empatica e4 bracelet  (Empatica Inc)

Journal: Biosensors

Article Title: Fusing Wearable Biosensors with Artificial Intelligence for Mental Health Monitoring: A Systematic Review

doi: 10.3390/bios15040202

Figure Lengend Snippet: Biosensing papers reviewed focusing on stress. The data collection environment (Env.) is specified as IL (In-Lab) or IW (In-the-Wild).

Article Snippet: (Campanella et al., 2023) [ ] Stress detection , 29 , ❖ Empatica E4 bracelet HRV derived from PPG and EDA/GSR , ❖ Binary classification ● Supervised ML ● 71% precision, 60% recall, and 65% F1-score (random forest (RF)) , IL , ● Labels: stress/no-stress classifications during cognitive, social, and physiological stress tasks ● 2 labels/participant (stressed and non-stressed) ● Monitoring: approximately 15–20 min per task , ● Unbalanced data impacts classifier performance ● Small sample size and lab setting limit generalizability ● Noise from Empatica E4 and lack of real-life stressors reduce accuracy ● Larger, diverse samples and varied stressors needed for robustness.

Techniques: Derivative Assay, Activity Assay, Plasmid Preparation, Biomarker Discovery, Introduce, Battery, Transmission Assay, Control, Standard Deviation, Selection

Journal: Biosensors

Article Title: Fusing Wearable Biosensors with Artificial Intelligence for Mental Health Monitoring: A Systematic Review

doi: 10.3390/bios15040202

Figure Lengend Snippet: Biosensing papers reviewed focusing on depression.

Article Snippet: (Campanella et al., 2023) [ ] Stress detection , 29 , ❖ Empatica E4 bracelet HRV derived from PPG and EDA/GSR , ❖ Binary classification ● Supervised ML ● 71% precision, 60% recall, and 65% F1-score (random forest (RF)) , IL , ● Labels: stress/no-stress classifications during cognitive, social, and physiological stress tasks ● 2 labels/participant (stressed and non-stressed) ● Monitoring: approximately 15–20 min per task , ● Unbalanced data impacts classifier performance ● Small sample size and lab setting limit generalizability ● Noise from Empatica E4 and lack of real-life stressors reduce accuracy ● Larger, diverse samples and varied stressors needed for robustness.

Techniques: Activity Assay, Expressing, Control, Introduce, Derivative Assay, Biomarker Discovery, Diagnostic Assay, Selection

Journal: Scientific Data

Article Title: Speech production under stress for machine learning: multimodal dataset of 79 cases and 8 signals

doi: 10.1038/s41597-024-03991-w

Figure Lengend Snippet: Dataset summary.

Article Snippet: Technology Types(s) , ECG - FAROS sensor; EDA - EMPATICA E4 bracelet; speech - microphones SHURE SM-35XLR7 headset/XY microphone connected to Zoom H4n recorder; Video sequences - Panasonic HC-VX9805 cameras; Stress - PSS14/STAI Y2; Cognitive load - NASA TLX.

Techniques: Activity Assay

Journal: Scientific Data

Article Title: Speech production under stress for machine learning: multimodal dataset of 79 cases and 8 signals

doi: 10.1038/s41597-024-03991-w

Figure Lengend Snippet: Available data streams.

Article Snippet: Technology Types(s) , ECG - FAROS sensor; EDA - EMPATICA E4 bracelet; speech - microphones SHURE SM-35XLR7 headset/XY microphone connected to Zoom H4n recorder; Video sequences - Panasonic HC-VX9805 cameras; Stress - PSS14/STAI Y2; Cognitive load - NASA TLX.

Techniques: Activity Assay

Journal: Scientific Data

Article Title: Speech production under stress for machine learning: multimodal dataset of 79 cases and 8 signals

doi: 10.1038/s41597-024-03991-w

Figure Lengend Snippet: Dataset summary.

Article Snippet: Heart electrical activity captured by the Faros 180 ( https://www.bittium.com/medical/bittium-faros ) sensor Electro-Dermal Activity (EDA) captured by the Empatica E4 ( https://www.empatica.com/research/e4/ ) bracelet Audio recordings captured by multiple microphones (1x Shure SM-35XLR7 headset; 1x XY microphone through the Zoom H4n recorder) Video recordings using 4x Panasonic HC-VX9805 camera.

Techniques: Activity Assay

Journal: Scientific Data

Article Title: Speech production under stress for machine learning: multimodal dataset of 79 cases and 8 signals

doi: 10.1038/s41597-024-03991-w

Figure Lengend Snippet: Available data streams.

Article Snippet: Heart electrical activity captured by the Faros 180 ( https://www.bittium.com/medical/bittium-faros ) sensor Electro-Dermal Activity (EDA) captured by the Empatica E4 ( https://www.empatica.com/research/e4/ ) bracelet Audio recordings captured by multiple microphones (1x Shure SM-35XLR7 headset; 1x XY microphone through the Zoom H4n recorder) Video recordings using 4x Panasonic HC-VX9805 camera.

Techniques: Activity Assay

Journal: Sensors (Basel, Switzerland)

Article Title: Workplace Well-Being in Industry 5.0: A Worker-Centered Systematic Review

doi: 10.3390/s24175473

Figure Lengend Snippet: Summary of the selected articles. Authors, year of publication, context of application, monitored activity, data collection technology, type of ergonomic parameter (when applicable), and data processing/analysis approach (when applicable) are reported. List of abbreviations: ML/DL Alg. = machine learning/deep learning algorithm, Ergo. = physical or mental ergonomics, Std. ergo. index/Phys. measure = standard ergonomic index/physiological measure.

Article Snippet: HRV data were collected using the Smartband 2 from Sony ® and the Activity bracelet E4 by Empatica ® .

Techniques: Activity Assay, Marker, Blocking Assay, Control