Carsten Babian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Carsten Babian has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
J. Dreßler
Institut für Rechtsmedizin Leipzig, Leipzig, Germany
3 shared publications- Stefan Lüdtke
University of Rostock
3 shared publications - Arash Keshavarz
Visual Computing, Fraunhofer-Institute for Computer Graphics Research IGD, 18059 Rostock, Germany
2 shared publications - Daniel Wulff
Institute of Visual & Analytic Computing, University of Rostock, 18059 Rostock, Germany
2 shared publications - Gerald Bieber
Fraunhofer IGD
2 shared publications - Benjamin Ondruschka
Institute of Legal Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
1 shared publication
Latest Publications
- Open Access
- 11 February 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (2) , 78
Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks
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 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.- 11 February 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (2) , 78
Open AccessAge Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks
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 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. - Open Access
- 24 December 2025
- Published by MDPI AG in Preprints.org
Age Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks
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.
- 24 December 2025
- Published by MDPI AG in Preprints.org
Open AccessAge Prediction of Hematoma from Hyperspectral Images Using Convolutional Neural Networks
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
- 11 September 2025
- Published by Springer Nature in Forensic Science, Medicine and Pathology
- Vol. 21 (4) , 1699-1708
Age prediction of hematoma using hyperspectral imaging (HSI)
Hyperspectral imaging (HSI) analyzes the reflected light spectrum of an object, providing insights into its material composition. In this experimental, prospective study, standardized hematomas were created in subjects and observed over 21 days using a portable hyperspectral camera, aiming to correlate changes in the reflected light spectrum with hematoma age and enable an objective method for age determination of hematomas. In 25 young and healthy subjects, a hematoma was induced by injecting 3 ml of autologous blood on the volar side of the forearm. The hematomas were documented over a 7-day period in 24-hour intervals and then for an additional 14 days in 48-hour intervals using a portable HSI camera covering the wavelength range from 400 nm to 1000 nm. The datasets from the hematomas were normalized using the spectrum of unaffected skin. Lasso regression models were trained on the normalized data to identify age-dependent changes in the reflection spectra and to make age predictions for hematomas of unknown age. The different regression models were compared based on their predictive accuracy and then evaluated against the actual hematoma age. The best age prediction model based on HSI recordings achieved a mean prediction error of 3.35 days over a 21-day observation period.Expanding the data set and methodology, as well as combining them with other techniques and testing on naturally occurring hematomas of various sizes and locations, are necessary steps to transfer this method into practical application.
- 11 September 2025
- Published by Springer Nature in Forensic Science, Medicine and Pathology
- Vol. 21 (4) , 1699-1708
Open AccessAge prediction of hematoma using hyperspectral imaging (HSI)
Hyperspectral imaging (HSI) analyzes the reflected light spectrum of an object, providing insights into its material composition. In this experimental, prospective study, standardized hematomas were created in subjects and observed over 21 days using a portable hyperspectral camera, aiming to correlate changes in the reflected light spectrum with hematoma age and enable an objective method for age determination of hematomas. In 25 young and healthy subjects, a hematoma was induced by injecting 3 ml of autologous blood on the volar side of the forearm. The hematomas were documented over a 7-day period in 24-hour intervals and then for an additional 14 days in 48-hour intervals using a portable HSI camera covering the wavelength range from 400 nm to 1000 nm. The datasets from the hematomas were normalized using the spectrum of unaffected skin. Lasso regression models were trained on the normalized data to identify age-dependent changes in the reflection spectra and to make age predictions for hematomas of unknown age. The different regression models were compared based on their predictive accuracy and then evaluated against the actual hematoma age. The best age prediction model based on HSI recordings achieved a mean prediction error of 3.35 days over a 21-day observation period.Expanding the data set and methodology, as well as combining them with other techniques and testing on naturally occurring hematomas of various sizes and locations, are necessary steps to transfer this method into practical application.
- Open Access
- 25 July 2025
- Published by Elsevier in Forensic Science International
- Vol. 376, 112583
Detection of tooth-colored composite restorations via optical analysis - hyperspectral imaging (400 nm–1000 nm) and near-infrared spectroscopy (1550 nm–1950 nm)
- 25 July 2025
- Published by Elsevier in Forensic Science International
- Vol. 376, 112583
Open AccessDetection of tooth-colored composite restorations via optical analysis - hyperspectral imaging (400 nm–1000 nm) and near-infrared spectroscopy (1550 nm–1950 nm)
- Open Access
- 15 November 2023
- Published by Springer Nature in Rechtsmedizin
- Vol. 34 (1) , 52-55
Selbst beigebrachte Schädeltrepanationen
Zusammenfassung: Ein 78 Jahre alt gewordener Mann hatte sich 8 Bohrlochverletzungen am Hirnschädel bei später diagnostiziertem Aneurysma cerebri beigebracht. Als wahrscheinlichstes Motiv für die außergewöhnliche Tat ist ein Vernichtungskopfschmerz anzunehmen. Nach 5‑tägiger Überlebenszeit trat der Tod im Multiorganversagen mit akutem Myokardinfarkt unter Therapiebegrenzung ein. Eine Fremdeinwirkung konnte ausgeschlossen werden.
- 15 November 2023
- Published by Springer Nature in Rechtsmedizin
- Vol. 34 (1) , 52-55
Open AccessSelbst beigebrachte Schädeltrepanationen
Zusammenfassung: Ein 78 Jahre alt gewordener Mann hatte sich 8 Bohrlochverletzungen am Hirnschädel bei später diagnostiziertem Aneurysma cerebri beigebracht. Als wahrscheinlichstes Motiv für die außergewöhnliche Tat ist ein Vernichtungskopfschmerz anzunehmen. Nach 5‑tägiger Überlebenszeit trat der Tod im Multiorganversagen mit akutem Myokardinfarkt unter Therapiebegrenzung ein. Eine Fremdeinwirkung konnte ausgeschlossen werden.
- Open Access
- 18 September 2023
- Published by Springer Nature in Forensic Science, Medicine and Pathology
- Vol. 20 (2) , 490-499
Discrimination of human and animal bloodstains using hyperspectral imaging
Blood is the most encountered type of biological evidence in violent crimes and contains pertinent information to a forensic investigation. The false presumption that blood encountered at a crime scene is human may not be realised until after costly and sample-consuming tests are performed. To address the question of blood origin, the novel application of visible-near infrared hyperspectral imaging (HSI) is used for the detection and discrimination of human and animal bloodstains. The HSI system used is a portable, non-contact, non-destructive method for the determination of blood origin. A support vector machine (SVM) binary classifier was trained for the discrimination of bloodstains of human (n = 20) and five animal species: pig (n = 20), mouse (n = 16), rat (n = 5), rabbit (n = 5), and cow (n = 20). On an independent test set, the SVM model achieved accuracy, precision, sensitivity, and specificity values of 96, 97, 95, and 96%, respectively. Segmented images of bloodstains aged over a period of two months were produced, allowing for the clear visualisation of the discrimination of human and animal bloodstains. The inclusion of such a system in a forensic investigation workflow not only removes ambiguity surrounding blood origin, but can potentially be used in tandem with HSI bloodstain age determination methods for rapid on-scene forensic analysis.
- 18 September 2023
- Published by Springer Nature in Forensic Science, Medicine and Pathology
- Vol. 20 (2) , 490-499
Open AccessDiscrimination of human and animal bloodstains using hyperspectral imaging
Blood is the most encountered type of biological evidence in violent crimes and contains pertinent information to a forensic investigation. The false presumption that blood encountered at a crime scene is human may not be realised until after costly and sample-consuming tests are performed. To address the question of blood origin, the novel application of visible-near infrared hyperspectral imaging (HSI) is used for the detection and discrimination of human and animal bloodstains. The HSI system used is a portable, non-contact, non-destructive method for the determination of blood origin. A support vector machine (SVM) binary classifier was trained for the discrimination of bloodstains of human (n = 20) and five animal species: pig (n = 20), mouse (n = 16), rat (n = 5), rabbit (n = 5), and cow (n = 20). On an independent test set, the SVM model achieved accuracy, precision, sensitivity, and specificity values of 96, 97, 95, and 96%, respectively. Segmented images of bloodstains aged over a period of two months were produced, allowing for the clear visualisation of the discrimination of human and animal bloodstains. The inclusion of such a system in a forensic investigation workflow not only removes ambiguity surrounding blood origin, but can potentially be used in tandem with HSI bloodstain age determination methods for rapid on-scene forensic analysis.
- Open Access
- 7 September 2020
- Published by Springer Nature in Rechtsmedizin
- Vol. 31 (1) , 62-66
Jäger tot. Rehbock tot
Zusammenfassung: Ein erfahrener Jäger kehrte nach einem Abendansitz auf Rehwild nicht nach Hause zurück. Er wurde am folgenden Morgen von seinem Sohn auf einem Feld, in einer Blutlache liegend, tot aufgefunden. Neben ihm lag ein toter Rehbock. Der Jäger wies eine stark blutende Bauchverletzung auf, welche die Polizei zur Spekulation veranlasste, der Rehbock hätte diese Verletzung mit seinem Geweih verursacht. Zudem lag zwischen dem Jäger und dem Kadaver ein Gewehr mit zerbrochenem Schaft. Im vorgestellten Fall wurden mehrere sicherheitsrelevante Aspekte im Umgang mit Schusswaffen und Grundregeln der Jagd ignoriert, was zu einem tödlichen Jagdunfall führte.
- 7 September 2020
- Published by Springer Nature in Rechtsmedizin
- Vol. 31 (1) , 62-66
Open AccessJäger tot. Rehbock tot
Zusammenfassung: Ein erfahrener Jäger kehrte nach einem Abendansitz auf Rehwild nicht nach Hause zurück. Er wurde am folgenden Morgen von seinem Sohn auf einem Feld, in einer Blutlache liegend, tot aufgefunden. Neben ihm lag ein toter Rehbock. Der Jäger wies eine stark blutende Bauchverletzung auf, welche die Polizei zur Spekulation veranlasste, der Rehbock hätte diese Verletzung mit seinem Geweih verursacht. Zudem lag zwischen dem Jäger und dem Kadaver ein Gewehr mit zerbrochenem Schaft. Im vorgestellten Fall wurden mehrere sicherheitsrelevante Aspekte im Umgang mit Schusswaffen und Grundregeln der Jagd ignoriert, was zu einem tödlichen Jagdunfall führte.