Endoskopie des Larynx mit Hyperspectral Imaging

Gerstner, Andreas O. H.; Laffers, Wiebke; Schade, G.; Göke, Friederike; Martin, R.; Thies, Boris

doi:10.1007/s00106-012-2632-9

Cited by 7 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hyperspectral imaging (HSI) has its origin in remote sensing and has been used in numerous nonmedical areas 1–3 . In recent years, however, its application has been extended to medicine, where it offers great potential for noninvasive disease diagnosis and surgical guidance 4–25 . This contact‐less and noninvasive tool uses light from the visible and near‐infrared spectrum for tissue illumination and the camera detects the diffuse remitted light.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of hyperspectral imaging to quantify perfusion changes during the modified Allen test

Linek¹,

Felicio‐Briegel

Freymüller³

et al. 2021

Lasers Surg Med

View full text Add to dashboard Cite

Objectives To evaluate the capability of hyperspectral imaging (HSI), a contact‐less and noninvasive technology, to monitor perfusion changes of the hand during a modified Allen test (MAT) and cuff occlusion test. Furthermore, the study aimed at obtaining objective perfusion parameters of the hand. Methods HSI of the hand was performed on 20 healthy volunteers with a commercially available HSI system during a MAT and a cuff occlusion test. Besides gathering red‐green‐blue (RGB) images, the perfusion parameters tissue hemoglobin index (THI), (superficial tissue) hemoglobin oxygenation (StO2), near‐infrared perfusion (NIR), and tissue water index (TWI) were calculated for four different regions of interest on the hand. For the MAT, occlusion (OI; the ratio between the condition during occlusion and before occlusion) and reperfusion (RI; the ratio between the non‐occlusion state and the prior occlusion state) indices were calculated for each perfusion parameter. All data were correlated to the clinical findings. Results False‐color images showed visible differences between the various perfusion conditions during the MAT and cuff occlusion test. THI, StO2, and NIR behaved as expected from physiology, while TWI did not in the context of this study. During rest, mean THI, StO2, and NIR of the hand were 34 ± 2, 72 ± 9, and 61 ± 6, respectively. The RI for THI showed a roundabout threefold increase after reperfusion of both radial and ulnar artery and was thus, distinctly pronounced when compared with StO2 and NIR (~1.25). The OI was lowest for THI when compared with StO2 and NIR. Conclusions HSI with its parameters THI, StO2, and NIR proved to be suitable to evaluate perfusion of the hand. By this, it could complement visual inspection during the MAT for evaluating the functionality of the superficial palmary arch before radial or ulnar artery harvest. The presented RI might deliver useful comparative values to detect pathological perfusion disorders at an early stage. As microcirculation monitoring is crucial for many medical issues, HSI shows potential to be used, besides further applications, in the monitoring of (free) flaps and transplants and microcirculation monitoring of critically ill patients.

show abstract

Section: Introductionmentioning

confidence: 99%

Evaluation of hyperspectral imaging to quantify perfusion changes during the modified Allen test

Linek¹,

Felicio‐Briegel

Freymüller³

et al. 2021

Lasers Surg Med

View full text Add to dashboard Cite

show abstract

“…The larynx is the site of the majority of head and neck malignancies and early detection of laryngeal tumors is critical for therapy [1]. In order to detect cancer early and from images, the long-wavelength range is recommended because the penetration depth into the tissue is higher [2]. Hyperspectral Imaging (HSI) potentially supports capturing this relevant spectral information due to extending spectral resolution into the visible-near infrared.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Wavelength Analysis with U-Net for Larynx Cancer Detection

Meyer-Veit¹,

Rayyes²,

Gerstner³

et al. 2022

ESANN 2022 Proceedings

View full text Add to dashboard Cite

Early detection of laryngeal tumors is critical for their successful therapy. In this paper, we investigate how hyperspectral (HS) imaging can contribute to this aim based on an in-vivo data set of 13 HS image cubes recorded in clinical practice. We perform semantic segmentation with a tailored U-Net trained on labels provided by the clinicians. We specifically investigate the influence of exposure time during image acquisition, the suitable wavelengths to determine the most informative image channels, and present quantitative results on accuracy and the AUC measure.* The data set collection was funded by the German Cancer Aid within the project framework "Early detection of laryngeal cancer by means of Hyperspectral Imaging (109825110275)".

show abstract

“…Hyperspectral imaging (HSI) is a noninvasive imaging technique for quantitative and objective surface analysis detecting reflecting light [10, 11]. HSI has been developed for satellites and geographical measurements and has been applied in medical research in recent years.…”

Section: Introductionmentioning

confidence: 99%

In vivo detection of head and neck tumors by hyperspectral imaging combined with deep learning methods

Eggert

Bengs

Westermann

et al. 2021

Journal of Biophotonics

View full text Add to dashboard Cite

Currently, there are no fast and accurate screening methods available for head and neck cancer, the eighth most common tumor entity. For this study, we used hyperspectral imaging, an imaging technique for quantitative and objective surface analysis, combined with deep learning methods for automated tissue classification. As part of a prospective clinical observational study, hyperspectral datasets of laryngeal, hypopharyngeal and oropharyngeal mucosa were recorded in 98 patients before surgery in vivo. We established an automated data interpretation pathway that can classify the tissue into healthy and tumorous using convolutional neural networks with 2D spatial or 3D spatio‐spectral convolutions combined with a state‐of‐the‐art Densenet architecture. Using 24 patients for testing, our 3D spatio‐spectral Densenet classification method achieves an average accuracy of 81%, a sensitivity of 83% and a specificity of 79%.

show abstract

Endoskopie des Larynx mit Hyperspectral Imaging

Cited by 7 publications

References 18 publications

Evaluation of hyperspectral imaging to quantify perfusion changes during the modified Allen test

Evaluation of hyperspectral imaging to quantify perfusion changes during the modified Allen test

Hyperspectral Wavelength Analysis with U-Net for Larynx Cancer Detection

In vivo detection of head and neck tumors by hyperspectral imaging combined with deep learning methods

Contact Info

Product

Resources

About