1,720,966 research outputs found
Predicting and monitoring blood glucose through nutritional factors in type 1 diabetes by artificial neural networks
Full text linkThe monitoring and management of Postprandial Glucose Response (PGR), by administering an insulin bolus before meals, is a crucial issue in Type 1 Diabetes (T1D) patients. Artificial Pancreas (AP), which combines autonomous insulin delivery and blood glucose sensor, is a promising solution; nevertheless, it still requires input from patients about meal carbohydrate intake for bolus administration. This is due to the limited knowledge of the factors that influence PGR. Even though meal carbohydrates are regarded as the major factor influencing PGR, medical experience suggests that other nutritional should be considered. To address this issue, in this work, we propose a Machine Learning (ML)-based approach for a more comprehensive analysis of the impact of nutritional factors (i.e., carbohydrates, protein, lipids, fiber, and energy intake) on the blood glucose levels (BGLs). In particular, the proposed ML-model takes into account BGLs, insulin doses, and nutritional factors in T1D patients to predict BGLs in 60-minute time windows after a meal. A Feed-Forward Neural Network was fed with different combinations of BGLs, insulin, and nutritional factors, providing a predicted glycaemia curve as output. The validity of the proposed system was demonstrated through tests on public data and on self-produced data, adopting intra- and inter-subject approach. Results anticipate that patient-specific data about nutritional factors of a meal have a major role in the prediction of postprandial BGLs
Interpreting the latent space of a Convolutional Variational Autoencoder for semi-automated eye blink artefact detection in EEG signals
Full text linkElectroencephalography (EEG) allows the investigation of brain activity. However, neural signals often contain artefacts, hindering signal analysis. For example, eye-blink artefacts are particularly challenging due to their frequency overlap with neural signals. Artificial intelligence, particularly Variational Autoencoders (VAE), has shown promise in EEG artefact removal. This research explores the design and application of Convolutional VAEs for automatically detecting and removing eye blinks in EEG signals. The latent space of CVAE, trained on EEG topographic maps, is used to identify latent components that are selective for eye blinks. Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) are employed to evaluate the discriminative performance of each latent component. The most discriminative component, determined by the highest AUC, is modified to eliminate eye blinks. The evaluation of artefact removal involves visual inspection and Pearson correlation index assessment of the original EEG signal and the reconstructed clean version, focusing on the Fp1 and Fp2 channels most affected by eye-blink artefacts. Results indicate that the proposed method effectively removes eye blinks without significant loss of information related to the neural signal, demonstrating Pearson correlation values around 0.60 for each subject. The contribution to the knowledge offered by this research study is the design and application of a novel offline pipeline for automatically detecting and removing eye blinks from multi-variate EEG signals without human intervention
Exploring the Latent Space of Person-Specific Convolutional Autoencoders for Eye-Blink Artefact Mitigation in EEG Signals
No full textElectroencephalography (EEG) is a non-invasive and
cost-effective technique that allows the investigation of brain
activity. However, EEG recordings often suffer from artefacts
that complicate signal analysis. Eye-blink artefacts pose a sig
nificant challenge among these artefacts due to their frequency
overlap with neural signals. Machine Learning, notably semi
supervised Autoencoders (AEs), appears promising in removing
EEGartefacts. This research investigates the use of Convolutional
Autoencoders (CAEs) for mitigating eye blinks in EEG signals,
deviating from a previous use of Convolutional Variational AEs.
This shift can offer a simpler approach with reduced computa
tional complexity. Specifically, the latent space of CAEs, trained
on spatially preserving EEG topographic maps, was explored
to identify latent components selective for eye blinks. Receiver
Operating Characteristic (ROC) curves and Area Under the
Curve (AUC) were employed to evaluate each latent component’s
discriminative performance. The most discriminative component,
determined by the highest AUC, is subsequently modified to
mitigate eye blinks. Specifically, the median is chosen to mask
this discriminative latent component for blink artefact removal.
Visual inspections and Pearson correlation indices between the
original EEG signal and the reconstructed clean version were
used to evaluate the effectiveness of artefact removal. This study
contributes to the knowledge for introducing an offline pipeline
able to detect and remove eye blinks from EEG recordings
without human intervention
Entropy-Based EEG Measures for Revealing Altered Neural Dynamics in Alzheimer's Disease: A Preliminary Study
No full textAlzheimer's disease (AD) is a prevalent age-related neurodegenerative condition, whose timely diagnosis is nowadays still complicated. In this context, investigating brain complexity can provide insight into the underlying mechanisms of neural dynamics in AD patients. In this regard, this study explores the differences in neural complexity between AD patients and healthy controls by using entropy measures. Specifically, multi-scale sample entropy (MSE) and multiscale approximate entropy (MAE) was computed on multi-channel electroencephalografic (EEG) signals at different time scales, highlighting the advantages and disadvantages of these two techniques, and outlining future improvements. Results show the potential of entropy measures as a promising strategy for early diagnosis of AD and for a better understanding of EEG dynamics alteration in AD, allowing for the development of more targeted therapies
Explainable AI Assessment of Meal-Related Features Impact in Predicting Basal Insulin for Type I Diabetes
No full textType 1 diabetes management using artificial pan-creas (AP) technology faces challenges in accurately regulating postprandial glucose levels. Current AP systems, based on heuris-tic control algorithms for basal insulin delivery, still require user input for mealtime insulin delivery. In the context of personalized patient-centered medicine, this study explores the impact of meal-related features on predicting postprandial basal insulin requirements using explainable artificial intelligence (XAI). Machine learning (ML) models were trained using preprandial blood glucose, mealtime insulin, preprandial basal insulin, and features from 15 T1D patients. The performance in terms of the root-mean-square-error (RMSE) and the mean-absolute-error (MSE) were (0.42 ± 0.1) insulin unit and (0.32 ± 0.07) insulin unit, respectively. Employing SHapley Additive exPlanations (SHAP) reveals significant influences of mealtime insulin bolus, carbohy-drates, and glycemic load, aligning with physiological knowledge on postprandial glycemia response (PGR)
Exploring Nutritional Influence on Blood Glucose Forecasting for Type 1 Diabetes Using Explainable AI
Full text link: Type 1 diabetes mellitus (T1DM) is characterized by insulin deficiency and blood sugar control issues. The state-of-the-art solution is the artificial pancreas (AP), which integrates basal insulin delivery and glucose monitoring. However, APs are unable to manage postprandial glucose response (PGR) due to limited knowledge of its determinants, requiring additional information for accurate bolus delivery, such as estimated carbohydrate intake. This study aims to quantify the influence of various meal-related factors on predicting postprandial blood glucose levels (BGLs) at different time intervals (15 min, 60 min, and 120 min) after meals by using deep neural network (DNN) models. The prediction models incorporate preprandial blood glucose values, insulin dosage, and various meal-related nutritional factors such as intake of energy, carbohydrates, proteins, lipids, fatty acids, fibers, glycemic index, and glycemic load as input variables. The impact of input features was assessed by exploiting eXplainable Artificial Intelligence (XAI) methodologies, specifically SHapley Additive exPlanations (SHAP), which provide insights into each feature's contribution to the model predictions. By leveraging XAI methodologies, this study aims to enhance the interpretability and transparency of BGL prediction models and validate clinical literature hypotheses. The findings can aid in the development of decision-support tools for individuals with T1DM, facilitating PGR management and reducing the risks of adverse events. The improved understanding of PGR determinants may lead to advancements in AP technology and improve the overall quality of life for T1DM patients
A Wearable AR-based BCI for Robot Control in ADHD Treatment: Preliminary Evaluation of Adherence to Therapy
Full text linkA wearable, single-channel Brain-Computer Interface (BCI) based on Augmented Reality (AR) and Steady-State Visually Evoked Potentials (SSVEPs) for robot control is
proposed as an innovative therapy for robot-based Attention Deficit Hyperactivity Disorder (ADHD) rehabilitation of children. The system manages to overcome the challenges
regarding immersivity and wearability, providing a direct path between human brain and social robots, already successfully employed for ADHD treatment. Through the proposed system, even without training, the user can drive a robot, in real-time, by brain signals. A preliminary evaluation of the children adherence to the therapy was conducted as a case study on 18 subjects, at an accredited rehabilitation center. After investigating the children acceptance of the proposed system, different tasks were assigned to the volunteers aiming to observe their level of involvement. The experimental activity showed encouraging results, where almost all the participants were satisfied with the
experience and keen to repeat it again in the future
Comparison of EEG Preprocessing Techniques for Complexity Measures in Alzheimer's Disease Detection
No full textAlzheimer's disease (AD) is a neurodegenerative condition characterized by memory loss, cognitive difficulties, and behavioral changes. Currently, there are no known permanent cure for this condition. Early and precise diagnosis is crucial for effective therapeutic intervention. This study introduces an unsupervised machine learning (ML)-based method for identifying patterns in electroencephalographic (EEG) signals from both healthy subjects and AD patients. Emphasis was placed on the preprocessing phase, examining the effects of various EEG data normalization techniques on the results. This generalized approach addresses the significant inter- and intra-subjective variability inherent in biological data, thereby enhancing method robustness and data consistency. Following the preprocessing phase, Multiscale Fuzzy Entropy (MFE) was derived from EEG signals. A k-means clustering algorithm was applied to identify distinct patterns. The efficacy of the clusters was assessed by using Silhouette Score (SS), Adjusted Rand Index Score (ARI), Adjusted Mutual Information Score (AMI) and V-Measure Score. The method was validated using a public EEG dataset. The results indicated enhanced clustering efficacy when normalization was applied, underscoring the critical importance of data preprocessing in the detection of AD
A Novel Metric for Alzheimer’s Disease Detection Based on Brain Complexity Analysis via Multiscale Fuzzy Entropy
Full text linkAlzheimer’s disease (AD) is a neurodegenerative brain disorder that affects cognitive functioning and memory. Current diagnostic tools, including neuroimaging techniques and cognitive questionnaires, present limitations such as invasiveness, high costs, and subjectivity. In recent years, interest has grown in using electroencephalography (EEG) for AD detection due to its non-invasiveness, low cost, and high temporal resolution. In this regard, this work introduces a novel metric for AD detection by using multiscale fuzzy entropy (MFE) to assess brain complexity, offering clinicians an objective, cost-effective diagnostic tool to aid early intervention and patient care. To this purpose, brain entropy patterns in different frequency bands for 35 healthy subjects (HS) and 35 AD patients were investigated. Then, based on the resulting MFE values, a specific detection algorithm, able to assess brain complexity abnormalities that are typical of AD, was developed and further validated on 24 EEG test recordings. This MFE-based method achieved an accuracy of 83% in differentiating between HS and AD, with a diagnostic odds ratio of 25, and a Matthews correlation coefficient of 0.67, indicating its viability for AD diagnosis. Furthermore, the algorithm showed potential for identifying anomalies in brain complexity when tested on a subject with mild cognitive impairment (MCI), warranting further investigation in future research
Assessment and Scientific Progresses in the Analysis of Olfactory Evoked Potentials
Full text linkThe human sense of smell is important for many vital functions, but with the current state of the art, there is a lack of objective and non-invasive methods for smell disorder diagnostics. In recent years, increasing attention is being paid to olfactory event-related potentials (OERPs) of the brain, as a viable tool for the objective assessment of olfactory dysfunctions. The aim of this review is to describe the main features of OERPs signals, the most widely used recording and processing techniques, and the scientific progress and relevance in the use of OERPs in many important application fields. In particular, the innovative role of OERPs is exploited in olfactory disorders that can influence emotions and personality or can be potential indicators of the onset or progression of neurological disorders. For all these reasons, this review presents and analyzes the latest scientific results and future challenges in the use of OERPs signals as an attractive solution for the objective monitoring technique of olfactory disorders
- …
