Internet of Things for Diagnosis of Alzheimer’s Disease: A Multimodal Machine Learning Approach Based on Eye Movement Features

Yin, Yunpeng; Wang, Han; Liu, Shuai; Sun, Jinglin; Jing, Peiguang; Liu, Yu

doi:10.1109/jiot.2023.3245067

Cited by 9 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, the technique employed can distinguish older adult patients with Alzheimer's disease, achieving an accuracy of 86%, a true positive rate of 78%, and a predictive value of 90%. The results confirmed that the algorithm implemented in the diagnosis of Alzheimer's disease has demonstrated high feasibility in solving the problem of eye tracking designed by IoT for the detection of Alzheimer's disease [17], [18].…”

Section: Literature Reviewsupporting

confidence: 68%

“…Similarly, there is alignment with a study that implemented a computer system to determine the diagnosis of mild cognitive impairment using deep learning techniques, successfully having the model correctly anticipate mild impairment and distinguishing it from our approach based on other data mining techniques [16]. On the other hand, in research that implemented an IoT-based model utilizing the cloud and deep learning techniques with eye tracking, there is not a meticulous match with our model, as it does not emphasize the Alzheimer's prediction goal [17], [18].…”

Section: Discussionmentioning

confidence: 97%

See 1 more Smart Citation

Predictive models in Alzheimer's disease: an evaluation based on data mining techniques

Andrade-Arenas,

Rubio-Paucar,

Yactayo-Arias

2024

IJECE

View full text Add to dashboard Cite

The increasing prevalence of Alzheimer's disease in older adults has raised significant concern in recent years. Aware of this challenge, this research set out to develop predictive models that allow early identification of people at risk for Alzheimer's disease, considering several variables associated with the disease. To achieve this objective, data mining techniques were employed, specifically the decision tree algorithm, using the RapidMiner Studio tool. The sample explore modify model and assess (SEMMA) methodology was implemented systematically at each stage of model development, ensuring an orderly and structured approach. The results obtained revealed that 45.00% of people with dementia present characteristics that identify them as candidates for confirmation of a diagnosis of Alzheimer's disease. In contrast, 52.78% of those who do not have dementia show no danger of contracting the disease. In the conclusion of the research, it was noted that most patients diagnosed with Alzheimer's are older than 65 years, indicating that this stage of life tends to trigger brain changes associated with the disease. This finding underscores the importance of considering age as a key factor in the early identification of the disease.

show abstract

Section: Literature Reviewsupporting

confidence: 68%

Section: Discussionmentioning

confidence: 97%

Predictive models in Alzheimer's disease: an evaluation based on data mining techniques

Andrade-Arenas,

Rubio-Paucar,

Yactayo-Arias

2024

IJECE

View full text Add to dashboard Cite

show abstract

“…DL in biomedical signal processing is not limited to diagnostic applications and can also be expanded to personalized medicine, monitoring in real-time, and therapeutic interventions. As an example, a recently reported study demonstrates the capability of DL models to analyze timeevolving data streams from wearable sensors to monitor disease progressions and recognize critical events in a patient with chronic illness such as heart failure or epilepsy [32]. In addition, DL-based predictive models can assist clinicians with more accurately identifying high-risk patient sub-cohorts that are susceptible to specific complications or adverse effects and subsequently administer timely and accurate preventive measures [33].…”

Section: Backgroundsmentioning

confidence: 99%

Exploring Cutting-Edge Developments in Deep Learning for Biomedical Signal Processing

Zhu,

Zhang,

Liu

et al. 2024

ijacsa

View full text Add to dashboard Cite

Biomedical condition monitoring devices are progressing quickly by incorporating cost-effective and noninvasive sensors to track vital signs, record medical circumstances, and deliver meaningful responses. These sophisticated innovations rely on breakthrough technology to provide intelligent platforms for health monitoring, quick illness recognition, and precise treatment. Biomedical signal processing determines patterns of signals and serves as the backbone for reliable applications, medical diagnostics, and research. Deep Learning (DL) methods have brought significant innovation in biomedical signal processing, leading to the transformation of the health sector and medical diagnostics. This article covers an entire range of technical innovations evolved for DL-based biomedical signal processing where different modalities have been considered, including Electrocardiography (ECG), Electromyography (EMG), and Electroencephalography (EEG). A vast amount of biomedical data in various forms is available, and DL concepts are required to extract and model this data in order to identify hidden complex patterns that can be utilized to improve the diagnosis, prognosis, and personalized treatment of diseases in an individual. The nature of this developing topic certainly gives rise to a number of challenges. First, the application of sensitive and noisy time series data requires truly robust models. Second, many inferences made at the bedside must have interpretability by design. Third, the field will require that processing be performed in real-time if used for therapeutic interventions. We systematically evaluate these challenges and highlight areas where continued research is needed. The general expansion of DL technologies into the biomedical domain gives rise to novel concerns about accountability and transparency of algorithmic decision-making, a subject which we briefly touch upon as well.

show abstract

“…Lu et al [22] used an fNIRS-based method for functional neurodegeneration analysis in Parkinson's Disease, demonstrating the utility of functional near-infrared spectroscopy in medical research. Yin et al [23] explored the Internet of Things (IoT) for Alzheimer's Disease diagnosis using eye movement features, highlighting the integration of IoT in medical diagnostics. Mishra et al [24] reviewed neuroimaging-driven brain age estimation techniques for identifying brain disorders, providing an overview of the state-of-the-art in brain age estimation methods.…”

Section: Literature Surveymentioning

confidence: 99%

IMNMAGN: Integrative Multimodal Approach for Enhanced Detection of Neurodegenerative Diseases Using Fusion of Multidomain Analysis With Graph Networks

Vijay Anand,

Shanmuga Priyan,

Guru Brahmam

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The burgeoning field of neurodegenerative disease detection and management necessitates the development of robust and comprehensive diagnostic approaches. Existing methodologies often fall short in effectively capturing the complex interplay of brain signals and genetic markers, which are crucial in the early detection and progression tracking of such diseases. This paper introduces a novel multimodal framework that leverages advanced signal processing and machine learning techniques to address these limitations, providing a more accurate and holistic understanding of neurodegenerative diseases. Our proposed model integrates multiple modalities: EEG signal analysis using Time-Frequency Analysis and Wavelet Transform, functional Magnetic Resonance Imaging (fMRI) analyzed through Independent Component Analysis (ICA) and Correlation Analysis, Magnetoencephalography (MEG) employing Beamforming and Source Localization Techniques, and Genomic Data analysis using Graph Neural Network for Genetic Pattern Recognition process. This integration is realized through the fusion of modalities using Gated Recurrent Units (GRU) and the classification into disease classes via an efficient 1D Convolutional Neural Network (CNN). The reasons for selecting these methods are twofold: they address the non-stationary characteristics of EEG signals and exploit spatial information of brain activity, while also identifying functional networks and genetic patterns associated with neurodegeneration conditions. The clinical impact of this work is profound. Tested on the BioGPS and BrainLat datasets, our framework demonstrated a 10.4% increase in precision, 8.5% increase in accuracy, 8.3% increase in recall, 9.4% increase in the Area Under the Curve (AUC), 7.5% increase in specificity, and a 2.9% reduction in delay compared to existing methods.

show abstract

Internet of Things for Diagnosis of Alzheimer’s Disease: A Multimodal Machine Learning Approach Based on Eye Movement Features

Cited by 9 publications

References 32 publications

Predictive models in Alzheimer's disease: an evaluation based on data mining techniques

Predictive models in Alzheimer's disease: an evaluation based on data mining techniques

Exploring Cutting-Edge Developments in Deep Learning for Biomedical Signal Processing

IMNMAGN: Integrative Multimodal Approach for Enhanced Detection of Neurodegenerative Diseases Using Fusion of Multidomain Analysis With Graph Networks

Contact Info

Product

Resources

About