Review of Clustering Methods for Functional Data

Zhang, Mimi; Parnell, Andrew

doi:10.1145/3581789

Cited by 15 publications

(2 citation statements)

References 122 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach enables a deeper understanding of patient groups and helps select the optimal treatment for patients with specific characteristics [ 29 ]. In clinical medicine, considering which cluster a patient belongs to when planning treatment can lead to predicting treatment response and improving patients' Quality of Life (QOL) while maximizing treatment effectiveness [ 30 ].…”

Section: Discussionmentioning

confidence: 99%

A Discrete Mathematics Approach for Understanding Risk Factors in Overactive Bladder Treatment

Okui

2024

Cureus

View full text Add to dashboard Cite

Introduction Discrete mathematics, a branch of mathematics that includes graph theory, combinatorics, and logic, focuses on discrete mathematical structures. Its application in the medical field, particularly in analyzing patterns in patient data and optimizing treatment methods, is invaluable. This study, focusing on post-void residual (PVR) urine following overactive bladder (OAB) treatment, utilized discrete mathematics techniques to analyze PVR and its associated risk factors. Methods A retrospective study was conducted on 128 OAB patients who received intradetrusor onabotulinum toxin A injections between 2020 and 2022. Network graphs based on graph theory were used to analyze correlations between clinical variables, and clustering analysis was performed with PVR as the primary variable. Results The network graph analysis revealed that frailty, daytime frequency, and nocturia episodes were closely related to PVR. Clustering analysis with PVR as the primary variable divided the patients into three groups, suggesting that the group with particularly high frailty (Cluster 1) is at high risk for PVR. Moreover, significant differences in clinical indicators such as age, voiding efficiency, Overactive Bladder Symptom Score, and International Consultation on Incontinence Questionnaire-Short Form were observed in the remaining two clusters (Cluster 0 and 2). Conclusion This study demonstrates the effectiveness of discrete mathematics methods in identifying risk factors for PVR after OAB treatment and in distinguishing clinical subgroups based on patient characteristics. This approach could contribute to the formulation of individualized treatment strategies and the improvement of patient care quality. Further development and clinical application of this methodology are expected in future research.

show abstract

Section: Discussionmentioning

confidence: 99%

A Discrete Mathematics Approach for Understanding Risk Factors in Overactive Bladder Treatment

Okui

2024

Cureus

View full text Add to dashboard Cite

show abstract

“…Thirdly, the presented algorithm assumes a homogeneous distribution of the latent population, and when multiple heterogeneous subpopulations are mixed, subpopulations must be identified and stratified. In this regard, future tasks include identifying subgroups based on dimension reduction 35 and clustering 36,37 and developing methods to test for heterogeneity based on functional data representation.…”

Section: Discussionmentioning

confidence: 99%

Unbiased estimation of the population-level motor module

Matsui

Uno

Nojima

2023

Preprint

View full text Add to dashboard Cite

Motor module is a functional neurophysiological command for muscle coordination. In clinical settings, population-level characterization and comparison of motor modules are necessary to evaluate pathophysiological mechanisms and intervention effects. Previous studies have estimated individual motor modules and then compared them, but the validity of capturing the distribution of the latent population has not been fully understood. Our study aimed to address this issue by investigating the accuracy of estimating the population mean of motor modules. Through simulation experiments, we found that previous individual-based approach did not converge regardless of sample size and was vulnerable to noise. We developed an unbiased estimation algorithm using the framework of functional data analysis, which significantly improved estimation accuracy. Our findings highlight statistical challenges for motor module analysis and suggest the need for further research on new computational algorithms using large-scale clinical data.

show abstract

Functional Data Analysis: An Introduction and Recent Developments

Gertheiss,

Rügamer,

Liew

et al. 2024

Biometrical J

View full text Add to dashboard Cite

Functional data analysis (FDA) is a statistical framework that allows for the analysis of curves, images, or functions on higher dimensional domains. The goals of FDA, such as descriptive analyses, classification, and regression, are generally the same as for statistical analyses of scalar‐valued or multivariate data, but FDA brings additional challenges due to the high‐ and infinite dimensionality of observations and parameters, respectively. This paper provides an introduction to FDA, including a description of the most common statistical analysis techniques, their respective software implementations, and some recent developments in the field. The paper covers fundamental concepts such as descriptives and outliers, smoothing, amplitude and phase variation, and functional principal component analysis. It also discusses functional regression, statistical inference with functional data, functional classification and clustering, and machine learning approaches for functional data analysis. The methods discussed in this paper are widely applicable in fields such as medicine, biophysics, neuroscience, and chemistry and are increasingly relevant due to the widespread use of technologies that allow for the collection of functional data. Sparse functional data methods are also relevant for longitudinal data analysis. All presented methods are demonstrated using available software in R by analyzing a dataset on human motion and motor control. To facilitate the understanding of the methods, their implementation, and hands‐on application, the code for these practical examples is made available through a code and data supplement and on GitHub.

show abstract

Review of Clustering Methods for Functional Data

Cited by 15 publications

References 122 publications

A Discrete Mathematics Approach for Understanding Risk Factors in Overactive Bladder Treatment

A Discrete Mathematics Approach for Understanding Risk Factors in Overactive Bladder Treatment

Unbiased estimation of the population-level motor module

Functional Data Analysis: An Introduction and Recent Developments

Contact Info

Product

Resources

About