Article
Chemistry, Multidisciplinary
Yun-ji Kim, Hyun Chin Cho, Hyun-chong Cho
Summary: A deep learning-based computer-aided diagnosis (CADx) system was proposed to classify gastroscopy images as normal or abnormal, with data augmentation and deep convolutional generative adversarial networks (DCGAN) used to improve training. The combination of these methods achieved the best performance in accuracy, effectively solving the medical-data problem and enhancing the accuracy of gastric disease diagnosis.
APPLIED SCIENCES-BASEL
(2021)
Article
Oncology
Huan Zheng, Zebin Xiao, Siwei Luo, Suqing Wu, Chuxin Huang, Tingting Hong, Yan He, Yanhui Guo, Guoqing Du
Summary: This study develops a computer aided diagnosis (CAD) system using deep learning to assist radiologists in diagnosing follicular thyroid carcinoma (FTC) on thyroid ultrasonography. The CAD system achieves better performance than radiologists and significantly improves their diagnosis of FTC. It provides a reliable reference for preoperative diagnosis of FTC and may assist in the development of a fast, accessible screening method for FTC.
FRONTIERS IN ONCOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Alvaro S. Hervella, Jose Rouco, Jorge Novo, Marcos Ortega
Summary: This study proposes a self-supervised learning method using unlabeled multimodal data to enhance the accuracy of retinal computer-aided diagnosis systems, without relying on manual annotation. Experimental results demonstrate satisfactory performance in diagnosing different ocular diseases, showcasing the potential of leveraging unlabeled multimodal visual data in the medical field.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Jung-Woo Chae, Hyun-Chong Cho
Summary: Gastric cancer is a high-risk cancer with a high incidence and mortality rate, but early diagnosis greatly improves survival chances. Gastroscopy is a reliable method for diagnosing gastric cancer and precancerous lesions, but it is subjective and can be influenced by specialist factors. To address this, we propose a computer-aided diagnosis system using a deep-learning model called Vision Transformer to classify healthy tissue, gastric lesions, and early gastric cancer. We also introduce the Multi-Filter AutoAugment method to enhance the model's classification performance with limited medical data. Experimental results show high accuracy in distinguishing between abnormalities and healthy tissue, as well as early gastric cancer and non-cancerous gastric lesions.
Article
Computer Science, Information Systems
Farnoosh Azour, Azzedine Boukerche
Summary: Breast cancer is the second most deadly cancer among women, but can be prevented through early detection. Researchers have developed Computer-Aided Diagnosis (CADx) systems, with the use of Deep Learning and Convolutional Neural Networks (CNNs) revolutionizing their development. By integrating the state-of-the-art pre-trained model EfficientNet with other models and applying ensemble learning, significant improvements in accuracy have been achieved.
Article
Computer Science, Information Systems
Boyu Zhang, Aleksandar Vakanski, Min Xian
Summary: This paper proposes a new machine learning method called BI-RADS-Net-V2 for breast cancer diagnosis using explainable artificial intelligence (XAI). The method can accurately distinguish malignant tumors from benign ones and provides clinically proven morphological features used for diagnosis in the Breast Imaging Reporting and Data System (BI-RADS). Experimental results on 1,192 Breast Ultrasound images demonstrate that the proposed method improves diagnostic accuracy by leveraging medical knowledge in BI-RADS and providing both semantic and quantitative explanations.
Review
Medicine, General & Internal
Xuejiao Pang, Zijian Zhao, Ying Weng
Summary: This study discussed the application prospects of CAD system based on deep learning in the medical field and its importance in clinical practice, and also proposed some prospects for future research directions.
Article
Computer Science, Hardware & Architecture
Mahmoud Ragab, Wajdi H. Aljedaibi, Alaa F. Nahhas, Ibrahim R. Alzahrani
Summary: This paper presents a deep learning-based computer aided diagnosis model for diabetic retinopathy detection and grading. The model includes preprocessing, image segmentation, feature extraction, and classification, and achieves high accuracy according to the performance validation on benchmark data set.
COMPUTERS & ELECTRICAL ENGINEERING
(2022)
Review
Oncology
Xin Yu Liew, Nazia Hameed, Jeremie Clos
Summary: Early detection and timely treatment of breast cancer can reduce the risk of death, with histopathology images and CAD systems being key technologies. Machine learning methods are increasingly applied in diagnosing breast cancer, helping to improve accuracy.
Article
Engineering, Biomedical
Amitava Halder, Debangshu Dey
Summary: Lung cancer is a life-threatening cancer and detecting lung nodules using CT images helps in early recognition. Different computer-aided algorithms can increase the survival rate of lung cancer patients. However, manually recognizing malignant nodules from benign ones is difficult. Deep learning-based CADx systems have been developed for nodule characterization.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Biotechnology & Applied Microbiology
Ahmed M. Zaalouk, Gamal A. Ebrahim, Hoda K. Mohamed, Hoda Mamdouh Hassan, Mohamed M. A. Zaalouk
Summary: In this paper, a computer-aided diagnosis system based on deep learning is developed to assist pathologists in diagnosing breast cancer. Multiple pre-trained convolutional neural network models are analyzed and tested, and a new approach for transfer learning is introduced. The experimental results show that the Xception model performs the best among the tested models.
BIOENGINEERING-BASEL
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Hyo Min Lee, Young Jae Kim, Je Bok Cho, Ji Young Jeon, Kwang Gi Kim
Summary: In this study, an automated computer-aided diagnosis (CAD) tool based on deep learning was developed to measure the sagittal alignment of the spine from X-ray images. The CAD system achieved high accuracy in segmenting the spine and measuring the thoracic kyphosis and lumbar lordosis angles. The performance of the CAD algorithm was verified using various analysis methods and showed high similarity and reliability. Therefore, CAD can assist clinicians in diagnosing spinal curvatures while reducing observer-based variability and required time or effort.
JOURNAL OF DIGITAL IMAGING
(2022)
Article
Computer Science, Interdisciplinary Applications
Rebecca Sawyer Lee, Jared A. Dunnmon, Ann He, Siyi Tang, Christopher Re, Daniel L. Rubin
Summary: This study compared machine learning methods for classifying mass lesions on mammography images and found that a common segmentation-free CNN model substantially outperforms other methods. This indicates that representation learning techniques are advantageous for mammogram analysis.
JOURNAL OF BIOMEDICAL INFORMATICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Chengze Zhu, Pingge Hu, Xingtong Wang, Xianxu Zeng, Li Shi
Summary: This study developed a deep learning-based CAD method that can recognize hydrops lesions in real-time, providing a robust and accurate solution with powerful feature extraction and segmentation capabilities for auxiliary diagnosis of hydatidiform mole.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2023)
Article
Gastroenterology & Hepatology
Sabrina Xin Zi Quek, Jonathan W. J. Lee, Zhu Feng, Min Min Soh, Masayuki Tokano, Yeoh Khay Guan, Jimmy B. Y. So, Tomohiro Tada, Calvin J. Koh
Summary: The study aimed to validate an AI-based endoscopic system in a Singaporean cohort. The results showed that the AI system was comparable to human endoscopists in diagnostic accuracy for static images and had shorter diagnostic time. AI may have a larger role in augmenting human diagnosis during endoscopy.
JOURNAL OF GASTROENTEROLOGY AND HEPATOLOGY
(2023)