Article
Public, Environmental & Occupational Health
Jinglun Liang, Guoliang Ye, Jianwen Guo, Qifan Huang, Shaohui Zhang
Summary: Malignant pulmonary nodules are a main feature of lung cancer in early CT screening. With the application of deep learning in image processing, researchers are exploring methods to diagnose pulmonary nodules. Imbalanced datasets may lead to higher false-positives, but a filtering step in this study has shown promising results in reducing false-positives and achieving high accuracy.
FRONTIERS IN PUBLIC HEALTH
(2021)
Review
Computer Science, Information Systems
K. K. Anilkumar, V. J. Manoj, T. M. Sagi
Summary: Leukemia is a non-tumor type of cancer and early diagnosis is crucial. This study categorizes the related works on computer aided diagnosis of leukemia into Machine Learning (ML) and Deep Learning (DL) based technologies. The review found that SVM was widely used in ML based works and CNN dominated the DL category. There is a lack of works on chronic leukemia in both ML and DL categories. The study highlights the need for public datasets and new diagnostic methods for different types of leukemia, particularly chronic leukemia. The shift towards DL based studies for leukemia diagnosis is evident from 2019 onwards and there is a scarcity of reviews classifying the related works into ML and DL techniques.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Sunyi Zheng, Ludo J. Cornelissen, Xiaonan Cui, Xueping Jing, Raymond N. J. Veldhuis, Matthijs Oudkerk, Peter M. A. van Ooijen
Summary: The study aims to develop an accurate deep learning framework for nodule detection through multiplanar candidate detection and multiscale false positive reduction. The proposed system achieves good performance for lung nodule detection on the LIDC-IDRI dataset, demonstrating its effectiveness across different nodule sizes.
Article
Radiology, Nuclear Medicine & Medical Imaging
Jingwen Chen, Rong Cao, Shengyin Jiao, Yunpeng Dong, Zilong Wang, Hua Zhu, Qian Luo, Lei Zhang, Han Wang, Xiaorui Yin
Summary: This study assessed the value of a CAD system for detecting lung nodules on chest CT images. The CAD system demonstrated higher sensitivity compared to manual detection by radiologists, with only a slight increase in false positive rate.
QUANTITATIVE IMAGING IN MEDICINE AND SURGERY
(2023)
Article
Computer Science, Interdisciplinary Applications
Sozan Abdullah Mahmood, Hunar Abubakir Ahmed
Summary: Lung cancer is a critical disease with a high death rate. Early diagnosis is crucial, and in this study, a CNN-based computer-aided diagnosis system is proposed to automatically classify lung nodules into benign or malignant, achieving excellent performance.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2022)
Article
Biochemistry & Molecular Biology
Rimsha Asad, Saif Ur Rehman, Azhar Imran, Jianqiang Li, Abdullah Almuhaimeed, Abdulkareem Alzahrani
Summary: Brain tumors can cause damage to the brain and surrounding tissues, blood vessels, and nerves if not treated promptly. Early detection is crucial to prevent fatal outcomes. Manual detection is challenging due to variations in tumor characteristics, thus an automatic system using a deep convolutional neural network is proposed. The system achieved high accuracy on the brain-tumor dataset and outperformed baseline methods, demonstrating its potential for application in other diseases.
Article
Radiology, Nuclear Medicine & Medical Imaging
Haozhe Huang, Dezhong Zheng, Hong Chen, Ying Wang, Chao Chen, Lichao Xu, Guodong Li, Yaohui Wang, Xinhong He, Wentao Li
Summary: The study developed a novel multimodal data fusion model using deep learning to predict the invasiveness risk of stage I lung adenocarcinoma. The model combines CT images and clinical variables, achieving higher accuracy and performance compared to radiologists.
Article
Computer Science, Interdisciplinary Applications
Xuejiao Pang, Zijian Zhao, Yanbing Wu, Yong Chen, Jin Liu
Summary: This paper proposes a transformer and convolutional neural network-based CAD system (TransMSF) that assists endoscopists in diagnosing multiple GI diseases, with superior performance compared to other state-of-the-art models and seasoned endoscopists.
JOURNAL OF COMPUTATIONAL DESIGN AND ENGINEERING
(2023)
Article
Health Care Sciences & Services
Mustafa Ghaderzadeh, Farkhondeh Asadi, Ramezan Jafari, Davood Bashash, Hassan Abolghasemi, Mehrad Aria
Summary: A highly efficient computer-aided detection (CAD) system using a neural search architecture network (NASNet)-based algorithm was designed for COVID-19 detection, achieving remarkable results in identifying patients with COVID-19 in the early stages. The performance of the CAD system demonstrated high detection sensitivity, specificity, and accuracy, showing potential to help radiologists detect COVID-19 early on and prevent the loss of healthcare resources during the pandemic.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Oncology
Ebtasam Ahmad Siddiqui, Vijayshri Chaurasia, Madhu Shandilya
Summary: This paper presents a novel automated diagnosis technique for lung nodules using three-dimensional deep convolutional neural networks and multi-layered filters. Volumetric computed tomographic images are utilized for lung nodule diagnosis, generating three-dimensional feature layers that retain temporal links between adjacent slices. Multiple activation functions at different levels of the network improve feature extraction and efficient classification. The proposed technique divides lung CT images into malignant and benign categories and outperforms the state-of-the-art in terms of various performance metrics.
JOURNAL OF CANCER RESEARCH AND CLINICAL ONCOLOGY
(2023)
Article
Automation & Control Systems
Siqi Cai, Yizhi Liao, Lixuan Lai, Haiyu Zhou, Longhan Xie
Summary: This article introduces a computer-aided diagnosis method based on convolutional neural networks for generating corrective solutions for patients with pectus excavatum. By training a CNN model to predict the corrected sternum contours for patients, the effectiveness of the approach was validated through experiments.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Ashish Semwal, Narendra D. Londhe
Summary: Pain assessment is crucial for medical diagnosis and treatment. Researchers have been focusing on objective methods, such as a fully automated pain assessment system based on facial expressions. Joint learning from both appearance and shape-based features results in a more robust pain assessment model.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Biomedical
Rama Vaibhav Kaulgud, Arun Patil
Summary: The imaging modality of computed tomography (CT) is significant for the diagnosis of lung cancer. This research evaluates 50 recent articles from the last five years to highlight the problems associated with current approaches to lung nodule identification. Deep learning optimization-based strategies are found to outperform convolutional techniques in terms of performance measures. The report also explores potential directions for future research and challenges in improving lung nodule detection accuracy.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Chemistry, Multidisciplinary
Imran Qureshi, Qaisar Abbas, Junhua Yan, Ayyaz Hussain, Kashif Shaheed, Abdul Rauf Baig
Summary: Hypertensive retinopathy is a retinal disorder associated with high blood pressure, and early identification and analysis are crucial for preventing blindness. A new CAD-HR system is proposed, which combines depth-wise separable CNN with linear support vector machine, to accurately diagnose hypertensive retinopathy. The system is shown to require less computational time and fewer parameters while achieving high accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Multidisciplinary
Junjie Xing, Minping Jia
Summary: An automatic detection method based on convolutional neural networks is proposed in this paper and its detection performance is evaluated and compared with other models, showing that the method has better performance in real-time automatic detection of workpiece surface defects.
