A biological continuum based approach for efficient clinical classification

Abstract

Clinical feature selection problem is the task of selecting and identifying a subset of informative clinical features that are useful for promoting accurate clinical diagnosis. This is a significant task of pragmatic value in the clinical settings as each clinical test is associated with a different financial cost, diagnostic value, and risk for obtaining the measurement. Moreover, with continual introduction of new clinical features, the need to repeat the feature selection task can be very time consuming. Therefore to address this issue, we propose a novel feature selection technique for diagnosis of myocardial infarction - one of the leading causes of morbidity and mortality in many high-income countries. This method adopts the conceptual framework of biological continuum, the optimization capability of genetic algorithm for performing feature selection and the classification ability of support vector machine. Together, a network of clinical risk factors, called the biological continuum based etiological network (BCEN), was constructed. Evaluation of the proposed methods was carried out using the cardiovascular heart study (CHS) dataset. Results demonstrate a significant speedup of 4.73-fold can be achieved for the development of MI classification model. The key advantage of this methodology is the provision of a reusable (feature subset) paradigm for efficient development of up-to-date and efficacious clinical classification models.

A novel neural-inspired learning algorithm with application to clinical risk prediction

“…radial basis function). UD approach was adopted for SVM as it has been shown to produce promising results, and at the same time alleviate the computational loads associated with the search for the optimal cost-gamma pair [36]. The parameter details for SVM are kernel function: radial basis function (RBF); cost: [2 À5 , 2 13 ]; gamma: [2 À15 , 2 3 ]; for EDC-AIRS are seed: 1; clonal rate: 10; hyper-mutation rate: 2; stimulation threshold: 0.9; initial memory pool size: The feature selection step aims to identify informative clinical markers that would contribute to the development of an accurate and parsimonious prediction model.…”

Assessing the performance of four different categories of histological criteria in brain tumours grading by means of a computer‐aided diagnosis image analysis system

“…Thus, treatment planning becomes subjective (Newman‐Toker et al ., ). (iii) It is well known that cancer develops along a biological continuum making it hard to discriminate between different stages, histological tumour grades and other diagnostic characteristics of the disease (Tay et al ., ). There are no systematic ways to identify this continuum when presenting to the pathologist only few slices from the total tumour area.…”

Detection of Cardiovascular Disease from Clinical Parameters Using a One-Dimensional Convolutional Neural Network

“…Additionally, manual feature extraction has similar problems as in group-(a); these studies must follow standardized guidelines to compare performance consistently. Due to the availability of several datasets with clinical parameters, the studies in group-(c) are prevalent among scientists [38][39][40][41][42][43][44][45][46][47][48][49][50]. Here, a specific list of clinical parameters is used as input for the system; the number and types of parameters depend on which dataset is selected.…”