Search Articles

View query in Help articles search

Search Results (1 to 10 of 41 Results)

Download search results: CSV END BibTex RIS


Authors’ Response to Peer Reviews of “Beyond Expected Patterns in Insulin Needs of People With Type 1 Diabetes: Temporal Analysis of Automated Insulin Delivery Data”

Authors’ Response to Peer Reviews of “Beyond Expected Patterns in Insulin Needs of People With Type 1 Diabetes: Temporal Analysis of Automated Insulin Delivery Data”

It employs various time series techniques to spot such patterns using matrix profiles and multivariate clustering. The Open APS Data Commons dataset, an extensive dataset collected in real-life conditions, was analyzed to discover temporal patterns in insulin need driven by well-known factors such as carbohydrates and potentially novel factors. The results are limited to disclosing interesting temporal patterns in insulin need that cannot be explained solely by carbohydrates through the performed analysis.

Isabella Degen, Kate Robson Brown, Henry W J Reeve, Zahraa S Abdallah

JMIRx Med 2024;5:e66643

Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study

Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study

The third novelty is in a new performance measure we apply to a conventional clustering algorithm to identify clusters that maximize the minority-to-majority (PE to non-PE) ratio and thereby help us focus on PE risk factors and groups of patients at risk on hospital admission from the additional angle of a clustering measure.

Ori Ben Yehuda, Edward Itelman, Adva Vaisman, Gad Segal, Boaz Lerner

J Med Internet Res 2024;26:e48595

User Engagement Clusters of an 8-Week Digital Mental Health Intervention Guided by a Relational Agent (Woebot): Exploratory Study

User Engagement Clusters of an 8-Week Digital Mental Health Intervention Guided by a Relational Agent (Woebot): Exploratory Study

The responses were transformed to a numeric Likert scale (1-5) for inclusion in the clustering models. Analyses focused on 4 mental health outcome variables of interest. Because only about half of the sample had clinically significant levels of depressive or anxiety symptoms at baseline, 2 more general measures of mental health wellness were selected for analysis: stress and resilience.

Valerie Hoffman, Megan Flom, Timothy Y Mariano, Emil Chiauzzi, Andre Williams, Andrew Kirvin-Quamme, Sarah Pajarito, Emily Durden, Olga Perski

J Med Internet Res 2023;25:e47198

Beyond Expected Patterns in Insulin Needs of People With Type 1 Diabetes: Temporal Analysis of Automated Insulin Delivery Data

Beyond Expected Patterns in Insulin Needs of People With Type 1 Diabetes: Temporal Analysis of Automated Insulin Delivery Data

To group similar days, we used time series k-means clustering. To prevent bias from different measurement scales, we applied min-max scaling of IOB, COB, and IG values to a range of 0 to 10 for each participant.

Isabella Degen, Kate Robson Brown, Henry W J Reeve, Zahraa S Abdallah

JMIRx Med 2024;5:e44384

Drug Recommendation System for Diabetes Using a Collaborative Filtering and Clustering Approach: Development and Performance Evaluation

Drug Recommendation System for Diabetes Using a Collaborative Filtering and Clustering Approach: Development and Performance Evaluation

Clustering is one of the most widely used machine-learning techniques in the field of health to identify patterns or groups of patients with similar characteristics [16,17]. Although the clustering technique has been the subject of research in the area of RSs, these systems have not yet been widely used in medicine.

Luis Fernando Granda Morales, Priscila Valdiviezo-Diaz, Ruth Reátegui, Luis Barba-Guaman

J Med Internet Res 2022;24(7):e37233

Behavior of Callers to a Crisis Helpline Before and During the COVID-19 Pandemic: Quantitative Data Analysis

Behavior of Callers to a Crisis Helpline Before and During the COVID-19 Pandemic: Quantitative Data Analysis

The call data were subjected to k-means clustering to discover the types of callers that used the service. In k-means clustering, data points are grouped together based on their closeness by Euclidean distance. In other words, the aim is to find k groups in n objects based on the similarity of their characteristics, where the characteristics in one group show high similarity with each other but low similarity with other groups [16,17].

Robin Turkington, Maurice Mulvenna, Raymond Bond, Edel Ennis, Courtney Potts, Ciaran Moore, Louise Hamra, Jacqui Morrissey, Mette Isaksen, Elizabeth Scowcroft, Siobhan O'Neill

JMIR Ment Health 2020;7(11):e22984

Filtering Entities to Optimize Identification of Adverse Drug Reaction From Social Media: How Can the Number of Words Between Entities in the Messages Help?

Filtering Entities to Optimize Identification of Adverse Drug Reaction From Social Media: How Can the Number of Words Between Entities in the Messages Help?

This method was evaluated by the comparison of 4 benchmark methods (example adaption for text categorization [EAT], positive examples and negative examples labeling heuristics [PNLH], active semisupervised clustering based two-stage text classification [ACTC], and Laplacian SVM) and the calculation of F scores (the harmonic mean of precision and recall) on ADRs posts. These 4 methods were improved by the use of this approach. The F score gains fluctuated between 1.94% and 6.14%.

Redhouane Abdellaoui, Stéphane Schück, Nathalie Texier, Anita Burgun

JMIR Public Health Surveill 2017;3(2):e36