Fenja T. Hesselmann
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Fenja T. Hesselmann has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Frank Wallhoff
Jade Hochschule, University of Applied Sciences, Institute for Assistive Technologies
8 shared publications - Alexander Pauls
Abt. Technik und Gesundheit für Menschen, Fachbereich Bauwesen Geoinformation Gesundheitstechnologie, Jade Hochschule Wilhelmshaven/Oldenburg/Elsfleth, Oldenburg, Germany
2 shared publications - Frauke Koppelin
Abt. Technik und Gesundheit für Menschen, Fachbereich Bauwesen Geoinformation Gesundheitstechnologie, Jade Hochschule Wilhelmshaven/Oldenburg/Elsfleth, Oldenburg, Germany
2 shared publications - Annelie Lorber
University of Bremen
1 shared publication Christine Scheve
Jade Hochschule, University of Applied Sciences
1 shared publication- Hannah Louisa Krüger
University of Bremen
1 shared publication
Latest Publications
- 8 August 2026
- Published by Springer Nature
- p. 413-426
Effective Sensor Position and Fusion Approaches for Nursing Activity Analysis
- 8 August 2026
- Published by Springer Nature
- p. 413-426
Effective Sensor Position and Fusion Approaches for Nursing Activity Analysis
- Open Access
- 2 June 2026
- Published by IOS Press
- p. 36-44
Robust Recognition of Nursing Tasks with Thermal Cameras and Body-Worn IMUs in a Simulated Care Setting
Accurate and low-burden documentation of nursing activities is essential to ensure the quality of care and for liability reasons, yet it remains time-consuming and contributes to workload and stress. Human Activity Recognition (HAR) can support automated documentation, but camera-based approaches raise privacy concerns in clinical environments. This work investigates a privacy-preserving multimodal HAR setup combining a low-resolution thermal camera with body-worn inertial measurement units (IMUs) to recognize care-related activities. In a simulated patient room, eleven trained nurses performed 13 typical nursing tasks on a simulation mannequin while thermal images and data from ten IMUs were recorded, synchronized, and classified using CNN-LSTM architectures. Using thermal images alone results an accuracy and F1 score of 90.2 %, while fusing thermal data with all IMUs increases the performance to 96.8 % accuracy (F1 96.8 %) and reduces the variance across folds. Confusion matrix analysis indicates that thermal imaging performs well for activities with distinctive spatial heat patterns, whereas IMU fusion substantially improves recognition of fine-grained actions and cases with partial occlusion.
- 2 June 2026
- Published by IOS Press
- p. 36-44
Open AccessRobust Recognition of Nursing Tasks with Thermal Cameras and Body-Worn IMUs in a Simulated Care Setting
Accurate and low-burden documentation of nursing activities is essential to ensure the quality of care and for liability reasons, yet it remains time-consuming and contributes to workload and stress. Human Activity Recognition (HAR) can support automated documentation, but camera-based approaches raise privacy concerns in clinical environments. This work investigates a privacy-preserving multimodal HAR setup combining a low-resolution thermal camera with body-worn inertial measurement units (IMUs) to recognize care-related activities. In a simulated patient room, eleven trained nurses performed 13 typical nursing tasks on a simulation mannequin while thermal images and data from ten IMUs were recorded, synchronized, and classified using CNN-LSTM architectures. Using thermal images alone results an accuracy and F1 score of 90.2 %, while fusing thermal data with all IMUs increases the performance to 96.8 % accuracy (F1 96.8 %) and reduces the variance across folds. Confusion matrix analysis indicates that thermal imaging performs well for activities with distinctive spatial heat patterns, whereas IMU fusion substantially improves recognition of fine-grained actions and cases with partial occlusion.
- 24 November 2025
- Published by Springer Nature
- p. 430-436
Identifying Employee Potential in Logistics Using AI-Supported HR Analytics
- 24 November 2025
- Published by Springer Nature
- p. 430-436
Identifying Employee Potential in Logistics Using AI-Supported HR Analytics
- 3 October 2025
- Published by IOS Press
- Vol. 330, 653-668
Nursing Activity Recognition for Automated Care Documentation in Clinical Settings
This paper introduces a concept for an assistance system that enables nursing staff in real-life clinical settings to reduce the time-consuming nursing documentation effort, e.g. from the morning routine. The overall goal is an AI-based documentation assistant that pre-fills the documentation record. An essential constraint in collecting and processing data is the use of sensors and features that preserve people’s privacy and the acceptance of being observed. Therefore, the sensors selected are known body-worn acceleration sensors as well as far-infrared based thermal scans. During the training and evaluation phase, the use of an Azure Kinect is foreseen as well. A crucial intermediate step within the concept is the autonomous identification and learning of actions that form an activity that is recognised as a part of the entire routine added to the documentation. Both open data sets and self-recorded material tailored to the use case are to be used for this purpose. By checking the automatically generated documentation results after each treatment by the nursing staff, additional training material is to be constantly generated during the application operation in order to improve the pattern recognition systems in the processing layer in the long term. The article concludes with the specific structure of the data collection and the data currently collected. In addition, three neural networks are presented that use different sensor modalities (only motion data, only thermal images or a combination of both modalities) of the recorded data to classify care activities. A 10-fold cross validation achieves an accuracy of 89.65 % for the model using motion data and thermal images.
- 3 October 2025
- Published by IOS Press
- Vol. 330, 653-668
Nursing Activity Recognition for Automated Care Documentation in Clinical Settings
This paper introduces a concept for an assistance system that enables nursing staff in real-life clinical settings to reduce the time-consuming nursing documentation effort, e.g. from the morning routine. The overall goal is an AI-based documentation assistant that pre-fills the documentation record. An essential constraint in collecting and processing data is the use of sensors and features that preserve people’s privacy and the acceptance of being observed. Therefore, the sensors selected are known body-worn acceleration sensors as well as far-infrared based thermal scans. During the training and evaluation phase, the use of an Azure Kinect is foreseen as well. A crucial intermediate step within the concept is the autonomous identification and learning of actions that form an activity that is recognised as a part of the entire routine added to the documentation. Both open data sets and self-recorded material tailored to the use case are to be used for this purpose. By checking the automatically generated documentation results after each treatment by the nursing staff, additional training material is to be constantly generated during the application operation in order to improve the pattern recognition systems in the processing layer in the long term. The article concludes with the specific structure of the data collection and the data currently collected. In addition, three neural networks are presented that use different sensor modalities (only motion data, only thermal images or a combination of both modalities) of the recorded data to classify care activities. A 10-fold cross validation achieves an accuracy of 89.65 % for the model using motion data and thermal images.
- 29 November 2024
- Published by Springer Nature
- p. 244-250
Nursing Activity Recognition for Automated Care Documentation in Clinical Settings
- 29 November 2024
- Published by Springer Nature
- p. 244-250
Nursing Activity Recognition for Automated Care Documentation in Clinical Settings
- 4 June 2024
- Published by Springer Nature
- p. 235-243
Activity Recognition of Nursing Tasks in a Hospital: Requirements and Challenges
- 4 June 2024
- Published by Springer Nature
- p. 235-243
Activity Recognition of Nursing Tasks in a Hospital: Requirements and Challenges
- 10 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Estimating Workload from Heart Rate and Game Precision in Exergames
Exergames aim to improve or maintain the user's physical and mental performance. However, to ensure an effective training while standing, users should be in a flow state in which they do not feel any effort or strain. For this purpose, the game should adapt to the player's emotional state. Therefore, a study is conducted to this treatise measuring workload and emotional state using heart rate and achieved hit accuracy at different difficulty levels to test the feasibility of these measurements in exergames. The discretely measurable data are combined with subjective assessments of users to gain insight into a correlation with individually perceived workload. A user study is adjoined in which a significant correlation (p<0.05) between the objective data (heart rate and error rate) and the subjective gaming experience can be observed. In addition, statistically significant differences are found between difficulty levels and data.
- 10 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Estimating Workload from Heart Rate and Game Precision in Exergames
Exergames aim to improve or maintain the user's physical and mental performance. However, to ensure an effective training while standing, users should be in a flow state in which they do not feel any effort or strain. For this purpose, the game should adapt to the player's emotional state. Therefore, a study is conducted to this treatise measuring workload and emotional state using heart rate and achieved hit accuracy at different difficulty levels to test the feasibility of these measurements in exergames. The discretely measurable data are combined with subjective assessments of users to gain insight into a correlation with individually perceived workload. A user study is adjoined in which a significant correlation (p<0.05) between the objective data (heart rate and error rate) and the subjective gaming experience can be observed. In addition, statistically significant differences are found between difficulty levels and data.