Article
Biology
R. Jenkin Suji, Sarita Singh Bhadauria, W. Wilfred Godfrey
Summary: This paper presents the critical steps of lung segmentation and lung nodule detection in lung cancer CAD system and discusses the background and taxonomy of 2.5D methods, providing future research directions.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Yifan Wang, Chuan Zhou, Heang-Ping Chan, Lubomir M. Hadjiiski, Aamer Chughtai, Ella A. Kazerooni
Summary: In this study, a hybrid deep learning model was developed for accurate segmentation of lung nodules with different characteristics. The model combined two deep convolutional neural networks as encoders to improve the segmentation performance of complex lung nodules. The results showed that the hybrid model achieved segmentation accuracy comparable to radiologists' performance and outperformed the individual shallow or deep models.
Article
Computer Science, Artificial Intelligence
Mohammad Hesam Hesamian, Wenjing Jia, Xiangjian He, Qingqing Wang, Paul J. Kennedy
Summary: This paper proposes an innovative approach to accurately detect and segment lung nodules in CT images by utilizing the changes of nodule shapes over continuous slices and learning the unique color patterns formed by nodules using synthetic images and deep learning techniques. The proposed method achieves simultaneous detection and segmentation of nodules with a 10% higher accuracy than existing methods, without introducing high computation cost. By leveraging inter-slice information and the proposed synthetic image, lung nodule segmentation is done more accurately and effectively.
APPLIED INTELLIGENCE
(2021)
Article
Biology
S. Akila Agnes, J. Anitha, A. Arun Solomon
Summary: This study proposes a novel computer-aided detection (CADe) system that improves the performance of lung cancer detection using deep learning. The system detects lung nodules from CT scans in two stages and achieves high sensitivity in detecting nodules of diverse sizes. The experimental results show that the proposed CADe system outperforms other methods in terms of average sensitivity and detection rate.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Information Systems
Reza Mousavi Moghaddam, Nasser Aghazadeh
Summary: For the past three years, the world has been dealing with an infectious disease that primarily affects the human respiratory system, causing numerous deaths and significant economic losses globally. Therefore, there is a need for more attention to computer-aided detection/diagnosis (CAD) systems to assist in diagnosing and treating respiratory-related diseases, enabling healthcare systems to better handle epidemic situations.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Biomedical
Hassan Mkindu, Longwen Wu, Yaqin Zhao
Summary: This study proposes an automated computer-aided diagnosis scheme based on Vision Transformer architecture with Bayesian Optimisation for lung nodule detection to assist radiologists in decision-making. The empirical results prove the effectiveness of the proposed algorithm compared to the state-of-the-art methods.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Biology
Yu Gu, Jingqian Chi, Jiaqi Liu, Lidong Yang, Baohua Zhang, Dahua Yu, Ying Zhao, Xiaoqi Lu
Summary: This paper summarizes the CAD approaches using deep learning for lung nodule detection on CT scan data. The technologies show promising results in improving the survival rate of lung cancer patients, but there are still challenges and limitations to be addressed.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Engineering, Electrical & Electronic
Hassan Mkindu, Longwen Wu, Yaqin Zhao
Summary: Lung cancer is the leading cause of cancer-related death, and radiologists use computed tomography (CT) to diagnose lung nodules. Manual analysis of hundreds of CT images by radiologists is burdensome and sometimes inaccurate. This study proposes a CAD scheme based on 3D multi-scale vision transformer (3D-MSViT) to enhance feature extraction and improve lung nodule prediction efficiency.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Review
Engineering, Biomedical
Haizhe Jin, Cheng Yu, Zibo Gong, Renjie Zheng, Yinan Zhao, Quanwei Fu
Summary: This study systematically analyzed and compared the performance of machine learning algorithms using the same dataset in the diagnosis of pulmonary nodules through a literature review.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Hassan Mkindu, Longwen Wu, Yaqin Zhao
Summary: Malignant lung nodules are the worse stage for lung cancer patients, early detection is essential for treatment. This study presents a 3D U-shaped encoding and decoding CNN integrated with channel attention mechanisms for lung nodule detection in chest CT images.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
R. Nurfauzi, H. A. Nugroho, I. Ardiyanto, E. L. Frannita
Summary: Male lung cancer has the highest mortality rate, with juxta-pleural and juxtavascular nodules being the most common types on the lung surface. The research aims for fast computational time and low error in covering nodule areas.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Soichiro Miki, Yukihiro Nomura, Naoto Hayashi, Shouhei Hanaoka, Eriko Maeda, Takeharu Yoshikawa, Yoshitaka Masutani, Osamu Abe
Summary: The study found that in a screening environment, radiologists tend to miss lung nodules in the hilar regions, and different radiologists have their own patterns of missed nodules.
ACADEMIC RADIOLOGY
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Yihui Du, Marcel J. W. Greuter, Mathias W. Prokop, Geertruida H. de Bock
Summary: This study aimed to determine appropriate pricing for DL-CAD to be cost-saving and to identify the potentially most cost-effective reading mode for lung cancer screening. The results showed that DL-CAD as a pre-screening reader had the largest potential for cost savings.
INSIGHTS INTO IMAGING
(2023)
Article
Computer Science, Information Systems
Ganesh Singadkar, Abhishek Mahajan, Meenakshi Thakur, Sanjay Talbar
Summary: This paper proposes a more robust and accurate lung segmentation method which can successfully handle juxtapleural nodules and pulmonary vessels, achieving high levels of segmentation accuracy. The method is faster than manual segmentation done by radiologists.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2021)
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, Artificial Intelligence
Rachel Lea Draelos, David Dov, Maciej A. Mazurowski, Joseph Y. Lo, Ricardo Henao, Geoffrey D. Rubin, Lawrence Carin
Summary: This study utilized a large-scale chest CT dataset with high-quality abnormality labels, developed a method for extracting labels from radiology reports, and used a deep learning CNN model for multi-organ, multi-disease classification of CT volumes, demonstrating good classification performance.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Critical Care Medicine
Takeshi Johkoh, Kyung Soo Lee, Mizuki Nishino, William D. Travis, Jay H. Ryu, Ho Yun Lee, Christopher J. Ryerson, Tomas Franquet, Alexander A. Bankier, Kevin K. Brown, Jin Mo Goo, Hans-Ulrich Kauczor, David A. Lynch, Andrew G. Nicholson, Luca Richeldi, Cornelia M. Schaefer-Prokop, Johny Verschakelen, Suhail Raoof, Geoffrey D. Rubin, Charles Powell, Yoshikazu Inoue, Hiroto Hatabu
Summary: Diagnosis and management of drug-related pneumonitis requires excluding other potential causes, understanding the incidence and risk factors, and evaluating imaging features based on the distribution of lung parenchymal abnormalities.
Article
Respiratory System
Martine Remy-Jardin, Christopher J. Ryerson, Mark L. Schiebler, Ann N. C. Leung, James M. Wild, Marius M. Hoeper, Philip O. Alderson, Lawrence R. Goodman, John Mayo, Linda B. Haramati, Yoshiharu Ohno, Patricia Thistlethwaite, Edwin J. R. van Beek, Shandra Lee Knight, David A. Lynch, Geoffrey D. Rubin, Marc Humbert
Summary: Pulmonary hypertension (PH) is characterized by a mean pulmonary artery pressure greater than 20 mmHg and classified into five groups with similar pathophysiologic mechanisms. Radiologists play a key role in the multidisciplinary assessment and management of PH, utilizing CT, MRI, and nuclear medicine to evaluate and treat the condition effectively.
EUROPEAN RESPIRATORY JOURNAL
(2021)
Article
Radiology, Nuclear Medicine & Medical Imaging
Martine Remy-Jardin, Christopher J. Ryerson, Mark L. Schiebler, Ann N. C. Leung, James M. Wild, Marius M. Hoeper, Philip O. Alderson, Lawrence R. Goodman, John Mayo, Linda B. Haramati, Yoshiharu Ohno, Patricia Thistlethwaite, Edwin J. R. van Beek, Shandra Lee Knight, David A. Lynch, Geoffrey D. Rubin, Marc Humbert
Summary: Pulmonary hypertension is a condition characterized by elevated mean pulmonary artery pressure, with radiologists playing a crucial role in its assessment and management. A working group was established to focus on the role of imaging techniques in diagnosing and managing PH, highlighting the importance of imaging in the recognition, work-up, treatment planning, and follow-up of PH.
Article
Radiology, Nuclear Medicine & Medical Imaging
Takeshi Johkoh, Kyung Soo Lee, Mizuki Nishino, William D. Travis, Jay H. Ryu, Ho Yun Lee, Christopher J. Ryerson, Tomas Franquet, Alexander A. Bankier, Kevin K. Brown, Jin Mo Goo, Hans-Ulrich Kauczor, David A. Lynch, Andrew G. Nicholson, Luca Richeldi, Cornelia M. Schaefer-Prokop, Johny Verschakelen, Suhail Raoof, Geoffrey D. Rubin, Charles Powell, Yoshikazu Inoue, Hiroto Hatabu
Summary: The frequency and broad spectrum of lung toxicity have increased with the use of molecular targeting agents and immune checkpoint inhibitors (ICIs), particularly in cancer patients. Diagnosis of drug-related pneumonitis (DRP) involves excluding other causes, and awareness of its incidence and risk factors is crucial. The severity of DRP symptoms can range from mild to life-threatening, requiring accurate diagnosis and prompt treatment.
Review
Radiology, Nuclear Medicine & Medical Imaging
Jeffrey P. Kanne, Harrison Bai, Adam Bernheim, Michael Chung, Linda B. Haramati, David F. Kallmes, Brent P. Little, Geoffrey D. Rubin, Nicola Sverzellati
Summary: The role of imaging, specifically CT scans, evolved during the pandemic, initially being seen as an alternative and potentially superior testing method compared to RT-PCR, but later having a more limited role based on specific indications.
Article
Cardiac & Cardiovascular Systems
Michael E. Zimmerman, Juan C. Batlle, Cathleen Biga, Ron Blankstein, Brian B. Ghoshhajra, Mark G. Rabbat, George E. Wesbey, Geoffrey D. Rubin
Summary: The study found that the direct cost of performing Coronary CT angiography (CCTA) is significantly higher than Contrast-enhanced thoracic CT (CECT), with both labor and equipment costs being more expensive for CCTA. This suggests that reimbursement schedules treating these procedures similarly undervalue the resources required for CCTA, potentially limiting access to this procedure.
JOURNAL OF CARDIOVASCULAR COMPUTED TOMOGRAPHY
(2021)
Article
Medical Informatics
Vincent M. D'Anniballe, Fakrul Islam Tushar, Khrystyna Faryna, Songyue Han, Maciej A. Mazurowski, Geoffrey D. Rubin, Joseph Y. Lo
Summary: This study developed a high-throughput multi-label annotator for body CT reports using a dictionary approach and rule-based algorithms for disease label extraction, with attention-guided recurrent neural networks for classification. The method showed excellent accuracy and performance, able to adapt to various cases and diseases, suitable for automated labeling of large-scale medical datasets.
BMC MEDICAL INFORMATICS AND DECISION MAKING
(2022)
Article
Cardiac & Cardiovascular Systems
Ricardo C. Cury, Jonathon Leipsic, Suhny Abbara, Stephan Achenbach, Daniel Berman, Marcio Bittencourt, Matthew Budoff, Kavitha Chinnaiyan, Andrew D. Choi, Brian Ghoshhajra, Jill Jacobs, Lynne Koweek, John Lesser, Christopher Maroules, Geoffrey D. Rubin, Frank J. Rybicki, Leslee J. Shaw, Michelle C. Williams, Eric Williamson, Charles S. White, Todd C. Villines, Ron Blankstein
Summary: Coronary Artery Disease Reporting and Data System (CAD-RADS) is a standardized reporting system for patients undergoing coronary CT angiography (CCTA), aiming to guide patient management. The updated CAD-RADS 2.0 improves the initial reporting system for CCTA by considering new technical developments and clinical guidelines, including the assessment of stenosis, plaque burden, and modifiers.
JOURNAL OF CARDIOVASCULAR COMPUTED TOMOGRAPHY
(2022)
Article
Cardiac & Cardiovascular Systems
Ricardo C. Cury, Jonathon Leipsic, Suhny Abbara, Stephan Achenbach, Daniel Berman, Marcio Bittencourt, Matthew Budoff, Kavitha Chinnaiyan, Andrew D. Choi, Brian Ghoshhajra, Jill Jacobs, Lynne Koweek, John Lesser, Christopher Maroules, Geoffrey D. Rubin, Frank J. Rybicki, Leslee J. Shaw, Michelle C. Williams, Eric Williamson, Charles S. White, Todd C. Villines, Ron Blankstein
Summary: The Coronary Artery Disease Reporting and Data System (CAD-RADS) was created to standardize reporting system for patients undergoing coronary CT angiography (CCTA) and guide patient management. The updated 2022 CAD-RADS 2.0 aims to improve the initial reporting system for CCTA by considering new technical developments in cardiac CT, including data from recent clinical trials and new clinical guidelines.
JACC-CARDIOVASCULAR IMAGING
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ricardo C. Cury, Jonathon A. Leipsic, Suhny Abbara, Stephan Achenbach, Daniel S. Berman, Marcio Bittencourt, Matthew Budoff, Kavitha Chinnaiyan, Andrew D. Choi, Brian Ghoshhajra, Jill Jacobs, Lynne Koweek, John Lesser, Christopher Maroules, Geoffrey D. Rubin, Frank J. Rybicki, Leslee J. Shaw, Michelle C. Williams, Eric Williamson, Charles S. White, Todd C. Villines, Ron Blankstein
Summary: Coronary Artery Disease Reporting and Data System (CAD-RADS) is a standardized reporting system for CCTA patients, aiming to guide patient management. The updated CAD-RADS 2.0 improves the initial reporting system by considering new technical developments, clinical trial data, and clinical guidelines.
JOURNAL OF THE AMERICAN COLLEGE OF RADIOLOGY
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ricardo C. Cury, Jonathon Leipsic, Suhny Abbara, Stephan Achenbach, Daniel Berman, Marcio Bittencourt, Matthew Budoff, Kavitha Chinnaiyan, Andrew D. Choi, Brian Ghoshhajra, Jill Jacobs, Lynne Koweek, John Lesser, Christopher Maroules, Geoffrey D. Rubin, Frank J. Rybicki, Leslee J. Shaw, Michelle C. Williams, Eric Williamson, Charles S. White, Todd C. Villines, Ron Blankstein
Summary: The Coronary Artery Disease Reporting and Data System (CAD-RADS) is a standardized reporting system for patients who undergo coronary CT angiography (CCTA) and aims to guide the management of patients. The updated 2022 CAD-RADS 2.0 improves the initial reporting system for CCTA by considering new technical developments, clinical trials, and guidelines. The classification in CAD-RADS follows a framework of stenosis, plaque burden, and modifiers, with the addition of assessing lesion-specific ischemia. It provides a standardized framework for communication, education, research, and improving patient care.
RADIOLOGY-CARDIOTHORACIC IMAGING
(2022)
Proceedings Paper
Engineering, Biomedical
Fakrul Islam Tushar, Vincent M. D'Anniballe, Geoffrey D. Rubin, Ehsan Samei, Joseph Y. Lo
Summary: This paper examines the effect of co-occurring diseases on weakly supervised learning in computer-aided diagnosis. The results show that binary classifiers outperform multi-label classifiers in every disease category. However, the performance of binary classifiers is heavily influenced by co-occurring diseases, indicating that they may not always learn the correct appearance of specific diseases.
MEDICAL IMAGING 2022: COMPUTER-AIDED DIAGNOSIS
(2022)
Article
Computer Science, Artificial Intelligence
Fakrul Islam Tushar, Vincent M. D'Anniballe, Rui Hou, Maciej A. Mazurowski, Wanyi Fu, Ehsan Samei, Geoffrey D. Rubin, Joseph Y. Lo
Summary: This study aims to design multidisease classifiers for body CT scans for three different organ systems using automatically extracted labels from radiology text reports. The results of the study demonstrate that weakly supervised deep learning models can effectively classify different diseases in multiple organ systems and achieve relatively high accuracy.
RADIOLOGY-ARTIFICIAL INTELLIGENCE
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ehsan Samei, Taylor Richards, William P. Segars, Melissa A. Daubert, Alex Ivanov, Geoffrey D. Rubin, Pamela S. Douglas, Udo Hoffmann
Summary: A computational framework was developed to objectively assess the precision of quantifying coronary stenosis in cardiac CTA, showing promising results in predicting image quality and estimation precision. This framework was successfully applied in a clinical trial and demonstrated potential for optimizing imaging protocols for targeted precision and measurement consistency in cardiac CT images.
JOURNAL OF MEDICAL IMAGING
(2021)