An explainable COVID-19 detection system based on human sounds.

Acoustic signals generated by the human body have often been used as biomarkers to diagnose and monitor diseases. As the pathogenesis of COVID-19 indicates impairments in the respiratory system, digital acoustic biomarkers of COVID-19 are under investigation. In this paper, we explore an accurate and explainable COVID-19 diagnosis approach based on human speech, cough, and breath data using the power of machine learning. We first analyze our design space considerations from the data aspect and model aspect. Then, we perform data augmentation, Mel-spectrogram transformation, and develop a deep residual architecture-based model for prediction. Experimental results show that our system outperforms the baseline, with the ROC-AUC result increased by 5.47%. Finally, we perform an interpretation analysis based on the visualization of the activation map to further validate the model.

An explainable COVID-19 detection system based on human sounds

Acoustic signals generated by the human body have often been used as biomarkers to diagnose and monitor diseases. As the pathogenesis of COVID-19 indicates impairments in the respiratory system, digital acoustic biomarkers of COVID-19 are under investigation. In this paper, we explore an accurate and explainable COVID-19 diagnosis approach based on human speech, cough, and breath data using the power of machine learning. We first analyze our design space considerations from the data aspect and model aspect. Then, we perform data augmentation, Mel-spectrogram transformation, and develop a deep residual architecture-based model for prediction. Experimental results show that our system outperforms the baseline, with the ROC-AUC result increased by 5.47%. Finally, we perform an interpretation analysis based on the visualization of the activation map to further validate the model.

Stress prediction using micro-EMA and machine learning during COVID-19 social isolation.

Accurately predicting users' perceived stress is beneficial to aid early intervention and prevent both mental illness and physical disease during the COVID-19 pandemic. However, the existing perceived stress predicting system needs to collect a large amount of previous data for training but has a limited prediction range (i.e., next 1-2 days). Therefore, we propose a perceived stress prediction system based on the history data of micro-EMA for identifying risks 7 days earlier. Specifically, we first select and deliver an optimal set of micro-EMA questions to users every Monday, Wednesday, and Friday for reducing the burden. Then, we extract time-series features from the past micro-EMA responses and apply an Elastic net regularization model to discard redundant features. After that, selected features are fed to an ensemble prediction model for forecasting fine-grained perceived stress in the next 7 days. Experiment results show that our proposed prediction system can achieve around 4.26 (10.65% of the scale) mean absolute error for predicting the next 7 day's PSS scores, and higher than 81% accuracy for predicting the next 7 day's stress labels.

Usability, Usefulness, and Acceptance of a Novel, Portable Rehabilitation System (mRehab) Using Smartphone and 3D Printing Technology: Mixed Methods Study.

BACKGROUND: Smart technology use in rehabilitation is growing and can be used remotely to assist clients in self-monitoring their performance. With written home exercise programs being the commonly prescribed form of rehabilitation after discharge, mobile health technology coupled with task-oriented programs can enhance self-management of upper extremity training. In the current study, a rehabilitation system, namely mRehab, was designed that included a smartphone app and 3D-printed household items such as mug, bowl, key, and doorknob embedded with a smartphone. The app interface allowed the user to select rehabilitation activities and receive feedback on the number of activity repetitions completed, time to complete each activity, and quality of movement. OBJECTIVE: This study aimed to assess the usability, perceived usefulness, and acceptance of the mRehab system by individuals with stroke and identify the challenges experienced by them when using the system remotely in a home-based setting. METHODS: A mixed-methods approach was used with 11 individuals with chronic stroke. Following training, individuals with stroke used the mRehab system for 6 weeks at home. Each participant completed surveys and engaged in a semistructured interview. Participants' qualitative reports regarding the usability of mRehab were integrated with their survey reports and quantitative performance data. RESULTS: Of the 11 participants, 10 rated the mRehab system between the 67.5th and 97.5th percentile on the System Usability Scale, indicating their satisfaction with the usability of the system. Participants also provided high ratings of perceived usefulness (mean 5.8, SD 0.9) and perceived ease of use (mean 5.3, SD 1.5) on a 7-point scale based on the Technology Acceptance Model. Common themes reported by participants showed a positive response to mRehab with some suggestions for improvements. Participants reported an interest in activities they perceived to be adequately challenging. Some participants indicated a need for customizing the feedback to be more interpretable. Overall, most participants indicated that they would like to continue using the mRehab system at home. CONCLUSIONS: Assessing usability in the lived environment over a prolonged duration of time is essential to identify the match between the system and users' needs and preferences. While mRehab was well accepted, further customization is desired for a better fit with the end users. TRIAL REGISTRATION: ClinicalTrials.gov NCT04363944; https://clinicaltrials.gov/ct2/show/NCT04363944.