A Novel Virtual Sample Generation Method to Overcome the Small Sample Size Problem in Computer Aided Medical Diagnosing

Abstract

Deep neural networks are successful learning tools for building nonlinear models. However, a robust deep learning-based classification model needs a large dataset. Indeed, these models are often unstable when they use small datasets. To solve this issue, which is particularly critical in light of the possible clinical applications of these predictive models, researchers have developed approaches such as virtual sample generation. Virtual sample generation significantly improves learning and classification performance when working with small samples. The main objective of this study is to evaluate the ability of the proposed virtual sample generation to overcome the small sample size problem, which is a feature of the automated detection of a neurodevelopmental disorder, namely autism spectrum disorder. Results show that our method enhances diagnostic accuracy from 84%–95% using virtual samples generated on the basis of five actual clinical samples. The present findings show the feasibility of using the proposed technique to improve classification performance even in cases of clinical samples of limited size. Accounting for concerns in relation to small sample sizes, our technique represents a meaningful step forward in terms of pattern recognition methodology, particularly when it is applied to diagnostic classifications of neurodevelopmental disorders. Besides, the proposed technique has been tested with other available benchmark datasets. The experimental outcomes showed that the accuracy of the classification that used virtual samples was superior to the one that used original training data without virtual samples.

Augmentation in Healthcare: Augmented Biosignal Using Deep Learning and Tensor Representation

“…Most of the techniques that were used to create virtual samples suffer from the lack of combining reasonableness and adaptability simultaneously. Accordingly, we followed an algorithm to generate virtual Gaussian samples [ 2 ]. This method started to calculate the mean and standard deviation for Gaussian distribution, as shown in Figure 4 .…”

“…In healthcare systems, data are not publicly available, and these data are limited in nature too. For example, in the current pandemic, the COVID-19, no data are publicly available and some institutes have very limited data [ 1 , 2 ]. As a result, machine learning and big data analytics cannot be performed on such limited data.…”

“…One possible solution is to augment limited data and increase the data for testing and training of various machine learning algorithms. The main purpose of Data Augmentation (DA) is to increase the data size [ 2 ]. Also, DA is a technique that strongly invades the field of data mining and processing for regression and classification purposes, particularly in healthcare applications.…”

“…DA algorithms become numerous. The manipulation between DA algorithms is to achieve high accuracy results and at the same time, implementing modest and rapid algorithm is a typical matter of talent [ 2 ]. The appropriately selected DA technique drives the accuracy values to a dramatic level.…”

Evolutionary Algorithms in Health Technologies

“…Wedyan et al [1] built a virtual sample generation for deep neural network-based nonlinear modelling with small samples. The main objective of their study was to evaluate the ability of the proposed virtual sample generation to overcome the small sample size problem, which is a feature of the automated detection of a neurodevelopmental disorder, namely autism spectrum disorder.…”