Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Gerald Bieber has a total of 54 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks

    by Arash Keshavarz, Gerald Bieber, Daniel Wulff, Carsten Babian, Stefan Lüdtke

    Accurate estimation of hematoma age remains a major challenge in forensic practice, as current assessments rely heavily on subjective visual interpretation. Hyperspectral imaging (HSI) captures rich spectral signatures that may reflect the biochemical evolution of hematomas over time. This study evaluates whether a convolutional neural network (CNN) integrating both spectral and spatial information improves hematoma age estimation accuracy. Additionally, we investigate whether performance can be maintained using a reduced, physiologically motivated subset of wavelengths. Using a dataset of forearm hematomas from 25 participants, we applied radiometric normalization and SAM-based segmentation to extract 64×64×204 hyperspectral patches. In leave-one-subject-out cross-validation, the CNN outperformed a spectral-only Lasso baseline, reducing the mean absolute error (MAE) from 3.24 days to 2.29 days. Band-importance analysis combining SmoothGrad and occlusion sensitivity identified 20 highly informative wavelengths; using only these bands matched or exceeded the accuracy of the full 204-band model across early, middle, and late hematoma stages. These results demonstrate that spectral–spatial modeling and physiologically grounded band selection can enhance estimation accuracy while significantly reducing data dimensionality. This approach supports the development of compact multispectral systems for objective clinical and forensic evaluation.

  • Illusion of Precision: How Averaging Undermines Pulse Measurement Accuracy

    by Gerald Bieber, Erik Endlicher, Timo Forstner, Milad Geravand

  • Open Access

    Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks

    by Arash Keshavarz, Gerald Bieber, Daniel Wulff, Carsten Babian, Stefan Lüdtke

    Accurate estimation of hematoma age remains a major challenge in forensic practice, where assessments rely largely on subjective visual interpretation. Hyperspectral imaging (HSI) captures rich spectral signatures that may reflect the biochemical evolution of hematomas. We evaluate whether integrating spectral and spatial information with a convolutional neural network (CNN) improves hematoma age estimation and whether a reduced, physiologically motivated subset of wavelengths can maintain performance. A forearm hematoma dataset from 25 participants was processed using radiometric normalization, SAM-based segmentation, and extraction of 64 × 64 × 204 hyperspectral patches. Using leave-one-subject-out cross-validation, the CNN achieved substantially lower mean absolute error (MAE 2.29 days) compared to a spectral-only Lasso baseline (MAE 3.24 days). Bandimportance analysis combining SmoothGrad and occlusion sensitivity identified 20 highly informative wavelengths, and using only these bands matched or exceeded the accuracy of the full 204-band model across early, middle, and late hematoma stages. These results show that spectral-spatial modeling and physiologically grounded band selection can significantly enhance hematoma age estimation while reducing data dimensionality, supporting the development of compact multispectral systems for objective clinical and forensic evaluation.


  • Open Access

    Technical Analysis for Electro-Sensory 3D Foot Scanning in a Shoe

    by Denys J.C. Matthies, Maximilian Kasbohm, Gerald Bieber, Troy Nachtigall


  • Supporting Thermal Imaging for Activity and Health Recognition by a Constant Temperature Device

    by Gerald Bieber, Erik Endlicher, Christopher Wald, Peter Gross, Bastian Kubsch

  • Objective Measurement of Stress Resilience: Is RSA a Possible Indicator?

    by Erik Endlicher, Gerald Bieber, Edda Jaleel, Angelina Schmidt, Michael Fellmann

  • Open Access

    Smartwatch-Based Interventions for People With Dementia: User-Centered Design Approach

    by Doreen Goerss, Stefanie Köhler, Eleonora Rong, Anna Gesine Temp, Ingo Kilimann, Gerald Bieber, Stefan Teipel

    Background: Assistive technologies can help people living with dementia maintain their everyday activities. Nevertheless, there is a gap between the potential and use of these materials. Involving future users may help close this gap, but the impact on people with dementia is unclear. Objective: We aimed to determine if user-centered development of smartwatch-based interventions together with people with dementia is feasible. In addition, we evaluated the extent to which user feedback is plausible and therefore helpful for technological improvements. Methods: We examined the interactions between smartwatches and people with dementia or people with mild cognitive impairment. All participants were prompted to complete 2 tasks (drinking water and a specific cognitive task). Prompts were triggered using a smartphone as a remote control and were repeated up to 3 times if participants failed to complete a task. Overall, 50% (20/40) of the participants received regular prompts, and 50% (20/40) received intensive audiovisual prompts to perform everyday tasks. Participants’ reactions were observed remotely via cameras. User feedback was captured via questionnaires, which included topics like usability, design, usefulness, and concerns. The internal consistency of the subscales was calculated. Plausibility was also checked using qualitative approaches. Results: Participants noted their preferences for particular functions and improvements. Patients struggled with rating using the Likert scale; therefore, we assisted them with completing the questionnaire. Usability (mean 78 out of 100, SD 15.22) and usefulness (mean 9 out of 12) were rated high. The smartwatch design was appealing to most participants (31/40, 76%). Only a few participants (6/40, 15%) were concerned about using the watch. Better usability was associated with better cognition. The observed success and self-rated task comprehension were in agreement for most participants (32/40, 80%). In different qualitative analyses, participants’ responses were, in most cases, plausible. Only 8% (3/40) of the participants were completely unaware of their irregular task performance. Conclusions: People with dementia can have positive experiences with smartwatches. Most people with dementia provided valuable information. Developing assistive technologies together with people with dementia can help to prioritize the future development of functional and nonfunctional features.


  • Open Access

    End-to-End Premature Ventricular Contraction Detection Using Deep Neural Networks

    by Dimitri Kraft, Gerald Bieber, Peter Jokisch, Peter Rumm

    In Holter monitoring, the precise detection of standard heartbeats and ventricular premature contractions (PVCs) is paramount for accurate cardiac rhythm assessment. This study introduces a novel application of the 1D U-Net neural network architecture with the aim of enhancing PVC detection in Holter recordings. Training data comprised the Icentia 11k and INCART DB datasets, as well as our custom dataset. The model’s efficacy was subsequently validated against traditional Holter analysis methodologies across multiple databases, including AHA DB, MIT 11 DB, and NST, as well as another custom dataset that was specifically compiled by the authors encompassing challenging real-world examples. The results underscore the 1D U-Net model’s prowess in QRS complex detection, achieving near-perfect balanced accuracy scores across all databases. PVC detection exhibited variability, with balanced accuracy scores ranging from 0.909 to 0.986. Despite some databases, like the AHA DB, showcasing lower sensitivity metrics, their robust, balanced accuracy accentuates the model’s equitable performance in discerning both false positives and false negatives. In conclusion, while the 1D U-Net architecture is a formidable tool for QRS detection, there’s a clear avenue for further refinement in its PVC detection capability, given the inherent complexities and noise challenges in real-world PVC occurrences.


  • Open Access

    End-to-End Premature Ventricular Contraction Detection using Deep Neural Networks

    by Dimitri Kraft, Gerald Bieber, Peter Jokisch, Peter Rumm

    In Holter monitoring, the precise detection of standard heartbeats and Ventricular Premature Contractions (PVCs) is paramount for accurate cardiac rhythm assessment. This study introduces a novel application of the 1D U-Net neural network architecture aimed at enhancing PVC detection in Holter recordings. Training data comprised the Icentia11k, INCART DB, and our custom dataset. The model’s efficacy was subsequently validated against traditional Holter analysis methodologies across multiple databases, including AHA DB, MIT 11 DB, NST, and another custom dataset encompassing challenging real-world examples. The results underscored the 1D U-Net model’s prowess in QRS complex detection, achieving near-perfect balanced accuracy scores across all databases. PVC detection exhibited variability, with balanced accuracy scores ranging from 0.909 to 0.986. Despite some databases, like the AHA DB, showcasing lower sensitivity metrics, their robust balanced accuracy accentuates the model’s equitable performance in discerning both false positives and negatives. In conclusion, while the 1D U-Net architecture is a formidable tool for QRS detection, there’s a clear avenue for further refinement in its PVC detection capability, given the inherent complexities and noise challenges in real-world PVC occurrences.