Article
Mathematical & Computational Biology
Yunju Im, Yuan Huang, Aixin Tan, Shuangge Ma
Summary: Cancer is a heterogeneous disease and finite mixture of regression (FMR) is an important analysis technique. Recently, histopathological images have been used in cancer research and high-dimensional image features have been effective for modeling cancer outcomes. This article proposes a Bayesian approach for cancer FMR analysis, incorporating clinical, demographic, environmental variables, as well as high-dimensional image features. The proposed method shows advantageous performance in simulations and provides interesting findings in the analysis of lung squamous cell cancer data from The Cancer Genome Atlas.
Article
Health Care Sciences & Services
Federico Ricciardi, Silvia Liverani, Gianluca Baio
Summary: The regression discontinuity design is a quasi-experimental design that estimates the causal effect of a treatment when its assignment is defined by a threshold for a continuous variable. Bandwidth selection is an important decision in this design analysis, and the proposed methodology considers units' exchangeability as the main criteria for selecting subjects. The validity of the methodology is demonstrated through simulated experiments and an example on the effect of statins on cholesterol levels.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2023)
Article
Mathematical & Computational Biology
Yunju Im, Yuan Huang, Jian Huang, Shuangge Ma
Summary: This article further advances cancer FMR analysis based on histopathological imaging data. By simultaneously using two types of imaging features extracted based on domain-specific biomedical knowledge and automated signal processing software, a significant modeling/methodological advancement is achieved in reflecting the increased resolution of the second type of imaging features.
STATISTICS IN MEDICINE
(2022)
Article
Statistics & Probability
Dawei Ding, George Karabatsos
Summary: We introduce Dirichlet process mixture (DPM) models for prediction and cluster-wise variable selection, showing that the proposed models outperform the standard DPM model in terms of predictive accuracy, variable selection, and clustering accuracy in a simulation study. Real data analysis further confirms the better predictive accuracy of the proposed DPM models.
Article
Statistics & Probability
Raffaele Argiento, Maria De Iorio
Summary: In this paper, a new class of priors, the Normalized Independent Point Process, is introduced and its probabilistic properties and special cases are investigated. Additionally, marginal and conditional algorithms for finite mixture models with a random number of components are designed, which overcome the challenges associated with the Reversible Jump algorithm. The performance and potential of the model are illustrated through simulation studies and real data applications.
ANNALS OF STATISTICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Kuo-Jung Lee, Martin Feldkircher, Yi-Chi Chen
Summary: A Bayesian framework is proposed to handle model selection and the selection of the number of mixture components in finite mixture models simultaneously. A feasible reversible jump Markov Chain Monte Carlo algorithm is introduced to model each component as a sparse regression model and make the approach robust to outliers. The study applied this framework to investigate early warning indicators in cross-sectional data, revealing two distinct country groups with varying effects of vulnerability indicators.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2021)
Article
Automation & Control Systems
Konstantinos Perrakis, Thomas Lartigue, Frank Dondelinger, Sach Mukherjee
Summary: Regularized regression models may not hold under the condition of distributional differences and latent group structure in large data. To address this, we propose a mixture model that combines the distribution of features with the conditional response variable's regression model, allowing for better learning of latent structure.
JOURNAL OF MACHINE LEARNING RESEARCH
(2023)
Article
Automation & Control Systems
Hiromasa Kaneko
Summary: This paper introduces the development and use of extended GMR (EGMR) which offers improved predictive ability over conventional GMR by optimizing hyperparameters and model selection, demonstrating the effectiveness of inverse analysis.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2021)
Review
Endocrinology & Metabolism
Mikahela A. Lopez-Morales, Maria Castello-Ruiz, Maria C. Burguete, David Hervas, Miguel A. Perez-Pinzon, Juan B. Salom
Summary: This systematic review evaluates the effects and mechanisms of Resveratrol (RSV) in animal models of ischemic stroke. The results show that RSV has beneficial effects on structural and functional outcomes, including reducing infarct size, edema size, BBB impairment, neurofunctional impairment, and improving motor performance. These effects may be attributed to the reduction of oxidative stress, inflammation, and apoptosis. However, further high-quality preclinical research is needed to better inform clinical research.
JOURNAL OF CEREBRAL BLOOD FLOW AND METABOLISM
(2023)
Article
Health Care Sciences & Services
Xiaoxiao Zhou, Xinyuan Song
Summary: This study introduces a method for conducting mediation analysis in the context of Cox proportional hazards cure models, addressing the cure fraction problem. Path-specific effects on survival time and probability are assessed using mediation formula approach and Bayesian estimation.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2021)
Review
Medicine, Research & Experimental
Mikkel Schou Andersen, Mikkel Seremet Kofoed, Asger Sand Paludan-Mueller, Christian Bonde Pedersen, Tiit Mathiesen, Christian Mawrin, Martin Wirenfeldt, Bjarne Winther Kristensen, Birgitte Brinkmann Olsen, Bo Halle, Frantz Rom Poulsen
Summary: This systematic review found high consistent tumor take rates (TTRs) in established cell line models and varying TTRs in primary patient-derived models and genetically engineered models. However, there were several issues identified regarding the quality of reporting and methodological approach, which reduced the validity, transparency, and reproducibility of the studies and suggested a high risk of publication bias. Finally, each tumor model type has specific roles in research based on their advantages (and disadvantages).
JOURNAL OF TRANSLATIONAL MEDICINE
(2023)
Review
Nursing
Lin Yu, Shu Yang, Chunmiao Zhang, Pingping Guo, Xuehui Zhang, Mengmeng Xu, Qi Tian, Xuan Cui, Wei Zhang
Summary: The study found that decision aids (DAs) can significantly improve pregnant women's knowledge and decision-making satisfaction, reduce decision conflicts, and help increase the proportion of women who make informed choices. However, DAs had no impact on anxiety and decision regret.
JOURNAL OF ADVANCED NURSING
(2021)
Article
Health Care Sciences & Services
Raphael Richert, Jean-Christophe Farges, Jean-Christophe Maurin, Jerome Molimard, Philippe Boisse, Maxime Ducret
Summary: This study aimed to classify the relative contributions of four biomechanical factors to the root stresses of the resected premolar. The results showed that the factors of preparation and bone height had a significant influence on root stresses, and neglecting the interactions between factors would result in missing nearly half of the biomechanical impact.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Mathematics, Interdisciplinary Applications
Minjung Kyung, Ju-Hyun Park, Ji Yeh Choi
Summary: The proposed Bayesian mixture extension of ERA classifies observations into multiple subpopulations and estimates ERA models within each subpopulation, demonstrating good performance in recovering parameters and empirical usefulness in real data applications.
Review
Computer Science, Information Systems
Xuan Song, Xinyan Liu, Fei Liu, Chunting Wang
Summary: The study found that machine learning models perform similarly to logistic regression models in predicting acute kidney injury (AKI), but some machine learning models show exceptional performance, with gradient boosting models performing the best.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2021)
Review
Criminology & Penology
NaeHyung Lee, Terri Deocampo Pigott, Ashley Watson, Katherine Reuben, Kathryn O'Hara, Greta Massetti, Xiangming Fang, Shannon Self-Brown
Summary: This scoping review examines the literature on polyvictimization and health outcomes, highlighting the varied constructions of polyvictimization and identifying gaps in knowledge. The findings emphasize the need for a standardized definition of polyvictimization and suggest specific health outcomes that should be investigated further. The study also underscores the importance of resilience and coping education for childhood polyvictims.
TRAUMA VIOLENCE & ABUSE
(2023)
Article
Mathematical & Computational Biology
Leila C. Kahwati, Bridget J. Kelly, Mihaela Johnson, Rachel T. Clark, Meera Viswanathan
Summary: Effective communication of systematic review results using qualitative comparative analyses (QCA) requires selecting appropriate formats. This study found that presenting results in a figure format can enhance subjective comprehension and parameter interpretation compared to text or table formats. However, the unfamiliarity with methods and terminology remains a barrier to full understanding of the findings.
RESEARCH SYNTHESIS METHODS
(2023)
Article
Education & Educational Research
Mikkel Helding Vembye, James Eric Pustejovsky, Therese Deocampo Pigott
Summary: Meta-analytic models for dependent effect sizes have become increasingly complex, posing challenges for a priori power calculations. This study introduces power approximations for tests of average effect sizes based on common approaches for handling dependent effect sizes. Through Monte Carlo simulation, the study demonstrates that the new power formulas can accurately estimate the true power of meta-analytic models for dependent effect sizes. Lastly, the study compares the Type I error rate and power of several common models, finding that tests using robust variance estimation offer better Type I error calibration than tests with model-based variance estimation.
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS
(2023)
Editorial Material
Medicine, General & Internal
Madelin R. Siedler, M. Hassan Murad, Rebecca L. Morgan, Yngve Falck-Ytter, Reem A. Mustafa, Shahnaz Sultan, Philipp Dahm
BMJ EVIDENCE-BASED MEDICINE
(2023)
Review
Endocrinology & Metabolism
Ghada El-Hajj Fuleihan, Gregory A. Clines, Mimi Hu, Claudio Marcocci, M. Hassan Murad, Thomas Piggott, Catherine Van Poznak, Joy Y. Wu, Matthew T. Drake
Summary: Hypercalcemia of malignancy (HCM) is a common complication of malignancies, but its incidence may be decreasing. Despite the availability of effective medications, there are no evidence-based recommendations for managing HCM. A panel of experts developed guidelines for the treatment of adults with HCM, recommending the use of denosumab or intravenous bisphosphonate. Treatment of the primary malignancy is crucial for controlling HCM.
JOURNAL OF CLINICAL ENDOCRINOLOGY & METABOLISM
(2023)
Article
Endocrinology & Metabolism
Anthony L. McCall, David C. Lieb, Roma Gianchandani, Heidemarie MacMaster, Gregory A. Maynard, M. Hassan Murad, Elizabeth Seaquist, Joseph Wolfsdorf, Robin Fein Wright, Wojtek Wiercioch
Summary: This study reviewed and updated the management of hypoglycemia in people with diabetes, aiming to reduce and prevent it. A multidisciplinary panel of experts identified 10 clinical questions and made 10 recommendations. These recommendations can be used to inform clinical practice and improve healthcare systems, as hypoglycemia is an important complication for people with diabetes.
JOURNAL OF CLINICAL ENDOCRINOLOGY & METABOLISM
(2023)
Editorial Material
Medicine, General & Internal
M. Hassan Murad, Melanie D. Swift, Raymund R. Razonable, Aaron J. Tande, John W. Wilson, Mary J. Kasten, Irene G. Sia, Jennifer N. Matey, Greg Vanichkachorn, Natalie A. Caine, Vijay Shah, Jack O'Horo, Molly J. Destro Borgen, Clayton T. Cowl, Elie F. Berbari
MAYO CLINIC PROCEEDINGS
(2023)
Article
Medicine, General & Internal
Zhen Wang, Muayad A. Alzuabi, Rebecca L. Morgan, Reem A. Mustafa, Yngve Falck-Ytter, Philipp Dahm, Shahnaz Sultan, Mohammad Hassan Murad
Summary: This study empirically evaluated five commonly used meta-analysis methods and their impact on imprecision judgements about effect estimates. The results showed that different pooling methods led to different conclusions about the precision of effect estimates, particularly when the number of studies was small and statistical heterogeneity was substantial. Therefore, sensitivity analyses using multiple methods may be necessary in these two scenarios.
BMJ EVIDENCE-BASED MEDICINE
(2023)
Review
Medicine, General & Internal
Daniel E. Jonas, Sean R. Riley, Lindsey C. Lee, Cory P. Coffey, Shu-Hua Wang, Gary N. Asher, Anne M. Berry, Niketa Williams, Casey Balio, Christiane E. Voisin, Leila C. Kahwati
Summary: Screening and treatment of latent tuberculosis infection (LTBI) in adults can reduce the risk of progression to active tuberculosis disease, but isoniazid treatment is associated with a higher risk of hepatotoxicity. No direct evidence evaluated the benefits and harms of LTBI screening compared with no screening.
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2023)
Letter
Medicine, General & Internal
Gerald Gartlehner, Leila Kahwati
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2023)
Article
Medicine, General & Internal
M. Hassan Murad, Zhen Wang, Ye Zhu, Samer Saadi, Haitao Chu, Lifeng Lin
Summary: Trading off benefits and harms requires knowledge of the absolute risk reduction or risk difference, making risk difference a critical measure for decision making. However, estimating risk difference is not straightforward and the available methods have various limitations. This article discusses four methods for estimating risk difference and provides recommendations on when to use each approach.
BMJ-BRITISH MEDICAL JOURNAL
(2023)
Article
Medicine, General & Internal
Gerald Gartlehner, Barbara Nussbaumer-Streit, Declan Devane, Leila Kahwati, Meera Viswanathan, Valerie J. King, Amir Qaseem, Elie Akl, Holger J. Schuenemann
Summary: This paper is a part of a series of methodological guidance provided by the Cochrane Rapid Reviews Methods Group, which explains that rapid reviews (RRs) are designed to speed up the review process while maintaining systematic and transparent methods. The paper focuses on considerations for rating the certainty of evidence (COE) in RRs and provides recommendations for the implementation of GRADE. It also suggests alternative approaches for rating COE if full implementation of GRADE is not feasible.
BMJ EVIDENCE-BASED MEDICINE
(2023)
Review
Medicine, General & Internal
Meera Viswanathan, Rachel Peragallo Urrutia, Kesha N. Hudson, Jennifer Cook Middleton, Leila C. Kahwati
Summary: Neural tube defects are common birth defects in the US. This study reviewed the evidence on the benefits and harms of folic acid supplementation for the prevention of these defects. The findings showed that folic acid supplementation can reduce the risk of neural tube defects, and there were no significant harms associated with pregnancy-related folic acid exposure.
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2023)
Article
Health Care Sciences & Services
M. Hassan Murad, Jos Verbeek, Lukas Schwingshackl, Tommaso Filippini, Marco Vinceti, Elie A. Akl, Rebecca L. Morgan, Reem A. Mustafa, Dena Zeraatkar, Emily Senerth, Renee Street, Lifeng Lin, Yngve Falck-Ytter, Gordon Guyatt, Holger J. Schunemann
Summary: This article discusses the impact of dose-response gradients on the certainty of evidence in intervention and exposure studies. By determining the credibility of the gradient and applying the gradient domain, ratings can be increased.
JOURNAL OF CLINICAL EPIDEMIOLOGY
(2023)
Article
Substance Abuse
Olivia K. Golan, Flora Sheng, Andrew W. Dick, Mark Sorbero, Daniel J. Whitaker, Barbara Andraka-Christou, Therese Pigott, Adam J. Gordon, Bradley D. Stein
Summary: This study examines the impact of the Affordable Care Act Medicaid expansion on the initiation rates of buprenorphine, a medication used to treat opioid use disorder. The results show that the expansion reduced income-related disparities in urban counties but had no significant effect in rural counties.
DRUG AND ALCOHOL DEPENDENCE REPORTS
(2023)
