Article
Veterinary Sciences
Clara Schoneberg, Lothar Kreienbrock, Amely Campe
Summary: Latent class analysis is a well-established method in human and veterinary medicine for evaluating the accuracy of diagnostic tests without a gold standard. An important assumption of this procedure is the conditional independence of the tests. The study extended the traditional latent class model with a term for the conditional dependency of the tests, which resulted in a more accurate estimation of test accuracy and prevalence compared to the Bayesian approach in certain scenarios.
FRONTIERS IN VETERINARY SCIENCE
(2021)
Article
Mathematics
Heba F. Nagy, Amer Ibrahim Al-Omari, Amal S. Hassan, Ghadah A. Alomani
Summary: This article discusses the application of ranked set sampling in the estimation of inverted Kumaraswamy distribution parameters. Through simulation studies and real data application, it is demonstrated that the RSS-based estimators outperform their simple random sampling counterparts.
Article
Multidisciplinary Sciences
Ye Lin, Sean B. Andersson
Summary: Single Particle Tracking (SPT) is a well-known tool for studying the dynamics of biological macromolecules inside living cells. The study focuses on the problem of localization and parameter estimation and proposes an Expectation Maximization (EM) based framework for simultaneous handling. Two representative methods, namely SMC-EM and U-EM, demonstrate better performance compared to standard techniques, especially at low signal levels.
Article
Engineering, Electrical & Electronic
Duc-Nghia Tran, Sebastien Li-Thiao-Te, Yves-Michel Frapart
Summary: Electron paramagnetic resonance (EPR) is a spectroscopy and imaging method used for detecting free radicals and oxidative stress biological markers. By introducing overmodulation acquisition procedures to improve signal-to-noise ratio (SNR), this study presents maximum likelihood estimates (MLEs) and formulas for expected relative error in single-line EPR spectra. The results are validated with simulated spectra and in vitro experimental results, showing that computed relative error is a good estimate of observed instrumental variation.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Mathematics, Applied
Refah Alotaibi, Hassan Okasha, Hoda Rezk, Abdullah M. Almarashi, Mazen Nassar
Summary: This paper introduces a new extension of the traditional Lomax distribution, discussing the hazard rate function, properties, and parameter estimation methods of the new distribution, as well as analyzing two real data sets. The estimates of the service times for Aircraft Windshield and the survival times of patients given chemotherapy medication are obtained based on data analysis.
Article
Chemistry, Analytical
Ming-Yan Gong, Bin Lyu
Summary: The existing expectation maximization (EM) and space-alternating generalized EM (SAGE) algorithms are applied to DOA estimation in unknown uniform noise. A new modified EM (MEM) algorithm applicable to the noise assumption is also proposed. The simulation results show that the improvement in EM-type algorithms ensures stability and the performance of the algorithms varies depending on the signal models.
Article
Automation & Control Systems
Yanling Chang, Alfredo Garcia, Zhide Wang, Lu Sun
Summary: This article discusses the (inverse) structural estimation of POMDPs based on observable sequences and implemented actions. The structural properties of an entropy regularized POMDP are analyzed, and conditions for model identifiability without knowledge of state dynamics are specified. A soft policy gradient algorithm is used to compute a maximum likelihood estimator, and an equipment replacement problem is used as an illustration.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2023)
Article
Engineering, Electrical & Electronic
Xiaoke Shang, Zhenkai Zhang, Yue Xiao
Summary: This paper proposes a two-step radar target parameter estimation algorithm based on OTFS, which combines maximum likelihood estimation and gradient descent principle. Simulation results show that the algorithm has better estimation performance and resolution while reducing computational complexity.
IET RADAR SONAR AND NAVIGATION
(2023)
Article
Acoustics
Changheng Li, Jorge Martinez, Richard Christian Hendriks
Summary: In this paper, two joint maximum likelihood estimators are proposed to estimate acoustic-scene related parameters, including relative transfer functions (RTFs), source power spectral densities (PSDs) and PSDs of the late reverberation. Experimental results demonstrate that our proposed joint MLE methods outperform existing MLE-based approaches that use only a single time frame in terms of estimation accuracy, noise reduction performance, speech quality, and speech intelligibility.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2023)
Article
Economics
Angelo Mele, Lingjiong Zhu
Summary: We have developed approximate estimation methods for exponential random graph models (ERGMs), where the likelihood is proportional to an intractable normalizing constant. Instead of using Monte Carlo simulations, which can converge slowly, we propose a deterministic method based on a variational mean-field approximation. Our method provides lower and upper bounds for the approximation error, allowing us to estimate the distance between true likelihood and mean-field likelihood. Monte Carlo simulations indicate that our deterministic method performs well in practice, surpassing the limitations of our theoretical approximation bounds.
REVIEW OF ECONOMICS AND STATISTICS
(2023)
Article
Computer Science, Information Systems
Buddhika Nettasinghe, Vikram Krishnamurthy
Summary: This article introduces a novel maximum likelihood estimation approach that leverages the friendship paradox to sample more efficiently from the tail of the degree distribution in undirected networks. The proposed method results in a smaller bias, variance, and Cramer-Rao lower bound compared to traditional methods. Detailed numerical and empirical results demonstrate the performance of the proposed method and its extensions to other parametric degree distributions.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2021)
Article
Engineering, Electrical & Electronic
Alessio Fascista, Angelo Coluccia, Giuseppe Ricci
Summary: This study addresses the problem of maximum likelihood direct position estimation of a multi-antenna receiver in dynamic multipath environments, proposing a reduced-complexity algorithm that utilizes non line-of-sight paths to improve efficiency and performance.
Article
Engineering, Electrical & Electronic
YunPeng Li, ZhaoHui Ye
Summary: This letter introduces a novel boosting-based method for univariate nonparametric estimation, deducing the boosting algorithm through second-order approximation of nonparametric log-likelihood, and choosing Gaussian kernel and smooth spline as weak learners to satisfy the smoothing assumptions. Simulations and real data experiments demonstrate the efficacy of the proposed approach.
IEEE SIGNAL PROCESSING LETTERS
(2021)
Article
Mathematics
Rashad Bantan, Mahmoud Elsehetry, Amal S. Hassan, Mohammed Elgarhy, Dreamlee Sharma, Christophe Chesneau, Farrukh Jamal
Summary: This study introduces a flexible model, the half logistic inverted Topp-Leone (HLITL) distribution, for fitting right-skewed, reversed J-shaped, and unimodal data, with parameter estimation based on the maximum likelihood method, showing more accurate results under ranked set sampling. Comparisons with other models reveal that the HLITL model represents a better alternative lifetime distribution.
Article
Engineering, Electrical & Electronic
Amir Weiss
Summary: A blind Direction-of-Arrivals (DOAs) estimate is proposed for narrowband signals in Acoustic Vector-Sensor (AVS) arrays, utilizing a unique parametric Canonical Polyadic Decomposition (CPD) of the observations' Second-Order Statistics (SOSs) tensor. The method does not require prior assumptions on the array configuration and achieves improved accuracy compared to existing methods. Simulation experiments validate the approach, showing a significant reduction in root mean squared error.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2021)
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)
