Jovana Paunovic Pantic , Svetlana Valjarevic , Jelena Cumic , Igor Pantic
{"title":"AI-enhanced EEG signal interpretation: A novel approach using texture analysis with random forests","authors":"Jovana Paunovic Pantic , Svetlana Valjarevic , Jelena Cumic , Igor Pantic","doi":"10.1016/j.mehy.2024.111405","DOIUrl":null,"url":null,"abstract":"<div><p>We hypothesize that the Gray-Level Co-occurrence Matrix (GLCM) and the Run-Length Matrix (RLM) techniques can effectively quantify discrete changes in EEG signals, and that the features extracted from these matrices can be utilized to train a Random Forest (RF) model. Our contribution includes the development of a robust code in sci-kit learn for a hypothetical model that, after adequate training and testing, could be used to detect and remove artifacts as well as differentiate between physiological and pathological EEG signals. Moreover, our approach envisions the RF model as a powerful tool capable of differentiating between normal and abnormal EEG signals. This approach could lead to the development of more potent AI tools that enhance clinical decision-making in neurology and psychiatry.</p></div>","PeriodicalId":18425,"journal":{"name":"Medical hypotheses","volume":"189 ","pages":"Article 111405"},"PeriodicalIF":2.1000,"publicationDate":"2024-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical hypotheses","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0306987724001488","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MEDICINE, RESEARCH & EXPERIMENTAL","Score":null,"Total":0}
引用次数: 0
Abstract
We hypothesize that the Gray-Level Co-occurrence Matrix (GLCM) and the Run-Length Matrix (RLM) techniques can effectively quantify discrete changes in EEG signals, and that the features extracted from these matrices can be utilized to train a Random Forest (RF) model. Our contribution includes the development of a robust code in sci-kit learn for a hypothetical model that, after adequate training and testing, could be used to detect and remove artifacts as well as differentiate between physiological and pathological EEG signals. Moreover, our approach envisions the RF model as a powerful tool capable of differentiating between normal and abnormal EEG signals. This approach could lead to the development of more potent AI tools that enhance clinical decision-making in neurology and psychiatry.
期刊介绍:
Medical Hypotheses is a forum for ideas in medicine and related biomedical sciences. It will publish interesting and important theoretical papers that foster the diversity and debate upon which the scientific process thrives. The Aims and Scope of Medical Hypotheses are no different now from what was proposed by the founder of the journal, the late Dr David Horrobin. In his introduction to the first issue of the Journal, he asks ''what sorts of papers will be published in Medical Hypotheses? and goes on to answer ''Medical Hypotheses will publish papers which describe theories, ideas which have a great deal of observational support and some hypotheses where experimental support is yet fragmentary''. (Horrobin DF, 1975 Ideas in Biomedical Science: Reasons for the foundation of Medical Hypotheses. Medical Hypotheses Volume 1, Issue 1, January-February 1975, Pages 1-2.). Medical Hypotheses was therefore launched, and still exists today, to give novel, radical new ideas and speculations in medicine open-minded consideration, opening the field to radical hypotheses which would be rejected by most conventional journals. Papers in Medical Hypotheses take a standard scientific form in terms of style, structure and referencing. The journal therefore constitutes a bridge between cutting-edge theory and the mainstream of medical and scientific communication, which ideas must eventually enter if they are to be critiqued and tested against observations.