Article
Mathematical & Computational Biology
Xiao Zou, Jintao Zhai, Shengyou Qian, Ang Li, Feng Tian, Xiaofei Cao, Runmin Wang
Summary: The study proposes a novel approach for improved tumor classification in breast ultrasound images through a dual-input model and global average pooling (GAP)-guided attention loss function. This method uses a convolutional neural network with a transformer architecture and modifies the single-input model for dual-input, enhancing the extraction of effective features from the target region and mitigating the impact of information loss or redundancy on misclassification.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaofeng Li, Yupeng Sang, Xianmin Ma, Yingjie Cai
Summary: The quantitative feature classification of breast ultrasound images based on improved Naive Bayes algorithm is proposed. Histogram features, grayscale co-generation matrix, and other texture features are extracted from the ultrasound image. NB combined with decision tree calculation to improve the classification accuracy of ultrasound images. The results show that the proposed algorithm achieves high accuracy in lesion recognition and classification of breast ultrasound images, with short classification time and application value in breast ultrasound diagnosis.
IET IMAGE PROCESSING
(2023)
Article
Computer Science, Interdisciplinary Applications
Zhemin Zhuang, Zengbiao Yang, Alex Noel Joseph Raj, Chuliang Wei, Pengcheng Jin, Shuxin Zhuang
Summary: This study proposed a reliable classification method for breast ultrasound images of tumors by combining image decomposition and fusion techniques with adaptive spatial feature fusion technology. The experimental results showed excellent performance in various metrics. The research has the potential to automate breast cancer detection and has strong clinical significance.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Computer Science, Information Systems
Qiang Liu, Zhong Liu, Wencong Xu, Huiying Wen, Ming Dai, Xin Chen
Summary: In this study, a deep learning model DCNN-MFF was proposed for liver fibrosis recognition by extracting and integrating features from ultrasound B-mode image and Nakagami parametric map, achieving superior performance compared to comparative methods.
Article
Computer Science, Artificial Intelligence
Mostafa Khojastehnazhand, Mozaffar Roostaei
Summary: This study used a machine vision system and texture feature extraction methods to classify seven varieties of wheat in the East Azerbaijan Province of Iran. By utilizing unsupervised and supervised methods, along with feature extraction, the different wheat varieties were identified with over 95% accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Acoustics
Ahmad Chowdhury, Rezwana R. Razzaque, Sabiq Muhtadi, Ahmad Shafiullah, Ehsan Ul Islam Abir, Brian S. Garra, S. Kaisar Alam
Summary: In this study, the potential of Nakagami parametric images for non-invasive classification of breast lesions and characterization of breast tissue was investigated. The optimal feature set acquired from Nakagami parametric images generated using a column window of length 0.75 mm provided the best results for characterizing breast lesions, with a high classification accuracy of 93.08% and a low false positive rate of 8.65%.
Article
Mathematics, Interdisciplinary Applications
Gang Xiong, Ziqin Xiong, Liqiong Jia, Trieu-Kien Truong
Summary: In this paper, the point-wise multifractal spectrum (MFS) of ultrasonic images in the time-frequency domain is studied. The spatial multifractal spectrum (SMFS) is proposed as an extension of the conventional MFS to analyze and characterize spatial multiscale multifractal features. The SMFS is derived from the two-dimensional multifractal spectrum and the Pseudo Wigner-Ville distribution (PWVD), and it allows for robust estimation of SMFS features under low SNR conditions. Furthermore, a novel BUSI classification method based on SMFS and deep learning network achieves a significantly improved classification accuracy compared to the state-of-the-art method.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Xiaolei Qu, Hongyan Lu, Wenzhong Tang, Shuai Wang, Dezhi Zheng, Yaxin Hou, Jue Jiang
Summary: In this study, a visual geometry group attention ViT (VGGA-ViT) network was proposed for breast ultrasound (BUS) image classification. By combining convolutional neural network (CNN) and vision transformer (ViT), the classification accuracy and objectivity were improved.
Article
Engineering, Electrical & Electronic
Xiaochun Xu, Yibing Li, Q. M. Jonathan Wu
Summary: This paper introduces a compact multi-pattern encoding descriptor for visual texture classification by developing local extremum and local center patterns, achieving state-of-the-art performance in texture classification.
DIGITAL SIGNAL PROCESSING
(2021)
Article
Computer Science, Information Systems
Kushangi Atrey, Bikesh Kumar Singh, Narendra Kuber Bodhey
Summary: This paper presents a novel semi-automated multimodal classification system for breast tumors, which combines features from mammogram and ultrasound images to improve classification accuracy. The experimental results demonstrate higher accuracy compared to single modality methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Sabiq Muhtadi, Rezwana R. Razzaque, Ahmad Chowdhury, Brian S. Garra, S. Kaisar Alam
Summary: This study evaluates the texture quantification of ultrasound Nakagami parametric images for the non-invasive characterization of breast tumors. The results show that the texture features obtained from specific window formations can be effectively used for breast lesion characterization.
JOURNAL OF MEDICAL IMAGING
(2023)
Article
Engineering, Electrical & Electronic
Tiecheng Song, Jie Feng, Lin Luo, Chenqiang Gao, Hongliang Li
Summary: In this paper, two novel operators, local grouped order pattern (LGOP) and non-local binary pattern (NLBP), are proposed for texture description. Experimental results demonstrate that combining LGOP and NLBP to construct discriminative histogram features as texture descriptor LGONBP shows superiority over state-of-the-art LBP variants for texture classification.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Article
Materials Science, Paper & Wood
Xiaoxia Yang, Hailan Jiang, Lixin Ma, Wenhu Yang, Xiaohan Zhao, Cuiping Hu, Zhedong Ge
Summary: This study used 19 high-value hardwood species (Papilionaceae, Ebenaceae, and Caesalpiniaceae) with similar microstructure as experimental objects. Images of transverse sections, radial sections, and tangential sections were collected using Micro CT. Local binary patterns (LBP) were used for feature extraction, and three fusion features were formed by fusing LBP deformed forms with Gray-Level Co-Occurrence Matrix (GLCM). The fusion features were combined with support vector machine (SVM) or BP neural network for wood classification. The texture feature fusion method showed better performance than single feature classification, with the best effect observed for the fusion of uniform LBP and GLCM, combined with SVM for the highest classification accuracy. The evaluation of 19 hardwood species mainly relied on transverse sections, which had higher accuracy than radial and tangential sections. Therefore, the classification results of transverse sections should be considered as the main evaluation basis for the classification of the 19 high-value hardwood species.
Article
Computer Science, Software Engineering
Bin Li, Yibing Li, Q. M. Jonathan Wu
Summary: This paper proposes a multi-scale threshold integration encoding strategy for texture classification, which introduces multi-scale local texture information and combines multiple sub-patterns to generate a distinctive and powerful texture descriptor.
Article
Computer Science, Interdisciplinary Applications
Xiaohui Di, Shengzhou Zhong, Yu Zhang
Summary: In this paper, a saliency map-guided hierarchical dense feature aggregation framework is proposed for breast lesion classification using breast ultrasound images. By generating saliency maps and constructing a triple-branch network, this method can extract and fuse features from foreground and background, thus improving the performance of breast lesion diagnosis.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Article
Materials Science, Biomaterials
Pei-Chen Li, Szu-Ching Chen, Yi-Jen Hsueh, Yang-Chun Shen, Meng-Yu Tsai, Li-Wen Hsu, Chih-Kuang Yeh, Hung-Chi Chen, Chieh-Cheng Huang
Summary: Corneal transplantation is the primary method for treating corneal blindness, but cell-based strategies utilizing CECs grown on biomaterials show promise as alternative therapies. This study focuses on developing a scaffold system that supports CEC function and prevents post-transplant complications through sustained release of curcumin, improving long-term engraftment success. Multiple experiments demonstrate the effectiveness of the Cur@MPs-loaded gelatin scaffold in promoting CEC proliferation and protecting against oxidative stress, inflammation, and angiogenesis.
MATERIALS SCIENCE AND ENGINEERING C-MATERIALS FOR BIOLOGICAL APPLICATIONS
(2021)
Article
Multidisciplinary Sciences
Wei-Chen Lo, Ching-Hsiang Fan, Yi-Ju Ho, Chia-Wei Lin, Chih-Kuang Yeh
Summary: The AVT technique, inspired by tornadoes, successfully captures and increases the local concentration of MBs. This compact, easy-to-use, and biocompatible method enables systemic drug administration with extremely low doses.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Medicine, General & Internal
Wei-Jung Chang, Yee-Chun Chen, Chen-Yang Hsu, Chih-Dao Chen, Sam Li-Sheng Chen, King-Jen Chang
Summary: The study found variations in PRPA of COVID-19 and its impact on death rates across different countries. By observing the changes in PRPA of COVID-19 at different stages, improvements in clinical management and medical capacity can be evaluated.
JOURNAL OF THE FORMOSAN MEDICAL ASSOCIATION
(2021)
Article
Biology
Chia-Hsin Wu, Chia-Shan Hsieh, Yo-Cheng Chang, Chi-Cheng Huang, Hsien-Tang Yeh, Ming-Feng Hou, Yuan-Chiang Chung, Shih-Hsin Tu, King-Jen Chang, Amrita Chattopadhyay, Liang-Chuan Lai, Tzu-Pin Lu, Yung-Hua Li, Mong-Hsun Tsai, Eric Y. Chuang
Summary: The study reveals the different effects of WGD on breast cancer subtypes and highlights the genomic association between WGD and homologous recombination deficiency in Taiwanese patients. Triple-negative breast cancer shows higher WGD and chromosomal instability, while HER2 + tumors exhibit early WGD events and varied CIN levels.
COMMUNICATIONS BIOLOGY
(2021)
Article
Physics, Applied
Zong-Han Hsieh, Ching-Hsiang Fan, Meng-Lin Li, Chih-Kuang Yeh
Summary: This study proposes a method to improve the local light intensity in scattering media by generating transient microbubbles using ultrasound and super hydrophobic polytetrafluoroethylene nanoparticles (PTFE NPs). Results show that only microbubbles of the proper size can effectively increase light propagation.
APPLIED PHYSICS LETTERS
(2022)
Article
Oncology
Po-Han Lin, Shin-Cheh Chen, Ling-Ming Tseng, King-Jen Chang, Ai-Chu Huang, Kuo-Chih Cheng, Karen Yang, Hui-Chen Wu, Tsu-Yi Chao, Yuan-Ching Chang, Peng-Chan Lin, Wen-Hung Kuo, Wen-Lin Kuo, Ching-Hung Lin, Huo-Mu Chen, Dah-Cherng Yeh, Liang-Chih Liu, Chun-Yu Liu, Ming-Yang Wang, Chiao Lo, Yen-Shen Lu, Chiun-Sheng Huang
Summary: The study found that Asian breast cancer patients with BRCA mutations had a higher risk of developing contralateral breast cancer, especially for BRCA1 mutation carriers with triple-negative breast cancer as their first breast cancer. However, there were no significant differences in relapse-free survival and overall survival between mutation carriers and control patients.
BREAST CANCER RESEARCH AND TREATMENT
(2022)
Review
Pharmacology & Pharmacy
Nan Wu, Ching-Hsiang Fan, Chih-Kuang Yeh
Summary: Despite the challenges in cancer treatment, sonodynamic therapy (SDT) shows promising potential as a new approach. This review provides an understanding of the mechanisms, efficacy, and future directions of SDT-based applications in tumor therapy.
DRUG DISCOVERY TODAY
(2022)
Article
Pharmacology & Pharmacy
Yi-Ju Ho, Dinh Thi Thao, Chih-Kuang Yeh
Summary: Hypoxia-induced drug resistance is a critical issue in cancer therapy, and this study found that the use of oxygen-loaded microbubbles can overcome this resistance by promoting drug uptake and reoxygenation, ultimately enhancing the treatment outcome.
Review
Pharmacology & Pharmacy
Ching-Hsiang Fan, Yi-Ju Ho, Chia-Wei Lin, Nan Wu, Pei-Hua Chiang, Chih-Kuang Yeh
Summary: In this article, the authors summarized recent progress on US-responsible drug carriers, focusing on the routes of enhancing delivery and applications. They discussed three specific mechanisms that enhance in vivo drug delivery via US with US-responsible carriers and reviewed the state-of-the-art materials for US-triggered drug delivery.
EXPERT OPINION ON DRUG DELIVERY
(2022)
Article
Biochemistry & Molecular Biology
Hao-Chih Tai, Kuen-Yuan Chen, Ming-Hsun Wu, King-Jen Chang, Chiung-Nien Chen, Argon Chen
Summary: This study validates the accuracy of computerized sonographic features (CSF) in thyroid nodules imaging and assesses how the reading performance of clinicians can be improved with the assistance of CSF. The study finds that the sequential reading performance of clinicians is significantly improved with CSF, indicating the importance of computerized features in assisting sonographic diagnosis.
Article
Medicine, Research & Experimental
Pei-Hua Chiang, Ching-Hsiang Fan, Qiaofeng Jin, Chih-Kuang Yeh
Summary: By coating the surface of ADS NPs with a superhydrophobic layer and modifying it with a phospholipid layer, it is possible to overcome the challenge of insufficient drug dose in chemotherapy. The release of drugs can be triggered by ultrasound, enhancing the fluorescence intensity in the tumor region.
MOLECULAR PHARMACEUTICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Chin-Kuo Chen, Yan-Heng Lai, Li-Chun Hsieh, Po-Hsiang Tsui
Summary: This study developed a new ultrasound signature, the multiregional-weighted Nakagami parameter (MNP), for assessing the severity and fluid properties of middle ear effusion (MEE) in pediatric patients. The MNP values were estimated through multiregional backscattering measurements of the mastoid, and compared with clinical examination results. The MNP was found to be capable of detecting MEE, assessing effusion severity, and potentially characterizing fluid properties.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Acoustics
Ching-Hsiang Fan, Nan Wu, Chih-Kuang Yeh
Summary: Sonodynamic therapy using carbon dots microbubbles as sonosensitizers with ultrasound irradiation effectively induced lipid peroxidation, intracellular ROS elevation, and apoptosis in cancer cells, suggesting its potential as a novel strategy for tumor therapy.
ULTRASONICS SONOCHEMISTRY
(2023)
Article
Cell & Tissue Engineering
Thi-Nhan Phan, Ching-Hsiang Fan, Chih-Kuang Yeh
Summary: Pluripotent stem cell therapy is an important regenerative approach for treating diseases, but it faces challenges such as low cell transplantation efficiency and uncontrollable differentiation. Ultrasound (US) has shown potential in enhancing stem cell therapy by promoting proliferation, differentiation, migration, and in vivo detection of stem cells. Combining US with US-responsive particles can provide enhanced physical-chemical stimulation and drug delivery. In-depth evaluation of the role and mechanism of US in stem cell therapy can improve understanding and encourage further research in this field.
STEM CELL REVIEWS AND REPORTS
(2023)