LDANet: Automatic lung parenchyma segmentation from CT images

Chen, Ying; Feng, Longfeng; Zheng, Cheng; Zhou, Taohui; Liu, Lan; Liu, Pengfei; Chen, Yi

doi:10.1016/j.compbiomed.2023.106659

Cited by 47 publications

(1 citation statement)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The key limitation of the Convolutional neural model is that it doesn't take the previous layer output as the input of the following layer, which has tended to cause less prediction score. So, the serial section in the convolutional neural model [ 33 ] was introduced for serially making the functional operation. Compared to the traditional Convolutional neural system, this serial section convolutional model afforded the finest outcome.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

Vijayanandh

Shenbagavalli²

2023

New Gener. Comput.

View full text Add to dashboard Cite

Nowadays, COVID severity prediction has attracted widely in medical research because of the disease severity. Hence, the image processing application is also utilized to analyze COVID severity identification using lungs X-ray images. Thus, several intelligent schemes were employed to detect the COVID-affected part of the lungs X-ray images. However, the traditional neural approaches reported less severity classification accuracy due to the image complexity score. So, the present study has presented a novel chimp-based Adaboost Severity Analysis (CbASA) implemented in the MATLAB environment. Hence, the lung's X-ray images are utilized to test the working performance of the designed model. All public imaging data sources contain more noisy features, so the noise features are removed in the initial hidden layer of the novel CbASA then the noise-free data is imported into the classification phase. Feature extraction, segmentation, and severity specification have been performed in the classification layer. Finally, the performance of the classification score has been measured and compared with other models. Subsequently, the presented novel CbASA has earned the finest classification outcome.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

Vijayanandh

Shenbagavalli²

2023

New Gener. Comput.

View full text Add to dashboard Cite

show abstract

A Resampling Ant Colony Optimization with Elite Exploration and Convergence Mechanism for Multithreshold Segmentation of Breast Cancer Images

Wang,

Zhao,

Heidari

et al. 2024

Advanced Intelligent Systems

View full text Add to dashboard Cite

Previous studies have emphasized the potential of threshold image segmentation for early breast cancer detection. However, traditional methods encounter challenges regarding low segmentation efficiency and accuracy. Addressing this, the ant colony optimization algorithm for continuous optimization (ACOR) shows promise. Yet, existing ACOR variants still grapple with poor initial population quality, affecting convergence speed and avoiding local optimization. These issues impact segmentation efficiency and accuracy. To tackle them, this study introduces RESACO, an enhanced ACOR version integrating three novel optimization strategies: resampling initialization (RIS), elite exploration (EES), and strengthened convergence mechanism (SCM). RIS enhances initial population quality by resampling regions with individuals demonstrating superior fitness and segmentation efficiency. EES promotes exploration across the search space, preventing local optima entrapment and enhancing model stability. SCM expediting convergence, segmentation efficiency, and precision. RESACO's performance is assessed through extensive experiments using IEEE CEC 2014 and IEEE CEC 2022 benchmark functions, including ablation experiments and comparisons with basic and improved algorithms and ACOR variants. Subsequently, the threshold image segmentation model based on RESACO is compared with other models using metaheuristic algorithms for segmenting realistic breast cancer medical images. Results demonstrate the proposed model's faster convergence and higher segmentation accuracy, preserving more lesion tissue details.

show abstract

Chest X-ray Images for Lung Disease Detection Using Deep Learning Techniques: A Comprehensive Survey

Al-qaness,

Zhu,

AL-Alimi

et al. 2024

Arch Computat Methods Eng

View full text Add to dashboard Cite

LDANet: Automatic lung parenchyma segmentation from CT images

Cited by 47 publications

References 37 publications

A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

A Resampling Ant Colony Optimization with Elite Exploration and Convergence Mechanism for Multithreshold Segmentation of Breast Cancer Images

Chest X-ray Images for Lung Disease Detection Using Deep Learning Techniques: A Comprehensive Survey

Contact Info

Product

Resources

About