Early Detection of Mild Cognitive Impairment Progression Using Non-Wearable Sensor Data – a Deep Learning Approach

Narasimhan, Rajaram; Muthukumaran, G.; McGlade, Charles; Ramakrishnan, Anantha

doi:10.1109/b-htc50970.2020.9297882

Cited by 3 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At each time step, the layer adds information to or removes information from the cell state, and this operation is regulated through special components called gates. For further details on these gates, the mathematical relationship among the gates, cell states, and output, the reader is referred to [16]. On the output layer, the SoftMax function is used to derive the final output "ỹ 11 " (corresponds to the 11th timestep given the 10 timesteps as input sequence).…”

Section: Lstm-based Disease Progression Modelingmentioning

confidence: 99%

“…An extension of the plain RNN, Long short-term memory (LSTM) RNN consists of LSTM cells, which are building units of the network, and these cells enable RNNs to capture long-term dependencies that exist in sequential input. Therefore, the LSTM RNN is an apt choice to model the AD progression by leveraging the temporal patterns found in activity trend data [16]. The activity trend data characterized by time series statistics as input to the LSTM network will further reinforce their prediction skill along with their natural ability to recognize the long-term dependencies in sequential data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Employing Deep-Learning Approach for the Early Detection of Mild Cognitive Impairment Transitions through the Analysis of Digital Biomarkers

Narasimhan,

Gopalan,

Sikkandar

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

Mild cognitive impairment (MCI) is the precursor to the advanced stage of Alzheimer’s disease (AD), and it is important to detect the transition to the MCI condition as early as possible. Trends in daily routines/activities provide a measurement of cognitive/functional status, particularly in older adults. In this study, activity data from longitudinal monitoring through in-home ambient sensors are leveraged in predicting the transition to the MCI stage at a future time point. The activity dataset from the Oregon Center for Aging and Technology (ORCATECH) includes measures representing various domains such as walk, sleep, etc. Each sensor-captured activity measure is constructed as a time series, and a variety of summary statistics is computed. The similarity between one individual’s activity time series and that of the remaining individuals is also computed as distance measures. The long short-term memory (LSTM) recurrent neural network is trained with time series statistics and distance measures for the prediction modeling, and performance is evaluated by classification accuracy. The model outcomes are explained using the SHapley Additive exPlanations (SHAP) framework. LSTM model trained using the time series statistics and distance measures outperforms other modeling scenarios, including baseline classifiers, with an overall prediction accuracy of 83.84%. SHAP values reveal that sleep-related features contribute the most to the prediction of the cognitive stage at the future time point, and this aligns with the findings in the literature. Findings from this study not only demonstrate that a practical, less expensive, longitudinal monitoring of older adults’ activity routines can benefit immensely in modeling AD progression but also unveil the most contributing features that are medically applicable and meaningful.

show abstract

Section: Lstm-based Disease Progression Modelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Employing Deep-Learning Approach for the Early Detection of Mild Cognitive Impairment Transitions through the Analysis of Digital Biomarkers

Narasimhan,

Gopalan,

Sikkandar

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…To predict the progression of MCI in older adults, Narasimhan et al [ 31 ] applied the RNN-LSTM model on activity data acquired from sensors and the subject’s health data recorded at different moments. The authors have focused on leveraging the temporal evolution of MCI in order to improve its future prediction according to its development in time.…”

Section: Related Workmentioning

confidence: 99%

Addressing Mild Cognitive Impairment and Boosting Wellness for the Elderly through Personalized Remote Monitoring

2022

View full text Add to dashboard Cite

Mild cognitive impairment (MCI) may occur with old age and is associated with increased cognitive deterioration compared to what is normal. This may affect the person’s quality of life, health, and independence. In this ageing worldwide context, early diagnosis and personalized assistance for MCI therefore become crucial. This paper makes two important contributions: (1) a system (RO-SmartAgeing) to address MCI, which was developed for Romania; and (2) a set of criteria for evaluating its impact on remote health monitoring. The system aims to provide customized non-invasive remote monitoring, health assessment, and assistance for the elderly within a smart environment set up in their homes. Moreover, it includes multivariate AI-based predictive models that can detect the onset of MCI and its development towards dementia. It was built iteratively, following literature reviews and consultations with health specialists, and it is currently being tested in a simulated home environment. While its main strength is the potential to detect MCI early and follow its evolution, RO-SmartAgeing also supports elderly people in living independently, and it is safe, comfortable, low cost, and privacy protected. Moreover, it can be used by healthcare institutions to continuously monitor a patient’s vital signs, position, and activities, and to deliver reminders and alarms.

show abstract

Recurrent Neural Network based Prediction of Transition to Mild Cognitive Impairment Using Unobtrusive Sensor Data

Narasimhan

Muthukumaran

Dey

et al. 2022

2022 IEEE International Conference on Data Science and Information System (ICDSIS)

View full text Add to dashboard Cite

Early Detection of Mild Cognitive Impairment Progression Using Non-Wearable Sensor Data – a Deep Learning Approach

Cited by 3 publications

References 26 publications

Employing Deep-Learning Approach for the Early Detection of Mild Cognitive Impairment Transitions through the Analysis of Digital Biomarkers

Employing Deep-Learning Approach for the Early Detection of Mild Cognitive Impairment Transitions through the Analysis of Digital Biomarkers

Addressing Mild Cognitive Impairment and Boosting Wellness for the Elderly through Personalized Remote Monitoring

Recurrent Neural Network based Prediction of Transition to Mild Cognitive Impairment Using Unobtrusive Sensor Data

Contact Info

Product

Resources

About