TraMiner: Vision-Based Analysis of Locomotion Traces for Cognitive Assessment in Smart-Homes

Zolfaghari, Samaneh; Khodabandehloo, Elham; Riboni, Daniele

doi:10.1007/s12559-020-09816-3

Cited by 15 publications

(18 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…“TraMiner” is a novel trajectory mining system to be used in smart homes where sensor infrastructures exist [ 125 ]. Both trajectory and speed images were analyzed, and several time intervals between consecutive sensor activations were proved (from 30 s to 180 s).…”

Section: Resultsmentioning

confidence: 99%

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer’s Disease Not Using Neuroimaging Biomarkers: A Systematic Review

Cabrera-León,

Báez,

Fernández-López

et al. 2024

JAD

View full text Add to dashboard Cite

Background: The growing number of older adults in recent decades has led to more prevalent geriatric diseases, such as strokes and dementia. Therefore, Alzheimer’s disease (AD), as the most common type of dementia, has become more frequent too. Background: Objective: The goals of this work are to present state-of-the-art studies focused on the automatic diagnosis and prognosis of AD and its early stages, mainly mild cognitive impairment, and predicting how the research on this topic may change in the future. Methods: Articles found in the existing literature needed to fulfill several selection criteria. Among others, their classification methods were based on artificial neural networks (ANNs), including deep learning, and data not from brain signals or neuroimaging techniques were used. Considering our selection criteria, 42 articles published in the last decade were finally selected. Results: The most medically significant results are shown. Similar quantities of articles based on shallow and deep ANNs were found. Recurrent neural networks and transformers were common with speech or in longitudinal studies. Convolutional neural networks (CNNs) were popular with gait or combined with others in modular approaches. Above one third of the cross-sectional studies utilized multimodal data. Non-public datasets were frequently used in cross-sectional studies, whereas the opposite in longitudinal ones. The most popular databases were indicated, which will be helpful for future researchers in this field. Conclusions: The introduction of CNNs in the last decade and their superb results with neuroimaging data did not negatively affect the usage of other modalities. In fact, new ones emerged.

show abstract

Section: Resultsmentioning

confidence: 99%

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer’s Disease Not Using Neuroimaging Biomarkers: A Systematic Review

Cabrera-León,

Báez,

Fernández-López

et al. 2024

JAD

View full text Add to dashboard Cite

show abstract

“…With this feature extraction method, the authors obtain accuracy rates close to 80% by using different machine learning algorithms, such as SVM or MLP. The TraMiner prototype [ 44 ] is also capable of recognizing locomotion patterns inside homes. In order to develop this prototype, authors have used information from the trajectories of 153 elderly people, including people with cognitive problems.…”

Section: Related Workmentioning

confidence: 99%

SDHAR-HOME: A Sensor Dataset for Human Activity Recognition at Home

Ramos

Domingo

Zalama

et al. 2022

Sensors

View full text Add to dashboard Cite

Nowadays, one of the most important objectives in health research is the improvement of the living conditions and well-being of the elderly, especially those who live alone. These people may experience undesired or dangerous situations in their daily life at home due to physical, sensorial or cognitive limitations, such as forgetting their medication or wrong eating habits. This work focuses on the development of a database in a home, through non-intrusive technology, where several users are residing by combining: a set of non-intrusive sensors which captures events that occur in the house, a positioning system through triangulation using beacons and a system for monitoring the user’s state through activity wristbands. Two months of uninterrupted measurements were obtained on the daily habits of 2 people who live with a pet and receive sporadic visits, in which 18 different types of activities were labelled. In order to validate the data, a system for the real-time recognition of the activities carried out by these residents was developed using different current Deep Learning (DL) techniques based on neural networks, such as Recurrent Neural Networks (RNN), Long Short-Term Memory networks (LSTM) or Gated Recurrent Unit networks (GRU). A personalised prediction model was developed for each user, resulting in hit rates ranging from 88.29% to 90.91%. Finally, a data sharing algorithm has been developed to improve the generalisability of the model and to avoid overtraining the neural network.

show abstract

“…The use of computer vision and deep neural networks for recognizing cognitive impairment based on locomotion traces acquired in smart-homes was investigated by Zolfaghari et al in [68]. While deep learning methods may achieve high recognition rates, they require large training sets of labeled data, which may be hard to acquire in real-world environments, especially in sensitive domains like the one addressed in this work.…”

Section: Cps For Healthcarementioning

confidence: 99%

FreeSia: A Cyber-physical System for Cognitive Assessment through Frequency-domain Indoor Locomotion Analysis

Khodabandehloo

Alimohammadi

Riboni

2022

ACM Trans. Cyber-Phys. Syst.

Self Cite

View full text Add to dashboard Cite

Thanks to the seamless integration of sensing, networking, and artificial intelligence, cyber-physical systems promise to improve healthcare by increasing efficiency and reducing costs. Specifically, cyber-physical systems are being increasingly applied in smart-homes to support independent and healthy ageing. Due to the growing prevalence of noncommunicable diseases in the senior population, a key application in this domain is the detection of cognitive issues based on sensor data. In this paper, we propose a novel cyber-physical system for cognitive assessment in smart-homes. Cognitive evaluation relies on clinical indicators characterizing symptoms of dementia based on the individual’s movement patterns. However, recognizing these patterns in smart-homes is challenging, because movement is constrained by the home layout and obstacles. Since different abnormal patterns are characterized by undulatory-like trajectories, we conjecture that frequency-based locomotion features may more effectively capture these patterns with respect to traditional features in the spatio-temporal domain. Based on this intuition, we introduce novel feature extraction techniques, and adopt state of the art machine learning algorithms for short- and long-term cognitive evaluation. Our system includes a user-friendly interface that enables clinicians to inspect the data and predictions. Extensive experiments carried out with a real-world dataset acquired from both cognitively healthy seniors and people with dementia show the superiority of our frequency-based features. Moreover, further experiments with an ensemble method show that prediction accuracy can be enhanced by combining features in the frequency and time domains.

show abstract

TraMiner: Vision-Based Analysis of Locomotion Traces for Cognitive Assessment in Smart-Homes

Cited by 15 publications

References 56 publications

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer’s Disease Not Using Neuroimaging Biomarkers: A Systematic Review

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer’s Disease Not Using Neuroimaging Biomarkers: A Systematic Review

SDHAR-HOME: A Sensor Dataset for Human Activity Recognition at Home

FreeSia: A Cyber-physical System for Cognitive Assessment through Frequency-domain Indoor Locomotion Analysis

Contact Info

Product

Resources

About