Search

Long-term diseases often develop at a slow rate, making it difficult to detect them. These medical conditions often arise due to approached health behavior patterns. The circumstance of negative behavioral patterns encouraging diseases results in treatment issues due to late treatment initialization. This paper presents a machine learning-based approach to predict people’s health behavior tendencies at an early stage based on their Future Time Perspective using objective resting-state EEG data. With a balanced accuracy of 95.00 percent based on the EEG frequency bands identified as relevant (3.5 Hz-4.5 Hz, 4.5 Hz-5.5 Hz, 20 Hz-21 Hz, and 27 Hz-28 Hz), our model sets a first benchmark in this field, building a base for early recognition and intervention of potential long-term diseases. [ABSTRACT FROM AUTHOR]

The human working memory and its cognitive functionality are essential for a range of fundamental to complex processes throughout our entire life. Damage and abnormalities can cause severe effects on an individual's life. Predictive healthcare analytics can support and accelerate the diagnosis of such effects in the early stages. Using restingstate EEG data and the results of a cognitive testbattery for attentional performance, we developed a machine learning approach to evaluate the human working memory and detect hyperor hypoactivation. By predicting the testbattery results using the EEG recording of a patient, we enable a fast, objective, and accurate evaluation. Furthermore, we identified the most relevant brain regions (prefrontal cortex and dorsolateral prefrontal cortex) and the corresponding frequency subbands (9.511.5 Hz and 1113 Hz). With a balanced accuracy of 87.50%, our results set a new benchmark in evaluating the working memory using only restingstate EEG recordings. [ABSTRACT FROM AUTHOR]

We applied machine learning to diagnose epilepsy based on the fine-graded spectral analysis of seizure-free (resting state) EEG recordings. Despite using unspecific agglomerated EEG spectra, our fine-graded spectral analysis specifically identified the two EEG resting state sub-bands differentiating healthy people from epileptics (1.5-2 Hz and 11-12.5 Hz). The rigorous evaluation of completely unseen data of 100 EEG recordings (50 belonging to epileptics and the other 50 to healthy people) shows that the approach works successfully, achieving an outstanding accuracy of 99 percent, which significantly outperforms the current benchmark of 70% to 95% by a panel of up to three experienced neurologists. Our epilepsy diagnosis classifier can be implemented in modern EEG analysis devices, especially in intensive care units where early diagnosis and appropriate treatment are decisive in life and death scenarios and where physicians' error rates are particularly high. Our approach is accurate, robust, fast, and cost-efficient and substantially contributes to Information Systems research in healthcare. The approach is also of high practical and theoretical relevance. [ABSTRACT FROM AUTHOR]