Predicting hepatocellular carcinoma recurrences: A data-driven multiclass classification method incorporating latent variables

Researchers across many domains of psychology increasingly wish to arrive at personalized and generalizable dynamic models of individuals’ processes. This is seen in psychophysiological, behavioral, and emotional research paradigms, across a range of data types. Errors of measurement are inherent in most data. For this reason, researchers typically gather multiple indicators of the same latent construct and use methods, such as factor analysis, to arrive at scores from these indices. In addition to accurately measuring individuals, researchers also need to find the model that best describes the relations among the latent constructs. Most currently available data-driven searches do not include latent variables. We present an approach that builds from the strong foundations of Group Iterative Multiple Model Estimation (GIMME), the idiographic filter, and model implied instrumental variables with two-stage least squares estimation (MIIV-2SLS) to provide researchers with the option to include latent variables in their data-driven model searches. The resulting approach is called Latent Variable GIMME (LV-GIMME). GIMME is utilized for the data-driven search for relations that exist among latent variables. Unlike other approaches such as the idiographic filter, LV-GIMME does not require that the latent variable model to be constant across individuals. This requirement is loosened by utilizing MIIV-2SLS for estimation. Simulated data studies demonstrate that the method can reliably detect relations among latent constructs, and that latent constructs provide more power to detect effects than using observed variables directly. We use empirical data examples drawn from functional MRI and daily self-report data.

Download Full-text

A survival classification method for hepatocellular carcinoma patients with chaotic Darcy optimization method based feature selection

Medical Hypotheses ◽

10.1016/j.mehy.2020.109626 ◽

2020 ◽

Vol 139 ◽

pp. 109626

Author(s):

Fahrettin Burak Demir ◽

Turker Tuncer ◽

Adnan Fatih Kocamaz ◽

Fatih Ertam

Keyword(s):

Hepatocellular Carcinoma ◽

Feature Selection ◽

Optimization Method ◽

Classification Method

Download Full-text

Quantification of Parkinsonian Bradykinesia Based on Axis-Angle Representation and SVM Multiclass Classification Method

IEEE Access ◽

10.1109/access.2018.2835463 ◽

2018 ◽

Vol 6 ◽

pp. 26895-26903 ◽

Cited By ~ 6

Author(s):

Zhirong Lin ◽

Yongsheng Xiong ◽

Guoen Cai ◽

Houde Dai ◽

Xuke Xia ◽

...

Keyword(s):

Multiclass Classification ◽

Classification Method ◽

Svm Multiclass

Download Full-text

EARLY FRAILTY PHENOTYPES AND PREDICTIONS OF COGNITIVE AGING: EVIDENCE FROM THE VICTORIA LONGITUDINAL STUDY

Innovation in Aging ◽

10.1093/geroni/igz038.1077 ◽

2019 ◽

Vol 3 (Supplement_1) ◽

pp. S292-S293

Author(s):

Linzy Bohn ◽

Yao Zheng ◽

G Peggy McFall ◽

Roger A Dixon

Keyword(s):

Longitudinal Study ◽

Cognitive Performance ◽

Cognitive Aging ◽

Latent Variables ◽

Latent Profile Analysis ◽

Cardiovascular Health ◽

Profile Analysis ◽

Community Dwelling ◽

Data Driven ◽

Data Driven Approach

Abstract Frailty is an aging condition that reflects multisystem decline. A prominent approach to frailty assessment is to create an index, whereby responses across multiple indicators of aging systems are summed to create a single score. These studies indicate that frailty is associated with adverse aging outcomes (e.g., mortality, dementia). We employ a data-driven approach to detecting and differentiating emerging frailty phenotypes and examine their associations with non-demented cognitive aging trajectories. Participants (n = 653; M age = 70.6, range 53-95) were community-dwelling older adults from the Victoria Longitudinal Study. Participants contributed (a) baseline data for 30 frailty-related items representing deficits across 7 domains (e.g., instrumental and cardiovascular health) and (b) longitudinal data for latent variables of executive function, speed, and memory. For each participant, we calculated the proportion of deficits present in each frailty-related domain and submitted these data to a latent profile analysis (LPA; Mplus 7.0). We used latent growth modeling (LGM) to test these frailty phenotypes for prediction of cognitive performance and decline. LPA results revealed three profiles, one large normal low-frailty profile and two emerging frailty phenotypes. Whereas the latter represented profiles of individuals with respiratory-type frailty (i.e., marked impairment in respiratory function; 7%) and mobility-type frailty (i.e., marked impairment in mobility function; 9%), the former featured limited impairment across frailty domains (83%). Findings from LGM indicated that these profiles were differentially related to cognitive performance and decline. Data-driven approaches can help detect early differentiation of frailty profiles and contribute to personalized intervention.

Download Full-text