17 September 2023 in Expert Systems with Applications
8 September 2023 in IEEE Transactions on Biomedical Engineering
Objective: We propose an efficient approach based on a convolutional denoising autoencoder (CDA) network to reduce motion and noise artifacts (MNA) from corrupted atrial fibrillation (AF) and non-AF photoplethysmography (PPG) data segments so that an accurate PPG-signal-derived heart rate can be obtained. Our method's main innovation is the optimization of the CDA performance for both rhythms using more AF than non-AF data for training the AF-specific CDA model and vice versa for the non-AF CDA network. Methods: To evaluate this unconventional training scheme, our proposed network was trained and tested on 25-sec PPG data segments from 48 subjects from two different databases–the Pulsewatch dataset and Stanford University's publicly available PPG dataset. In total, our dataset contains 10,773 data segments: 7,001 segments for training and 3,772 independent segments from out-of-sample subjects for testing. Results: Using real-life corrupted PPG segments, our approach significantly reduced the average heart rate root mean square error (RMSE) of the reconstructed PPG segments by 45.74% and 23% compared to the corrupted non-AF and AF data, respectively. Further, our approach exhibited lower RMSE, and higher sensitivity and PPV for detected peaks compared to the reconstructed data produced by the alternative methods. Conclusion: These results show the promise of our approach as a reliable denoising method, which should be used prior to AF detection algorithms for an accurate cardiac health monitoring involving wearable devices. Significance: PPG signals collected from wearables are vulnerable to MNA, which limits their use as a reliable measurement, particularly in uncontrolled real-life environments.
30 May 2024 in JMIR Human Factors
Background: Studies evaluating the usability of mobile-phone assessments in older adults are limited. Objective: This study aims to identify design-based barriers and facilitators to mobile app survey completion among 2 samples of older adults; those in the Framingham Heart Study and a more diverse sample from a hospital-based setting. Methods: We used mixed methods to identify challenging and beneficial features of the mobile app in participants from the electronic Framingham Heart Study (n=15; mean age of 72 years; 6/15, 40% women; 15/15, 100% non-Hispanic and White) and among participants recruited from a hospital-based setting (n=15; mean age of 71 years; 7/15, 47% women; 3/15, 20% Hispanic; and 8/15, 53% non-White). A variety of app-based measures with different response formats were tested, including self-reported surveys, pictorial assessments (to indicate body pain sites), and cognitive testing tasks (eg, Trail Making Test and Stroop). Participants completed each measure using a think-aloud protocol, while being audio- and video-recorded with a qualitative interview conducted at the end of the session. Recordings were coded for participant usability errors by 2 pairs of coders. Participants completed the Mobile App Rating Scale to assess the app (response range 1=inadequate to 5=excellent). Results: In electronic Framingham Heart Study participants, the average total Mobile App Rating Scale score was 7.6 (SD 1.1), with no significant differences in the hospital-based sample. In general, participants were pleased with the app and found it easy to use. A large minority had at least 1 navigational issue, most committed only once. Most older adults did not have difficulty completing the self-reported multiple-choice measures unless it included lengthy instructions but participants had usability issues with the Stroop and Trail Making Test. Conclusions: Our methods and results help guide app development and app-based survey construction for older adults, while also giving consideration to sociodemographic differences.
17 April 2026 in JMIR Rehabilitation and Assistive Technologies
Background: Wearable devices enabling remote monitoring by surgeons of their patients have gained prominence around total joint arthroplasty (TJA), offering continuous patient data to identify those not meeting postoperative goals, thereby facilitating timely interventions. While multiple studies highlight the utility of these devices in tracking postoperative progress, a standardized approach to their application is lacking. This review aims to synthesize existing literature on the use of wearable device-tracked activity for monitoring TJA outcomes. Objective: We examined the current literature to evaluate how wearable devices are used in monitoring and improving patient rehabilitation and outcomes following TJA. Methods: A systematic review was conducted following Cochrane methodology. A literature search of all available literature was performed in April 2024 and identified 102 studies to undergo full-text review. Systematic reviews, duplicate papers, and theoretical papers were excluded. Ultimately, 35 studies met the selection criteria. Results: The review revealed that 32 of 35 (91.4%) studies used wearable devices to monitor step counts. A total of 21 (60%) studies incorporated joint-specific patient-reported outcome measures, though the specific measures varied. Further, 9 studies used standardized performance-based outcome measures, which also differed across studies. Finally, 7 (20%) studies collected sleep data; however, the methods and outcomes for sleep measurement were inconsistent among these studies. Conclusions: Remote monitoring via wearable devices offers a novel approach to tracking outcomes in TJA patients. Although the use of these devices in perioperative care is expanding, significant variability exists in the data reported across studies. Wearable monitoring is often integrated with patient-reported outcome measures and standardized functional assessments, yet the optimal data parameters that best correlate with established outcome metrics remain undefined. Additionally, data collected by wearables has not yet been shown to predict patient recovery or satisfaction. Further research is essential to refine these data parameters and the development of postoperative protocols that leverage wearable devices to enhance patient compliance and improve clinical outcomes.
22 January 2024 in JMIR Preprints
BACKGROUND Studies evaluating the usability of mobile-phone assessments in older adults are limited. OBJECTIVE This study aims to identify design-based barriers and facilitators to mobile app survey completion among 2 samples of older adults; those in the Framingham Heart Study and a more diverse sample from a hospital-based setting. METHODS We used mixed methods to identify challenging and beneficial features of the mobile app in participants from the electronic Framingham Heart Study (n=15; mean age of 72 years; 6/15, 40% women; 15/15, 100% non-Hispanic and White) and among participants recruited from a hospital-based setting (n=15; mean age of 71 years; 7/15, 47% women; 3/15, 20% Hispanic; and 8/15, 53% non-White). A variety of app-based measures with different response formats were tested, including self-reported surveys, pictorial assessments (to indicate body pain sites), and cognitive testing tasks (eg, Trail Making Test and Stroop). Participants completed each measure using a think-aloud protocol, while being audio- and video-recorded with a qualitative interview conducted at the end of the session. Recordings were coded for participant usability errors by 2 pairs of coders. Participants completed the Mobile App Rating Scale to assess the app (response range 1=inadequate to 5=excellent). RESULTS In electronic Framingham Heart Study participants, the average total Mobile App Rating Scale score was 7.6 (SD 1.1), with no significant differences in the hospital-based sample. In general, participants were pleased with the app and found it easy to use. A large minority had at least 1 navigational issue, most committed only once. Most older adults did not have difficulty completing the self-reported multiple-choice measures unless it included lengthy instructions but participants had usability issues with the Stroop and Trail Making Test. CONCLUSIONS Our methods and results help guide app development and app-based survey construction for older adults, while also giving consideration to sociodemographic differences. CLINICALTRIAL
23 September 2025 in JMIR Preprints
BACKGROUND Wearable devices enabling remote monitoring by surgeons of their patients have gained prominence around total joint arthroplasty (TJA), offering continuous patient data to identify those not meeting postoperative goals, thereby facilitating timely interventions. While multiple studies highlight the utility of these devices in tracking postoperative progress, a standardized approach to their application is lacking. This review aims to synthesize existing literature on the use of wearable device-tracked activity for monitoring TJA outcomes. OBJECTIVE We examined the current literature to evaluate how wearable devices are utilized and in monitoring and improving patient rehabilitation and outcomes following TJA. METHODS A systematic review was conducted following Cochrane methodology. A literature search of all available literature was performed in April 2024 and identified 102 studies to undergo full-text review. Systematic reviews, duplicate articles, and theoretical articles were excluded. Ultimately, 35 studies met the selection criteria. RESULTS The review revealed that 32 out of 35 studies (91.4%) employed wearable devices to monitor step counts. Twenty-one studies (60%) incorporated joint-specific patient-reported outcome measures (PROMs), though the specific measures varied. Nine studies utilized standardized performance-based outcome measures, which also differed across studies. Finally, seven studies (20%) collected sleep data, however the methods and outcomes for sleep measurement were inconsistent among these studies. CONCLUSIONS Remote monitoring via wearable devices offers a novel approach to tracking outcomes in TJA patients. Although the use of these devices in peri-operative care is expanding, significant variability exists in the data reported across studies. Wearable monitoring is often integrated with PROMs and standardized functional assessments, yet the optimal data parameters that best correlate with established outcome metrics remain undefined. Additionally, data collected by wearable has not yet been shown to predict patient recovery or satisfaction. Further research is essential to refine these data parameters and the development of post-operative protocols that leverage wearable devices to enhance patient compliance and improve clinical outcomes.