Article
Medicine, General & Internal
Yujeong Kim, Jong-Hwan Jang, Namgi Park, Na-Young Jeong, Eunsun Lim, Soyun Kim, Nam-Kyong Choi, Dukyong Yoon
Summary: The study developed a machine learning-based active surveillance system for detecting possible factors that can induce adverse events using health claim and vaccination databases. The system successfully predicted health outcomes of interest with high accuracy, contributing to the establishment of a system for conducting active surveillance on vaccination.
JOURNAL OF KOREAN MEDICAL SCIENCE
(2021)
Article
Public, Environmental & Occupational Health
Gihong Seo, Sewon Park, Munjae Lee
Summary: This study analyzed Korean and foreign systems and proposed a method to determine the replacement time for high-risk medical devices, revealing a lack of specific regulations regarding the life cycle of such devices. The study also identified important factors for device replacement evaluation, including year of introduction, repair cost, component discontinuation, and multiple failures.
FRONTIERS IN PUBLIC HEALTH
(2022)
Article
Medicine, General & Internal
Rachael Gotlieb, Corinne Praska, Marissa A. Hendrickson, Jordan Marmet, Victoria Charpentier, Emily Hause, Katherine A. Allen, Scott Lunos, Michael B. Pitt
Summary: Despite the acknowledgement that medical jargon should be avoided, healthcare practitioners frequently use it when communicating with patients. A survey found that the general public has varied understanding of common medical jargon terms, with interpreted meanings often being the opposite of the intended ones.
Article
Multidisciplinary Sciences
Kai Packhaeuser, Sebastian Guendel, Nicolas Muenster, Christopher Syben, Vincent Christlein, Andreas Maier
Summary: With the rise of deep learning techniques, publicly available medical datasets are crucial for developing diagnostic algorithms in the medical field. However, it has been shown that deep learning systems can recover patient identities from chest X-ray data, posing a risk of privacy breach.
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Ahmed N. Balshi, Mohammed A. Al-Odat, Abdulrahman M. Alharthy, Rayan A. Alshaya, Hanan M. Alenzi, Alhadzia S. Dambung, Huda Mhawish, Saad M. Altamimi, Waleed Th. Aletreby
Summary: Automated activation of the RRT significantly reduced CPR events and rates, improved CPR success rate, reduced hospital length of stay and mortality, but increased the number of RRT activations. There were no differences in unplanned ICU admission or readmission.
Article
Biology
Hansle Gwon, Imjin Ahn, Yunha Kim, Hee Jun Kang, Hyeram Seo, Heejung Choi, Ha Na Cho, Minkyoung Kim, Jiye Han, Gaeun Kee, Seohyun Park, Kye Hwa Lee, Tae Joon Jun, Young-Hak Kim
Summary: Electronic medical records have potential in advancing healthcare technologies, but privacy issues hinder their full utilization. Deep learning-based generative models can mitigate this problem by creating synthetic data similar to real patient data. However, the risk of data leakage due to malicious attacks poses a challenge to traditional generative models. To address this, we propose a method that employs local differential privacy (LDP) to protect the model from attacks and preserve the privacy of training data, while generating medical data with reasonable performance.
COMPUTERS IN BIOLOGY AND MEDICINE
(2024)
Article
Public, Environmental & Occupational Health
Krista F. Huybrechts, Martin Kulldorff, Sonia Hernandez-Diaz, Brian T. Bateman, Yanmin Zhu, Helen Mogun, Shirley Wang
Summary: The scientific community relies on postmarketing approaches to assess the risks of medication use during pregnancy, but existing studies often focus on specific outcomes. The study used a new tree-based scan statistic data-mining method, TreeScan, identified known safety concerns, and found only one new risk alert in one of the cases.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2021)
Article
Oncology
N. E. Omar, A. I. Fahmy Soliman, M. Eshra, T. Saeed, A. Hamad, A. Abou-Ali
Summary: Analysis of FAERS data revealed significant safety signals for several ALK inhibitors, including risks related to eye disorders, respiratory disorders, and metabolism disorders. Further regulatory investigation is needed to verify the significance of these signals and potentially update product labels.
Article
Public, Environmental & Occupational Health
Elizabeth A. Suarez, Michael Nguyen, Di Zhang, Yueqin Zhao, Danijela Stojanovic, Monica Munoz, Jane Liedtka, Abby Anderson, Wei Liu, Inna Dashevsky, David Cole, Sandra DeLuccia, Talia Menzin, Jennifer Noble, Judith C. Maro
Summary: The US FDA places great importance on monitoring the safety of medications used during pregnancy. Traditional methods like pregnancy exposure registries and cohort studies have limitations, such as small sample sizes and limited outcome assessment. TreeScan, a statistical data mining tool, can simultaneously identify potential adverse outcomes for neonates and infants after maternal medication exposure. This study applied TreeScan to compare the use of fluoroquinolones and cephalosporins during the first trimester and did not observe any new safety signals. TreeScan, combined with tailored or high-dimensional propensity scores, is a valuable tool for safety surveillance of medications used during pregnancy.
PHARMACOEPIDEMIOLOGY AND DRUG SAFETY
(2023)
Article
Computer Science, Information Systems
David Lyell, Ying Wang, Enrico Coiera, Farah Magrabi
Summary: This study analyzed 266 safety events involving machine learning medical devices reported to the US FDA between 2015 and October 2021. Most problems were caused by the devices themselves, but use problems were more likely to cause harm. Data input issues were the top contributor to events. The study highlights the importance of a whole-system approach to safety and emphasizes the interaction between users and devices.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Computer Science, Artificial Intelligence
Xiaotong Wu, Jiaquan Gao, Muhammad Bilal, Fei Dai, Xiaolong Xu, Lianyong Qi, Wanchun Dou
Summary: With the explosive development of the Internet of Things, it has become convenient and important to collect health data and construct medical knowledge graphs. However, there is a risk of sensitive information leakage. This paper proposes the use of federated learning to construct a knowledge graph with local differential privacy for epidemic risk surveillance.
Article
Medicine, General & Internal
Michael Bretthauer, Sara Gerke, Cesare Hassan, Omer F. Ahmad, Yuichi Mori
Summary: The European Union has implemented stricter regulations for medical devices through the Medical Device Regulation (MDR), which includes increased requirements for clinical trials and surveillance throughout the device's life cycle. New expert panels have been established to assess devices for certification, and the role of previous notified bodies has been expanded. All existing medical devices must be recertified under the MDR, with a deadline extension to 2027 or 2028. Meeting these new requirements may be uncertain for most manufacturers and the MDR is expected to have significant consequences for various stakeholders.
ANNALS OF INTERNAL MEDICINE
(2023)
Article
Computer Science, Information Systems
Hongjiao Wu, Ashutosh Dhar Dwivedi, Gautam Srivastava
Summary: In this article, a method to protect private information in medical systems using blockchain technology is designed, incorporating the Patient-oriented Privacy Preserving Access Control model and file authorization contracts to ensure the security of private information. Simulation results demonstrate strong performance in information storage and transmission efficiency of the method.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Article
Health Care Sciences & Services
Sanket S. Dhruva, Jonathan J. Darrow, Aaron S. Kesselheim, Rita F. Redberg
Summary: The increasing availability of drugs and medical devices approved through expedited pathways has clinical and ethical implications. Although expedited development and review have advantages, there is concern about physician's awareness and communication with patients about uncertainties in knowledge and the use of surrogate measures. Effective approaches are needed to ensure the appropriate clinical use of drugs and devices approved through expedited pathways.
JOURNAL OF GENERAL INTERNAL MEDICINE
(2022)
Article
Engineering, Multidisciplinary
Guangyu Peng, Aiqing Zhang, Xiaodong Lin
Summary: This study proposes a privacy-preserving EMR sharing architecture based on blockchain technology to address the issue of privacy leakage in cloud-based storage. Through a dual-blockchain system and identity-based tripartite authentication key agreement scheme, patients have fine-grained control over their EMR access, ensuring trust between healthcare institutions and supervising doctors' identities. The results of the study demonstrate that the proposed protocol is secure, efficient, and suitable for the EMR sharing of comatose patients.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
