Federated learning for violence incident prediction in a simulated cross-institutional psychiatric setting

Borger, Thomas; Mosteiro, Pablo; Kaya, Heysem; Rijcken, Emil; Salah, Albert Ali; Scheepers, Floortje; Spruit, Marco

doi:10.1016/j.eswa.2022.116720

Cited by 22 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FL is a technique to train algorithms across multiple decentralized servers holding local data samples without exchanging them. A simulation study using federated learning within psychiatry shows promising results as the federated model outperforms local models and performs similarly to a data-centralized model (Borger et al, 2022). For these reasons, we expect the field's performance to increase from using Transformer-based language models and federated learning.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Psychiatry - A Systematic Literature Review

Rijcken

Zervanou

Mosteiro

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Throughout the history of artificial intelligence, various algorithm branches have predominantly been used at different times. The last decade has been characterized by a shift from rule-based methods to self-learning methods. However, while the shift towards using ML methods is evident, there is no comparison of both methods for document classification. This systematic literature review focuses on the document classification in healthcare notes from electronic health records within psychiatry. We assess how these methods compare to each other in terms of classification performance and how they have developed throughout time, and we discuss potential directions of the field. We find that rule-based methods have had a higher performance for most of the last decade than machine-learning methods.Yet, the shift in representation techniques and algorithms used in recent years resulted in machine learning methods performing better.Dense document representation techniques, with mostly non-zero cells, outperform sparse representation techniques, with mostly zeros. Also, many neural networks outperform other self-learning- and rule-based methods. We find that state-of-the-art language models are barely employed in the psychiatric domain and expect an increase in the application of federated learning can increase the data availability for model training.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Psychiatry - A Systematic Literature Review

Rijcken

Zervanou

Mosteiro

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…FL is a technique to train algorithms across multiple decentralized servers holding local data samples without exchanging them. A simulation study using federated learning within mental healthcare shows promising results as the federated model outperforms local models and performs similarly to a data-centralized model [101]. For these reasons, we expect the field's performance to increase from using federated learning.…”

Section: Data Availabilitymentioning

confidence: 99%

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Mental Health Care - A Systematic Literature Review

Rijcken,

Zervanou,

Mosteiro

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Document classification is a widely used approach for analysing mental healthcare texts. This systematic literature review focuses on document classification in healthcare notes obtained from electronic health records within mental health care. We observe that the last decade has been characterized by a shift from rule-based methods to machine-learning methods. However, while the shift towards machine-learning methods is evident, there is currently no systematic comparison of both methods for document classification in applications in mental healthcare. In this work, we perform a systematic literature review to assess how these methods compare in terms of performance, which are the specific applications and tasks, and how the approaches have developed throughout time. We find that for most of the last decade, rule-based methods have performed better than machine-learning methods. However, recent developments towards healthcare data availability in combination with self-learning neural networks and transformer-based large language models result in higher performance.

show abstract

“…[56] Image Melanoma FL was used in a study on melanoma detection using skin images. [57,58] Free text Psychiatric disease The potential of FL for clinical psychiatry was highlighted.…”

Section: Data Typesmentioning

confidence: 99%

Survey of Medical Applications of Federated Learning

Choi,

Cha,

Lee

et al. 2024

Healthc Inform Res

View full text Add to dashboard Cite

Objectives: Medical artificial intelligence (AI) has recently attracted considerable attention. However, training medical AI models is challenging due to privacy-protection regulations. Among the proposed solutions, federated learning (FL) stands out. FL involves transmitting only model parameters without sharing the original data, making it particularly suitable for the medical field, where data privacy is paramount. This study reviews the application of FL in the medical domain.Methods: We conducted a literature search using the keywords "federated learning" in combination with "medical," "healthcare," or "clinical" on Google Scholar and PubMed. After reviewing titles and abstracts, 58 papers were selected for analysis. These FL studies were categorized based on the types of data used, the target disease, the use of open datasets, the local model of FL, and the neural network model. We also examined issues related to heterogeneity and security.Results: In the investigated FL studies, the most commonly used data type was image data, and the most studied target diseases were cancer and COVID-19. The majority of studies utilized open datasets. Furthermore, 72% of the FL articles addressed heterogeneity issues, while 50% discussed security concerns.Conclusions: FL in the medical domain appears to be in its early stages, with most research using open data and focusing on specific data types and diseases for performance verification purposes. Nonetheless, medical FL research is anticipated to be increasingly applied and to become a vital component of multi-institutional research.

show abstract

Federated learning for violence incident prediction in a simulated cross-institutional psychiatric setting

Cited by 22 publications

References 23 publications

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Psychiatry - A Systematic Literature Review

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Psychiatry - A Systematic Literature Review

Machine Learning vs. Rule-Based Methods for Document Classification of Electronic Health Records within Mental Health Care - A Systematic Literature Review

Survey of Medical Applications of Federated Learning

Contact Info

Product

Resources

About