Search Articles

View query in Help articles search

Search Results (1 to 5 of 5 Results)

Download search results: CSV END BibTex RIS


Framework for Ranking Machine Learning Predictions of Limited, Multimodal, and Longitudinal Behavioral Passive Sensing Data: Combining User-Agnostic and Personalized Modeling

Framework for Ranking Machine Learning Predictions of Limited, Multimodal, and Longitudinal Behavioral Passive Sensing Data: Combining User-Agnostic and Personalized Modeling

In this paper, we present our ML modeling and ranking framework to address these challenges. The framework is designed to induce improved predictions for multimodal sensing. It balances both user-agnostic and personalized modeling of small data sets encountered often in mental and physical health–based studies.

Tahsin Mullick, Sam Shaaban, Ana Radovic, Afsaneh Doryab

JMIR AI 2024;3:e47805

Dashboard With Bump Charts to Visualize the Changes in the Rankings of Leading Causes of Death According to Two Lists: National Population-Based Time-Series Cross-Sectional Study

Dashboard With Bump Charts to Visualize the Changes in the Rankings of Leading Causes of Death According to Two Lists: National Population-Based Time-Series Cross-Sectional Study

Ranking of the 10 leading causes of death for both sexes of all ages in the United States according to (A) the World Health Organization (WHO) list and (B) the National Center for Health Statistics (NHCS) list. This dashboard is accessible [17].

Shu-Yu Tai, Ying-Chen Chi, Yu-Wen Chien, Ichiro Kawachi, Tsung-Hsueh Lu

JMIR Public Health Surveill 2023;9:e42149

Treatment Discontinuation Prediction in Patients With Diabetes Using a Ranking Model: Machine Learning Model Development

Treatment Discontinuation Prediction in Patients With Diabetes Using a Ranking Model: Machine Learning Model Development

We designed a prediction model of TD as a ranking problem with imbalanced data to compare patients by length of time until TD. The ranking problem [26] is an application of survival time analysis [27]. Cox regression [28] is generally used in statistical analysis, whereas the ranking model is used in ML [29-31]. Cox regression is a model of the hazard function in which the effects of the explanatory variables on outcomes are predetermined, requiring an assumption that they remain constant over time [28].

Hisashi Kurasawa, Kayo Waki, Akihiro Chiba, Tomohisa Seki, Katsuyoshi Hayashi, Akinori Fujino, Tsuneyuki Haga, Takashi Noguchi, Kazuhiko Ohe

JMIR Bioinform Biotech 2022;3(1):e37951

Using Automated Machine Learning to Predict the Mortality of Patients With COVID-19: Prediction Model Development Study

Using Automated Machine Learning to Predict the Mortality of Patients With COVID-19: Prediction Model Development Study

After ranking variables by importance, we chose the top 10 most influential variables and retrained the auto ML models to generate new models that only used these 10 variables. This was done to create high-performing models with low dimensionality. In addition, we sought to provide interpretable black-box model results to clinicians and patients.

Kenji Ikemura, Eran Bellin, Yukako Yagi, Henny Billett, Mahmoud Saada, Katelyn Simone, Lindsay Stahl, James Szymanski, D Y Goldstein, Morayma Reyes Gil

J Med Internet Res 2021;23(2):e23458

The Anatomy of the SARS-CoV-2 Biomedical Literature: Introducing the CovidX Network Algorithm for Drug Repurposing Recommendation

The Anatomy of the SARS-CoV-2 Biomedical Literature: Introducing the CovidX Network Algorithm for Drug Repurposing Recommendation

As search capabilities advanced, ranking algorithms also emerged. Luo et al [21] presented a ranking algorithm to produce a drug repurposing recommendation system. Karatszas et al [22] presented a web-based tool, which they called composite drug reranking scoring, to identify the most promising drugs and chemical substances to test.

Lyndsey Elaine Gates, Ahmed Abdeen Hamed

J Med Internet Res 2020;22(8):e21169