Combined use of wearable devices and Machine Learning for the measurement of thermal sensation in indoor environments

Cosoli, Gloria; Mansi, Silvia Angela; Arnesano, Marco

doi:10.1109/metrolivenv54405.2022.9826956

Cited by 8 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present manuscript, the authors used features extracted from different types of physiological signals acquired through wearable devices to train ML-based models for the classification of thermal sensation (TS). In particular, EEG, PPG, EDA, and SKT signals were considered, since a previous preliminary study by the same authors [18] verified the possibility to exploit them for TS assessment. The post-processing of these signals, carried out in both time and frequency domains, allows to obtain features that can be ingested by ML classifiers, after a proper evaluation and selection based on statistical significance with respect to TS.…”

Section: Discussionmentioning

confidence: 99%

“…In this perspective, the authors carried out a preliminary study [18] to demonstrate the feasibility of thermal sensation assessment through the combined use of Machine Learning (ML) classifiers and wearable sensors, measuring physiological signals in a fully controlled environment, and exposing the participants to predetermined thermal conditions. Results showed classification accuracy values up to 0.80, encouraging the expansion and deepening of this line of research.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Wearable devices and Machine Learning algorithms to assess indoor thermal sensation: metrological analysis

Cosoli,

Mansi,

Revel

et al. 2023

Acta IMEKO

View full text Add to dashboard Cite

Personal comfort modeling is considered the most promising solution for indoor thermal comfort management in buildings. The use of wearable sensors is investigated for the real-time measurement of physiological signals to train comfort models for buildings monitoring and control. To achieve the required reliability, different uncertainty sources should be considered and weighted in the measurement results evaluation. This study presents an example of personal comfort model (PCM) development based on wearable sensors (i.e., Empatica E4 smartband and MUSE headband) acquiring multimodal signals (i.e., photoplethysmographic – PPG, electrodermal activity – EDA, skin temperature – SKT, and electroencephalographic – EEG ones), together with a metrological characterization of the modeling procedure. Starting from the data collected within an experimental campaign on 76 subjects, different Machine Learning (ML) algorithms were exploited to create comfort models capable of predicting the human thermal sensation (TS). The most accurate model was considered to investigate the impact of sensors uncertainty through a Monte Carlo simulation. Results showed that the Random Forest model is the best performing one (accuracy: 0.86). Monte Carlo simulation method proved that the model is very robust towards measurement uncertainties of input features (expanded uncertainty of the model accuracy: ± 0.04, k = 2). This confirms the possibility to derive the subject’s TS exploiting only physiological signals; measurement uncertainty is influenced mostly by PPG and EDA signals. This kind of investigation could lead to the development of PCMs, exploitable within control systems to optimize subjects’ well-being and building energy efficiency.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Wearable devices and Machine Learning algorithms to assess indoor thermal sensation: metrological analysis

Cosoli,

Mansi,

Revel

et al. 2023

Acta IMEKO

View full text Add to dashboard Cite

show abstract

“…Krigolson et al [ 29 ] used MUSE to explore the relationship between perceived cognitive fatigue and human event-related potential (ERP) and electroencephalography (EEG) oscillations in a large sample, while validating that portable EEG devices can be a viable research tool to effectively measure relevant changes in the brain. In addition, there are also research teams that have applied MUSE to stroke disease detection [ 30 ], heat sensation measurement [ 31 ], engagement classification [ 32 ], and mood detection [ 33 ]. All of these studies have validated the MUSE wearable device as a reliable and effective EEG data collection device by exploring the changes in biomarkers extracted from EEG data.…”

Section: Methodsmentioning

confidence: 99%

Screening for Mild Cognitive Impairment with Speech Interaction Based on Virtual Reality and Wearable Devices

Wu,

Li,

Xue

et al. 2023

Brain Sciences

View full text Add to dashboard Cite

Significant advances in sensor technology and virtual reality (VR) offer new possibilities for early and effective detection of mild cognitive impairment (MCI), and this wealth of data can improve the early detection and monitoring of patients. In this study, we proposed a non-invasive and effective MCI detection protocol based on electroencephalogram (EEG), speech, and digitized cognitive parameters. The EEG data, speech data, and digitized cognitive parameters of 86 participants (44 MCI patients and 42 healthy individuals) were monitored using a wearable EEG device and a VR device during the resting state and task (the VR-based language task we designed). Regarding the features selected under different modality combinations for all language tasks, we performed leave-one-out cross-validation for them using four different classifiers. We then compared the classification performance under multimodal data fusion using features from a single language task, features from all tasks, and using a weighted voting strategy, respectively. The experimental results showed that the collaborative screening of multimodal data yielded the highest classification performance compared to single-modal features. Among them, the SVM classifier using the RBF kernel obtained the best classification results with an accuracy of 87%. The overall classification performance was further improved using a weighted voting strategy with an accuracy of 89.8%, indicating that our proposed method can tap into the cognitive changes of MCI patients. The MCI detection scheme based on EEG, speech, and digital cognitive parameters proposed in this study provides a new direction and support for effective MCI detection, and suggests that VR and wearable devices will be a promising direction for easy-to-perform and effective MCI detection, offering new possibilities for the exploration of VR technology in the field of language cognition.

show abstract

“…in America 21% of citizens between 18 and 49 years wear a smartwatch or a smartband [5] ). There is a plethora of application fields: from health [6 , 7] to industry [8 , 9] , through sport [10 , 11] , and rehabilitation [12 , 13] , just to cite some. The role of wearable sensors in the remote monitoring of physiological parameters is pivotal, given their capability to acquire data 24 hours a day, 7 days per week; for this reason, their use has been often combined to Artificial Intelligence (AI) techniques for both classification and regression purposes [14 , 15] .…”

Section: Overviewmentioning

confidence: 99%

Methods for the metrological characterization of wearable devices for the measurement of physiological signals: state of the art and future challenges

2023

View full text Add to dashboard Cite

Combined use of wearable devices and Machine Learning for the measurement of thermal sensation in indoor environments

Cited by 8 publications

References 21 publications

Wearable devices and Machine Learning algorithms to assess indoor thermal sensation: metrological analysis

Wearable devices and Machine Learning algorithms to assess indoor thermal sensation: metrological analysis

Screening for Mild Cognitive Impairment with Speech Interaction Based on Virtual Reality and Wearable Devices

Methods for the metrological characterization of wearable devices for the measurement of physiological signals: state of the art and future challenges

Contact Info

Product

Resources

About