China
Africa
Explore chapters and articles related to this topic
Online Communities and Social Networks
Published in Rupa S. Valdez, Richard J. Holden, The Patient Factor, 2021
Annie T. Chen, Albert Park, Andrea L. Hartzler
In addition, OCSNs can help patients to find and retrieve relevant health information by integrating extant research literature, computational techniques, and human-centered design principles. Extant research literature pertaining to patient information needs at different temporal points and in different patient work contexts could inform the selection of topics of interest. Systems can be designed to weave expert opinions from health professionals into OCSNs discussions (Huh & Pratt, 2014). Computational techniques could be used to identify content containing potentially unreliable or inaccurate information (Starbird et al., 2016; Hou et al., 2019). Cluster analysis could be used to identify and present topics of interest to patients (Chen, 2012). In addition, human-centered design techniques could be employed to iteratively develop visual indicators that facilitate patients’ interpretation of information quality and relevance.
A Typology of Service Patterns in End-Stage AIDS Care: Relationships to the Transprofessional Model
Published in David Alex Cherin, G. J. Huba, AIDS Capitation, 2021
G. J. Huba, Diana E. Brief, David A. Cherin, A. T. Panter, Lisa A. Melchior
Service Profiles for Patients with AIDS. To identify a service-usage profile for end-stage AIDS patients, cluster analysis was performed. Cluster analysis is a method for taking a group of individuals, each of whom has scores on a number of variables, and then asking what the “clusters” or groups of individuals are based upon the individuals’ scores on all of the variables. Thus, cluster analysis is a method of empirically generating a typology. In this case, we are clustering utilization rates of a number of types of end-stage AIDS-care indices. Thus, people who are identified as being in the same cluster will tend to have the same pattern of service utilization. To interpret the clusters of people obtained, it is typical to calculate the mean for the group of people in the cluster on each of the variables. Thus, cluster analysis is a method in which one looks at the scores of many individuals on many variables and identifies groups (or clusters) of people who tend to have the same patterns on the variables. Then, the mean-score profile for the people in the cluster is calculated and compared to the mean score profiles for the other identified groups. While there are numerous empirical ways to determine the clustering of individuals from their variable scores, one method that tends to work well with service utilization data is Ward’s method (see Hartigan, 1975).
Developing and Validating a Clinical Case Definition for the Fibromyalgia Syndrome for Use in Clinical Practice
Published in I. Jon Russell, The Fibromyalgia Syndrome: A Clinical Case Definition for Practitioners, 2020
All data collected then would be entered into detailed cluster analysis to determine clusters of patients and the determinants of each cluster. This analysis could lead to identification of more than one FMS-like cluster. Hence, the results would need to go back to the initial content validation committee to group potential clusters and to provide a final, testable case definition for fibromyalgia. An advantage of this ap proach is that investigators may be able to identify FMS subgroups which may exhibit different features of natural history, different associations with co-morbid conditions, different risk factors, different laboratory abnormalities, and different responses to treatment.
The electronic tongue: an advanced taste-sensing multichannel sensory tool with global selectivity for application in the pharmaceutical and food industry
Published in Pharmaceutical Development and Technology, 2023
PCA is a technique used to emphasize variation and bring out strong patterns in a dataset. It is often used to make data comparatively easier to explore and visualize. DFA is used to determine which variables discriminate between two or more naturally occurring groups. Cluster Analysis is the process to find similar groups of objects in order to form clusters. It is an unsupervised machine learning-based algorithm that acts on unlabelled data. A group of data points would comprise together to form a cluster in which all the objects would belong to the same group. PLS is a comparatively quicker, efficient, and optimal regression method based on covariance. It is recommended in cases of regression where the number of explanatory variables is high, and where it is likely that there is multicollinearity among the variables, i.e. that the explanatory variables are correlated. It is the most widely used method to obtain calibration models for numerical predictions of various quality parameters (Kirsanov et al. 2019; Lozano-Torres et al. 2022).
Different clusters of perfectionism in inpatients with anorexia nervosa and healthy controls
Published in Eating Disorders, 2022
Paola Longo, Matteo Aloi, Nadia Delsedime, Marianna Rania, Cristina Segura Garcia, Giovanni Abbate-Daga, Enrica Marzola
Cluster analyses, hierarchical and k-means ones, were performed to classify participants according to perfectionism scores. Firstly, we conducted hierarchical cluster analysis on patients group to decide the number of clusters to adopt. Hierarchical cluster analysis includes a series of steps in which cases are joined together according to their similarities. Each step of the process is described in a dendrogram, a schematic part of the analysis output. Examining the output the researcher can detect the appropriate number of clusters to adopt (Clatworthy et al., 2005). In particular, we used the process of tree cutting, included in the stopping rules, which consists of cutting the dendrogram at the stage where the cluster coefficients rise exponentially, namely at the step in which the dissimilarity increases inconsistently (Clatworthy et al., 2005; Everitt, 2005); then we applied the formula [number of cases—number of stages = number of clusters]. Then k-means cluster analysis was run introducing the number of clusters detected by the previous analysis also for the HCs group to make the study homogenous. K-means enables to create a new categorical variable assigning each subject to a cluster; this variable was used for the following analysis.
Early Sexual Debut and the Effects on Well-Being among South African Adolescent Girls and Young Women Aged 15 to 24 Years
Published in International Journal of Sexual Health, 2022
Tracy McClinton Appollis, Kim Jonas, Roxanne Beauclair, Carl Lombard, Zoe Duby, Mireille Cheyip, Kealeboga Maruping, Janan Dietrich, Catherine Mathews
Demographic information was obtained by asking questions around age, marital status, education and orphanhood status. A binary indicator of socio-economic status (SES) was created using a Cluster Analysis with the K-modes algorithm. Cluster Analysis is a machine learning technique that allows analysts to group data based upon shared features. The features used in the creation of this indicator were: (1) AGYW was away from home for more than one month in the past 12 months; (2) piped water in household; (3) flushing toilet(s) in household; (4) working electricity in household; (5) household has car; (6) household has computer; (7) household has access to internet; (8) household has refrigerator; (9) household has stove; (10) AGYW went a day/night without eating in past month; (11) AGYW has own money; (12) AGYW saves money; (13) AGYW owes money. These variables were grouped into clusters by the algorithm indicating those with a relatively low SES and those with a high SES.