method. Through proposed COOT method, aim contribute literature leveraging COOT's robust exploration exploitation processes, its dynamic flexible structure. Comprehensive experimental studies were conducted evaluate consistency effectiveness COOT-based algorithm using randomly generated synthetic widely used Iris dataset literature. The same underwent analysis traditional K-Means, renowned for simplicity speed, comparative purposes. performance algorithms was assessed cluster validity measures Silhouette Global, Davies-Bouldin, Krznowski-Lai, Calinski-Harabasz indices, along with Total Squared Error (SSE) objective function. Experimental results indicate performs at competitive level K-Means shows potential, especially multidimensional real-world problems. Despite not being previously purposes, impressive some tests compared showcases success potential pioneer different aimed expanding usage domain.

Language: Английский

Citations

1