TY  - JOUR
AB  - INTRODUCTION: The detrimental consequences of stress highlight the need for precise stress detection, as this offers a window for timely intervention. However, both objective and subjective measurements suffer from validity limitations. Contactless sensing technologies using machine learning methods present a potential alternative and could be used to estimate stress from externally visible physiological changes, such as emotional facial expressions. Although previous studies were able to classify stress from emotional expressions with accuracies of up to 88.32%, most works employed a classification approach and relied on data from contexts where stress was induced. Therefore, the primary aim of the present study was to clarify whether stress can be detected from facial expressions of six basic emotions (anxiety, anger, disgust, sadness, joy, love) and relaxation using a prediction approach. METHOD: To attain this goal, we analyzed video recordings of facial emotional expressions collected from n = 69 participants in a secondary analysis of a dataset from an interventional study. We aimed to explore associations with stress (assessed by the PSS-10 and a one-item stress measure). RESULTS: Comparing two regression machine learning models [Random Forest (RF) and XGBoost], we found that facial emotional expressions were promising indicators of stress scores, with model fit being best when data from all six emotional facial expressions was used to train the model (one-item stress measure: MSE (XGB) = 2.31, MAE (XGB) = 1.32, MSE (RF) = 3.86, MAE (RF) = 1.69; PSS-10: MSE (XGB) = 25.65, MAE (XGB) = 4.16, MSE (RF) = 26.32, MAE (RF) = 4.14). XGBoost showed to be more reliable for prediction, with lower error for both training and test data. DISCUSSION: The findings provide further evidence that non-invasive video recordings can complement standard objective and subjective markers of stress.
AU  - Rupp, L.H.*
AU  - Kumar, A.*
AU  - Sadeghi, M.*
AU  - Schindler-Gmelch, L.*
AU  - Keinert, M.*
AU  - Eskofier, B.M.
AU  - Berking, M.*
C1  - 74557
C2  - 57511
CY  - Avenue Du Tribunal Federal 34, Lausanne, Ch-1015, Switzerland
TI  - Stress can be detected during emotion-evoking smartphone use: A pilot study using machine learning.
JO  - Front. Digit. Health
VL  - 7
PB  - Frontiers Media Sa
PY  - 2025
SN  - 2673-253X
ER  - 

TY  - JOUR
AB  - OBJECTIVE: A study participant's informed consent, based on study information and expressed using a consent form (CF), is the ethical and legal basis for research with humans. Timely automatic access to a participant's consent status in different systems is crucial for knowing which medical data, images, and biological samples can be collected for research. To support time-critical (pandemic) research, this article evaluates a fully electronic consent management system and a consent collection process using a tablet PC in comparison to traditional paper-and-pencil-based approaches and assesses their impact on patient recruitment. MATERIALS AND METHODS: The evaluation is based on a COVID-19 study [the Sektorenübergreifende Plattform (SÜP) study; 2,753 study participants] that offered both paper-and-pencil- and tablet-based consent collection approaches and focused on the following: (a) initial CF validity and its impact on patient recruitment, (b) time-to-initial availability of structured consent information for other systems, (c) time-to-research based on completed quality assurance of CFs, and (d) feedback on both approaches from study staff and participants. RESULTS: The initial CF validity increased significantly from 67.38% for paper-and-pencil-based CFs to 99.46% for tablet-based CFs. This quality increase also reduced the number of invalid CFs or CFs requiring corrections, which can lead to study exclusion and, consequently, lower recruitment rates and lost research data. The time lag between recruitment and the availability of data decreased significantly when using tablet-based CFs, supporting time-critical research while protecting participants' privacy. Overall, the participants' and study staff's feedback on tablet-based CF collection was positive and highlighted the benefits of tablet-based CF collection in reducing the documentational burden on study staff and enabling participants to adjust the CF's appearance, for example, by choosing a bigger font size. DISCUSSION: Although tablet-based CF collection has measurable positive effects, especially on patient recruitment rates due to an increase in initially valid CFs, the majority of the National Pandemic Cohort Network (German: Nationales Pandemie Kohorten Netz, NAPKON) study sites still solely use paper-and-pencil-based processes. Since the feedback from study staff and participants was mainly positive, other barriers beyond technical availability and workflows likely exist and need to be evaluated in further settings. CONCLUSION: Fully electronic informed consent collection is the "best practice" approach to ensure valid CFs and increase initial patient inclusion rates in studies. Due to the additional benefits, including shorter time-to-research, electronic consent form collection should be integrated into pandemic response schemes and other time-critical research.
AU  - Stahl, D.*
AU  - Rau, H.*
AU  - Blumentritt, A.*
AU  - Fiedler-Lacombe, L.*
AU  - Heim, E.*
AU  - Valentin, H.*
AU  - Bialke, M.*
AU  - Kraus, M.
AU  - Hoffmann, W.*
C1  - 75783
C2  - 57995
TI  - The benefits of fully electronic consent management and consent collection via a tablet PC in supporting time-critical pandemic research-an example from a NAPKON COVID-19 project.
JO  - Front. Digit. Health
VL  - 7
PY  - 2025
SN  - 2673-253X
ER  -