Article
Computer Science, Artificial Intelligence
Sang Ho Oh, Jongyoul Park, Su Jin Lee, Seungyeon Kang, Jeonghoon Mo
Summary: In this study, the researchers aim to develop a reinforcement learning-based treatment recommendation model using electronic health records of South Korean patients. By considering various details, the proposed contextual bandits model offers a practical solution to address clinical challenges in prescribing diabetes medication.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Elsie M. F. Horne, Susannah McLean, Mohammad A. Alsallakh, Gwyneth A. Davies, David B. Price, Aziz Sheikh, Athanasios Tsanas
Summary: This study defined and validated asthma subtypes using large longitudinal primary care electronic health records (EHRs). The results showed that asthma subtypes were primarily defined by the level of steroid use, level of healthcare utilization, and the presence of comorbidities. This has important clinical implications for defining asthma subtypes, facilitating patient stratification, and developing more personalized monitoring and treatment strategies.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Jeong Min Lee, Milos Hauskrecht
Summary: Clinical event sequences are crucial for patient care, but predicting them accurately is challenging due to patient-specific variability. In this study, we propose multiple event sequence prediction models that focus on refining population-wide models to individual patients and their unique conditions. We analyze and test the performance of these models on clinical event sequences from the MIMIC-III database.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2023)
Review
Computer Science, Information Systems
Kevin B. Johnson, Michael J. Neuss, Don Eugene Detmer
Summary: This study aims to provide an introductory tutorial on the history of medical documentation, sources of clinician burnout, and opportunities to improve electronic health records (EHRs). It highlights both the successes and current limitations of EHRs, and suggests that improving the clinician's experience through collaboration and policy changes can address burnout.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2021)
Article
Health Care Sciences & Services
Laura Ryan, Kelly A. Weir, Jessica Maskell, Lily Bevan, Robyne Le Brocque
Summary: This study aimed to understand clinician attitudes and behaviors when a patient covertly records a hospital clinical encounter using a smartphone. Interviews with 20 hospital clinicians revealed that most had either suspected or experienced such recordings. Findings suggest the need for nuanced strategies to support clinicians in managing covert recordings while balancing patient needs.
Article
Pharmacology & Pharmacy
Pierpaolo Pellicori, Alex McConnachie, Christopher Carlin, Ann Wales, John G. F. Cleland
Summary: Using electronic health records, a pragmatic and parsimonious multivariable model was developed to predict 90-day mortality in patients hospitalized with chronic obstructive pulmonary disease (COPD). The model, based on a small number of variables including age, sex, length of hospital stay, prior diagnosis, prescription, and laboratory data, showed excellent calibration and reasonable discrimination. The risk-calculator can be useful for service-evaluation, clinical management, risk-stratification, and patient selection for clinical research.
PHARMACOLOGICAL RESEARCH
(2022)
Article
Health Care Sciences & Services
Xi Yang, Aokun Chen, Nima PourNejatian, Hoo Chang Shin, Kaleb E. Smith, Christopher Parisien, Colin Compas, Cheryl Martin, Anthony B. Costa, Mona G. Flores, Ying Zhang, Tanja Magoc, Christopher A. Harle, Gloria Lipori, Duane A. Mitchell, William R. Hogan, Elizabeth A. Shenkman, Jiang Bian, Yonghui Wu
Summary: This study develops a large clinical language model and evaluates it on five clinical NLP tasks. By scaling up the number of parameters and increasing the size of the training data, the model improves accuracy and shows potential for enhancing medical AI systems.
NPJ DIGITAL MEDICINE
(2022)
Article
Business
Zheming Zuo, Jie Li, Han Xu, Noura Al Moubayed
Summary: Disruptive technologies play a crucial role in pervasive healthcare, with machine learning widely utilized in patient-centric solutions. The feature selection method CFS has shown outstanding performance in analyzing EHR data.
TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE
(2021)
Editorial Material
Health Care Sciences & Services
Roseann S. Gammal, Lucas A. Berenbrok, Philip E. Empey, Mylynda B. Massart
Summary: The article highlights the importance of documenting pharmacogenomic test results in a patient's electronic health record, especially in the absence of dedicated clinical pharmacogenomics teams and support. Practical tips are provided for clinicians to maximize the visibility and utility of these results over time.
JOURNAL OF PERSONALIZED MEDICINE
(2021)
Article
Multidisciplinary Sciences
Kenney Ng, Uri Kartoun, Harry Stavropoulos, John A. Zambrano, Paul C. Tang
Summary: A machine-learning precision cohort treatment option workflow was developed to support point-of-care decision making by presenting outcomes of past treatment choices for cohorts of similar patients, based on observational data from electronic health records. The workflow demonstrated that better treatment options were available for a majority of cases, and models were deployed in a pilot study with primary care physicians for hypertension and type 2 diabetes mellitus. By creating patient-similarity models, personalized treatment insights can be generated dynamically at the point-of-care, integrating knowledge-driven treatment guidelines with data-driven EHR data for medical decision-making.
SCIENTIFIC REPORTS
(2021)
Article
Public, Environmental & Occupational Health
Nadia M. Penrod, Jason H. Moore
Summary: Yoga may be an effective strategy for blood pressure control and the prevention of hypertension at the population level, according to a retrospective observational study.
Article
Computer Science, Information Systems
Charlotte A. Nelson, Riley Bove, Atul J. Butte, Sergio E. Baranzini
Summary: By embedding individual patient data into a biomedical knowledge graph, SPOKEsigs can effectively classify patients with chronic diseases, outperforming predictions using EHRs alone.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2022)
Review
Medicine, General & Internal
Jillian C. Ryan, John Noel Viana, Hamza Sellak, Shakuntla Gondalia, Nathan O'Callaghan
Summary: Precision health is a new field that requires clearer definition and differentiation from precision medicine. This study aims to conduct a scoping review to define precision health and map research in this area. By analyzing data from scientific databases and grey literature sources, the study will identify gaps and future directions for precision health research.
Article
Health Care Sciences & Services
Andre de Wit, John de Heide, Paul Cummins, Ada van Bruchem-van de Scheur, Rohit Bhagwandien, Mattie Lenzen
Summary: With the shorten duration of hospital admissions, providing necessary information and instructions to patients within a limited time becomes a challenge for healthcare professionals. This study aimed to evaluate patient comprehension of discharge information using a computer-generated patient-tailored discharge document. The results showed that the personalized discharge tool could potentially lead to improved care, especially for patients who initiated teleconsultation post-discharge. However, further research with a larger sample size is needed to confirm this trend.
Article
Multidisciplinary Sciences
Vivek Rudrapatna, Yao-Wen Cheng, Colin Feuille, Arman Mosenia, Jonathan Shih, Yongmei Shi, Olivia J. Roberson, Benjamin Rubin, Atul Butte, Uma Mahadevan, Nicholas Skomrock, Ngozi Erondu, Christel Chehoud, Saquib Rahim, David S. Apfel, Mark Curran, Najat Khan, Christopher D. O'Brien, Natalie V. Terry, Benjamin V. Martini, Sreeram V. Ramagopalan
Summary: This study developed an external control arm for Crohn's disease using electronic health records (EHR) data and a combination of informatics and manual methods. However, significant missing data was found when standard-of-care clinical data were repurposed. Further improvements are needed to align trial design with typical patterns of clinical practice, enabling more robust external control arms in chronic diseases like Crohn's disease.
Editorial Material
Mathematical & Computational Biology
Jason H. Moore, Ian Barnett, Mary Regina Boland, Yong Chen, George Demiris, Graciela Gonzalez-Hernandez, Daniel S. Herman, Blanca E. Himes, Rebecca A. Hubbard, Dokyoon Kim, Jeffrey S. Morris, Danielle L. Mowery, Marylyn D. Ritchie, Li Shen, Ryan Urbanowicz, John H. Holmes
Article
Obstetrics & Gynecology
Snigdha Alur-Gupta, Mary Regina Boland, Kurt T. Barnhart, Mary D. Sammel, Anuja Dokras
Summary: The study reveals that women with polycystic ovary syndrome are at increased risk of cardiovascular and psychiatric complications during the postpartum period, showing significantly higher odds of cardiovascular disease complications and depression than those without the syndrome.
AMERICAN JOURNAL OF OBSTETRICS AND GYNECOLOGY
(2021)
Review
Biochemical Research Methods
Lena Davidson, Mary Regina Boland
Summary: Through a systematic review, it was found that supervised learning methods are more popular in the field of artificial intelligence and machine learning, and AI and ML methods are mainly used in prenatal care, perinatal care, and preterm birth in the pregnancy domain. Future research should focus on less-studied areas such as postnatal and postpartum care, and more emphasis should be placed on the clinical adoption of AI methods and the ethical implications of such adoption.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Public, Environmental & Occupational Health
Jessica R. Meeker, Heather Burris, Mary Regina Boland
Summary: The study aimed to develop an algorithm called REMAP to identify residential mobility events during pregnancy and avoid exposure misclassification. Results showed that REMAP was 95.7% accurate and revealed that 41% of patients in the large urban cohort moved during pregnancy. REMAP outperformed the use of ZIP codes alone, with a potential misclassification rate of 39% if residential mobility was not taken into account.
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
(2021)
Article
Obstetrics & Gynecology
Jessica R. Meeker, Silvia P. Canelon, Ray Bai, Lisa D. Levine, Mary Regina Boland
Summary: Individual and neighborhood-level risk factors are associated with severe maternal morbidity, potentially contributing to rising rates in the United States. Recognition of these factors is crucial for developing interventions to reduce severe maternal morbidity and mortality.
OBSTETRICS AND GYNECOLOGY
(2021)
Article
Health Care Sciences & Services
Mary Regina Boland, Lena M. Davidson, Silvia P. Canelon, Jessica Meeker, Trevor Penning, John H. Holmes, Jason H. Moore
Summary: Environmental disasters, such as the Seveso disaster and Fukushima-Daiichi nuclear meltdown, have catastrophic health consequences. Traditional methods for studying these disasters are costly and time-intensive. By utilizing electronic health records and informatics methods, researchers can study the health impacts of emergent environmental disasters in a cost-effective manner, leading to more easily accessible data on large populations with diverse backgrounds.
NPJ DIGITAL MEDICINE
(2021)
Article
Medicine, General & Internal
Silvia P. Canelon, Samantha Butts, Mary Regina Boland
Summary: This study found that pregnant women with sickle cell trait (SCT) have an increased risk of stillbirth.
Article
Multidisciplinary Sciences
Anna Bushong, Thomas McKeon, Mary Regina Boland, Jeffrey Field
Summary: The study found a significant association between the increase in asthma hospitalization rates and unconventional natural gas development (UNGD) in rural counties in Pennsylvania. After analyzing publicly available data, it was discovered that there was a positive relationship between annual well density and asthma hospitalization rates. Additionally, there was a stronger association between asthma hospitalization rates and PM2.5 levels.
Article
Medical Informatics
Lena Davidson, Silvia P. Canelon, Mary Regina Boland
Summary: This study investigated the association between preconception and periconception medication exposure and the risk of multiple gestational births. The results identified several medications that were associated with multiple births and suggested potential candidates for further research, demonstrating the effectiveness of using electronic health record data in medication-wide association studies.
JMIR MEDICAL INFORMATICS
(2022)
Article
Multidisciplinary Sciences
Lena Davidson, Silvia P. Canelon, Mary Regina Boland
Summary: This retrospective study examines the safety profiles of repurposed COVID medication therapies on pregnancy outcomes using electronic health record (EHR) data. The findings suggest that some medications may be considered for future studies involving pregnant individuals, while further investigation is needed for drugs associated with pregnancy outcomes.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Information Systems
Aditya Kashyap, Chris Callison-Burch, Mary Regina Boland
Summary: This study aims to develop two predictive algorithms for opioid prescription and opioid use disorder (OUD) using deep learning models trained on patient Electronic Health Records (EHRs). The results show that these models can accurately predict these challenging outcomes.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2023)
Proceedings Paper
Mathematical & Computational Biology
Silvia P. Canelon, Mary Regina Boland
Summary: This study utilized EHR data from Penn Medicine to assess risk factors for emergency admissions at the time of delivery. The findings indicated that preterm birth, age younger than 25, Black/African American, Asian, and Other/Mixed race patients were at increased risk, while later pregnancies and repeat cesareans decreased the risk. Similar trends were observed for cesarean deliveries, with some variations among different racial groups.
PACIFIC SYMPOSIUM ON BICOMPUTING 2021
(2021)
Article
Computer Science, Information Systems
John H. Holmes, James Beinlich, Mary R. Boland, Kathryn H. Bowles, Yong Chen, Tessa S. Cook, George Demiris, Michael Draugelis, Laura Fluharty, Peter E. Gabriel, Robert Grundmeier, C. William Hanson, Daniel S. Herman, Blanca E. Himes, Rebecca A. Hubbard, Charles E. Kahn, Dokyoon Kim, Ross Koppel, Qi Long, Nebojsa Mirkovic, Jeffrey S. Morris, Danielle L. Mowery, Marylyn D. Ritchie, Ryan Urbanowicz, Jason H. Moore
Summary: This study aims to provide a comprehensive reference for users of EHR in clinical and research settings, as well as for health information systems professionals involved in designing, implementing, and maintaining EHR systems. A panel of 24 experts with extensive experience in EHR systems collaborated to share their knowledge and experience in addressing various challenges related to EHR use in clinical and research contexts. The study highlights key aspects such as usability, data quality, standards, governance, data integration, clinical care, and clinical research.
METHODS OF INFORMATION IN MEDICINE
(2021)
Article
Mathematical & Computational Biology
Owen Wetherbee, Jessica R. Meeker, Caroline DeVoto, Trevor M. Penning, Jason H. Moore, Mary Regina Boland
DATABASE-THE JOURNAL OF BIOLOGICAL DATABASES AND CURATION
(2020)
Article
Computer Science, Information Systems
Rui Duan, Chongliang Luo, Martijn J. Schuemie, Jiayi Tong, C. Jason Liang, Howard H. Chang, Mary Regina Boland, Jiang Bian, Hua Xu, John H. Holmes, Christopher B. Forrest, Sally C. Morton, Jesse A. Berlin, Jason H. Moore, Kevin B. Mahoney, Yong Chen
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2020)