Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection

Brain Informatics, Journal Year: 2024, Volume and Issue: 11(1)

Published: April 5, 2024

Abstract Explainable artificial intelligence (XAI) has gained much interest in recent years for its ability to explain the complex decision-making process of machine learning (ML) and deep (DL) models. The Local Interpretable Model-agnostic Explanations (LIME) Shaply Additive exPlanation (SHAP) frameworks have grown as popular interpretive tools ML DL This article provides a systematic review application LIME SHAP interpreting detection Alzheimer’s disease (AD). Adhering PRISMA Kitchenham’s guidelines, we identified 23 relevant articles investigated these frameworks’ prospective capabilities, benefits, challenges depth. results emphasise XAI’s crucial role strengthening trustworthiness AI-based AD predictions. aims provide fundamental capabilities XAI enhancing fidelity within clinical decision support systems prognosis.

Language: Английский

---

Integration of wearable electronics and heart rate variability for human physical and mental well-being assessment

Journal of Semiconductors, Journal Year: 2025, Volume and Issue: 46(1), P. 011603 - 011603

Published: Jan. 1, 2025

Abstract Heart rate variability (HRV) that can reflect the dynamic balance between sympathetic nervous and parasympathetic of human autonomic system (ANS) has attracted considerable attention. However, traditional electrocardiogram (ECG) devices for HRV analysis are bulky, hard wires needed to attach measuring electrodes chest, resulting in poor wearable experience during long-term measurement. Compared with that, electronics enabling continuously cardiac signals monitoring assessment provide a desirable promising approach helping subjects determine sleeping issues, cardiovascular diseases, or other threats physical mental well-being. Until now, significant progress advances have been achieved applications predicting In this review, latest integration as well practical health included. The commonly used methods physiological briefly summarized. Furthermore, we highlighted research on concerning diverse such stress estimation, drowsiness detection, etc. Lastly, current limitations integrated concluded, possible solutions direction outlined.

Language: Английский

---

Assessing the clinical reliability of short-term heart rate variability: insights from controlled dual-environment and dual-position measurements

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 15, 2025

Heart rate variability (HRV) is a widely recognized biomarker for autonomic nervous system regulation, applicable in clinical and athletic settings to monitor health recovery. Despite its extensive use, HRV measurement reliability influenced by numerous factors, necessitating controlled conditions accurate assessments. This study investigates the of short-term measurements various positions, aiming establish consistent protocols monitoring interpretation. We assessed morning 34 healthy, physically active adults across supine standing at home laboratory, over 24-hour period. Environment significantly impacted HRV. Home exhibited slightly lower variance compared lab settings, underscoring importance environment control. Our findings confirm high measurements, indicating their robustness capturing changes, provided rigorous methodology employed. Here we show that effective reliable assessment possible conditions, contingent upon strict management confounding factors. research supports utility as non-invasive diagnostic tool, emphasizing potential broadening applications diverse populations. Future studies are encouraged expand these assessments include varied demographic profiles, enhancing integration into routine evaluations.

Language: Английский

---

Impact of Built Environments on Human Perception: A Systematic Review of Physiological Measures and Machine Learning

Journal of Building Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 112319 - 112319

Published: March 1, 2025

Language: Английский

---

EEG Data Analysis Techniques for Precision Removal and Enhanced Alzheimer’s Diagnosis: Focusing on Fuzzy and Intuitionistic Fuzzy Logic Techniques

Signals, Journal Year: 2024, Volume and Issue: 5(2), P. 343 - 381

Published: May 31, 2024

Effective management of EEG artifacts is pivotal for accurate neurological diagnostics, particularly in detecting early stages Alzheimer’s disease. This review delves into the cutting-edge domain fuzzy logic techniques, emphasizing intuitionistic systems, which offer refined handling uncertainties inherent data. These methods not only enhance artifact identification and removal but also integrate seamlessly with other AI technologies to push boundaries analysis. By exploring a range approaches from standard protocols advanced machine learning models, this paper provides comprehensive overview current strategies emerging management. Notably, fusion neural network models illustrates significant advancements distinguishing between genuine activity noise. synthesis improves diagnostic accuracy enriches toolset available researchers clinicians alike, facilitating earlier more precise neurodegenerative diseases. The ultimately underscores transformative potential integrating diverse computational setting new analysis paving way future innovations medical diagnostics.

Language: Английский

---

HyEpiSeiD: a hybrid convolutional neural network and gated recurrent unit model for epileptic seizure detection from electroencephalogram signals

Brain Informatics, Journal Year: 2024, Volume and Issue: 11(1)

Published: Aug. 21, 2024

Abstract Epileptic seizure (ES) detection is an active research area, that aims at patient-specific ES with high accuracy from electroencephalogram (EEG) signals. The early of crucial for timely medical intervention and prevention further injuries the patients. This work proposes a robust deep learning framework called HyEpiSeiD extracts self-trained features pre-processed EEG signals using hybrid combination convolutional neural network followed by two gated recurrent unit layers performs prediction based on those extracted features. proposed evaluated public datasets, UCI Epilepsy Mendeley datasets. model achieved 99.01% 97.50% classification accuracy, respectively, outperforming most state-of-the-art methods in epilepsy domain.

Language: Английский

---

Automated classification of stress and relaxation responses in major depressive disorder, panic disorder, and healthy participants via heart rate variability

Frontiers in Psychiatry, Journal Year: 2025, Volume and Issue: 15

Published: Jan. 9, 2025

Stress is a significant risk factor for psychiatric disorders such as major depressive disorder (MDD) and panic (PD). This highlights the need advanced stress-monitoring technologies to improve treatment. affects autonomic nervous system, which can be evaluated via heart rate variability (HRV). While machine learning has enabled automated stress detection HRV in healthy individuals, its application patients remains underexplored. study feasibility of using machine-learning algorithms detect automatically MDD PD patients, well controls (HCs), based on features. The included 147 participants (MDD: 41, PD: 47, HC: 59) who visited laboratory up five times over 12 weeks. data were collected during relaxation tasks, with 20 features extracted. Random forest multilayer perceptron classifiers applied distinguish between tasks. Feature importance was analyzed SHapley Additive exPlanations, differences tasks (ΔHRV) compared across groups. impact personalized longitudinal scaling classification accuracy also assessed. accuracies 0.67 MDD, 0.69 PD, 0.73 HCs, indicating higher HC group. Longitudinal improved 0.94 0.90 0.96 suggesting potential monitoring patients' conditions HRV. group demonstrated greater ΔHRV fluctuation larger number more than patient groups, potentially contributing accuracy. Multilayer models provided consistent results random forest, confirming robustness findings. that differentiating challenging groups group, underscoring metrics biomarkers. Psychiatric exhibited altered responses, may influence their reactivity. indicates tailored approach these Additionally, we emphasized significance enhancing accuracy, utilized develop patients.

Language: Английский

---

Machine Learning-Based Human Stress Detection Model Employing Physiological Sensory Data

Arabian Journal for Science and Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 21, 2025

Language: Английский

---

Detection of Cognitive Performance Deterioration Due to Cold-Air Exposure in Females Using Wearable Electrodermal Activity and Electrocardiogram

Biosensors, Journal Year: 2025, Volume and Issue: 15(2), P. 78 - 78

Published: Jan. 29, 2025

Prolonged exposure to cold air can impair reaction time and cognitive function, which lead serious consequences. One mitigation strategy is develop models that predict performance by tracking physiological metrics associated with stress. As females are evidenced be more sensitive exposure, this study investigated the relationship between deterioration of female subjects under Wearable electrodermal activity (EDA) electrocardiogram (ECG) were collected from nineteen who underwent five sessions a task battery—assessing time, memory, attention—in (10 °C) environment. Machine learning classifiers showed higher classification accuracies heart rate variability (HRV) features than EDA features. Particularly in detecting assessing short-term our support vector machine classifier HRV an 82.4% accuracy, sensitivity 84.2% specificity 80.6%, whereas 55.4% accuracy 44.7% 66.7% was obtained Our results demonstrate feasibility using wearable ECG, allowing for preventive measures reduce risk environments, especially military personnel.

Language: Английский

---

Patient performance assessment methods for upper extremity rehabilitation in assist-as-needed therapy strategies: a comprehensive review

Medical & Biological Engineering & Computing, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 7, 2025

This paper aims to comprehensively review patient performance assessment (PPA) methods used in assist-as-needed (AAN) robotic therapy for upper extremity rehabilitation. AAN strategies adjust assistance according the patient's performance, aiming enhance engagement and recovery individuals with motor impairments. categorizes implemented PPA literature first time such a wide scope suggests future research directions improve adaptive personalized therapy. At first, studies are examined evaluate methods, which subsequently categorized their underlying implementation strategies: position error-based force-based electromyography (EMG), electroencephalography (EEG)-based indicator-based physiological signal-based methods. The advantages limitations of each method discussed. In addition classification current study also examines clinically tested applied rehabilitation clinical outcomes. Clinical findings from these trials demonstrate potential improving function engagement. Nevertheless, more extensive testing is necessary establish long-term benefits over conventional therapies. Ultimately, this guide developments field rehabilitation, providing researchers insights into optimizing enhanced

Language: Английский

---