Potential role of artificial intelligence in craniofacial surgery

The field of artificial intelligence (AI) is rapidly advancing, and AI models are increasingly applied in the medical field, especially in medical imaging, pathology, natural language processing, and biosignal analysis. On the basis of these advances, telemedicine, which allows people to receive medical services outside of hospitals or clinics, is also developing in many countries. The mechanisms of deep learning used in medical AI include convolutional neural networks, residual neural networks, and generative adversarial networks. Herein, we investigate the possibility of using these AI methods in the field of craniofacial surgery, with potential applications including craniofacial trauma, congenital anomalies, and cosmetic surgery.

Reservoir computing with biocompatible organic electrochemical networks for brain-inspired biosignal classification

Early detection of malign patterns in patients’ biological signals can save millions of lives. Despite the steady improvement of artificial intelligence–based techniques, the practical clinical application of these methods is mostly constrained to an offline evaluation of the patients’ data. Previous studies have identified organic electrochemical devices as ideal candidates for biosignal monitoring. However, their use for pattern recognition in real time was never demonstrated. Here, we produce and characterize brain-inspired networks composed of organic electrochemical transistors and use them for time-series predictions and classification tasks using the reservoir computing approach. To show their potential use for biofluid monitoring and biosignal analysis, we classify four classes of arrhythmic heartbeats with an accuracy of 88%. The results of this study introduce a previously unexplored paradigm for biocompatible computational platforms and may enable development of ultralow–power consumption hardware-based artificial neural networks capable of interacting with body fluids and biological tissues.

Intelligent Biosignal Analysis Methods

This Editorial presents the accepted manuscripts for the special issue “Intelligent Biosignal Analysis Methods” of the Sensors MDPI journal [...]

Overview of Biosignal Analysis Methods for the Assessment of Stress

Objectives: Stress is a normal reaction of the human organism induced in situations that demand a level of activation. This reaction has both positive and negative impact on the life of each individual. Thus, the problem of stress management is vital for the maintenance of a person’s psychological balance. This paper aims at the brief presentation of stress definition and various factors that can lead to augmented stress levels. Moreover, a brief synopsis of biosignals that are used for the detection and categorization of stress and their analysis is presented. Methods: Several studies, articles and reviews were included after literature research. The main questions of the research were: the most important and widely used physiological signals for stress detection/assessment, the analysis methods for their manipulation and the implementation of signal analysis for stress detection/assessment in various developed systems. Findings: The main conclusion is that current researching approaches lead to more sophisticated methods of analysis and more accurate systems of stress detection and assessment. However, the lack of a concrete framework towards stress detection and assessment remains a great challenge for the research community. Doi: 10.28991/esj-2021-01267 Full Text: PDF

Multiparameter Biosignal Analysis in Elderly Ischemic Stroke Patients

Biosignal can provide information about body conditions, including physiological conditions of ischemic stroke. The regulation of blood in the brain is regulated through the mechanism of Cerebral Autoregulation (CA). Some parameters that can be used to determine this mechanism are Blood Flow Velocity (BFV) and Blood Pressure (BP). Stroke is also related to nervous system activity, which is represented through the Heart Rate Variability (HRV). This study aims to determine the relationship between those biosignals and their effects on the physiology of ischemic stroke sufferers. The subjects were divided into two groups (20 strokes and 20 controls). BFV data is obtained in the Middle Cerebral Artery (MCA), BP is obtained through the arteries of the upper arms, and 3 leads electrocardiogram is placed in the chest. The results showed that there was a relationship between BP and BFV in the control group (p-value < 0.05; r = -0.574). This correlation was not found in the stroke group. The relationship between BP and HRV was only found in the stroke group, which was associated with high sympathetic activity and lower parasympathetic activity (p-values < 0.05 and r > 0.4). It was based on SDRR, RMSSD, CVRR, LF, and SD1 parameters. In the control group, there was no relationship between HRV and BP. The relationship between BFV and HRV in the control group was not found statistically. Still, in the stroke group, this relationship was found in the LF and LF/HF Ratio parameters (p-value < 0.05; r > 0.4). Based on this research, parameters on HRV that can be used to determine the characteristics of stroke patients in all positions are MeanRR, VLF, and LF

Fronto-Temporal Analysis of EEG Signals of Patients with Depression: Characterisation, Nonlinear Dynamics and Surrogate Analysis

The recent advances in signal processing techniques have enabled the analysis of biosignals from brain so as to enhance the predictive capability of mental states. Biosignal analysis has been successfully used to characterise EEG signals of unipolar depression patients. Methods of characterisation of EEG signals and the use of nonlinear parameters are the major highlights of this chapter. Bipolar frontopolar-temporal EEG recordings obtained under eyes open and eyes closed conditions are used for the analysis. A discussion on the reliability of the use of energy distribution and Relative Wavelet Energy calculations for distinguishing unipolar depression patients from healthy controls is presented. The potential of the application of Wavelet Entropy to differentiate states of the brain under normal and pathologic condition is introduced. Details are given on the suitability of ascertaining certain nonlinear indices on the feature extraction, assuming the time series to be highly nonlinear. The assumption of nonlinearity of the measured EEG time series is further verified using surrogate analysis. The studies discussed in this chapter indicate lower values of nonlinear measures for patients. The higher values of signal energy associated with the delta bands of depression patients in the lower frequency range are regarded as a major characteristic indicative of a state of depression. The chapter concludes by presenting the important results in this direction that may lead to better insight on the brain activity and cognitive processes. These measures are hence posited to be potential biomarkers for the detection of depression.