High-performance detection of alcoholism by unfolding the amalgamated EEG spectra using the Random Forests method

Abstract

We show that by unfolding the outdated EEG standard bandwidths in a fine-grade equidistant 99-point spectrum we can precisely detect alcoholism. Using this novel pre-processing step prior to entering a random forests classifier, our method substantially outperforms all previous results with a balanced accuracy of 97.4 percent. Our machine learning work contributes to healthcare and information systems. Due to its drastic and protracted consequences, alcohol consumption is always a critical issue in our society. Consequences of alcoholism in the brain can be recorded using electroencephalography (EEG). Our work can be used to automatically detect alcoholism in EEG mass data within milliseconds. In addition, our results challenge the medically outdated EEG standard bandwidths.

Development of a Machine Learning Based Algorithm To Accurately Detect Schizophrenia based on One-minute EEG Recordings

“…This insight was supported by Howells et al [43]. For this reason we first decided to apply the unmodified method from the study of Rieg et al [34] considering the frequency range up to 50 Hz. Thus, we achieved a balanced accuracy of 96.01 %.…”

“…While we intensively evaluated other traditional machine learning approaches such as clustering [46] and also most modern convolutional neural networks, which are outstanding in other domains such as image recognition [47][48][49], we achieved the best results here with our novel tree-based method proposed in [34]. However, the method of choice always limits scientific understanding.…”

“…In this work, the commonly used division of frequency bands into alpha, beta, theta, delta and gamma bands, as shown in table 1, was not used as a feature extraction criterion [31]. According to the study of Rieg et al [34], we decided to apply their method on EEG data of schizophrenic and non-schizophrenic persons. To take the upper gamma frequency bands into account, we additionally decided to extend the frequency range up to 100 Hz.…”

“…This classifier shows substantial improvements in classification accuracy that results from growing collections of trees and letting them vote for the most popular class. The process works as illustrated in figure 3, and is described as follows [34][35][36][37]: 1) ntree bootstrap training samples are randomly produced from the original data. 2) Each training sample generates the corresponding decision tree.…”

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

“…Machine learning (ML) is one of the possibilities for use in evaluating large amounts of data in real time [17][18][19][20][21][22][23][24][25][26]. Over the last decades with more powerful hardware, more and better ML architectures were developed and refined.…”