Plattenepithelkarzinome der hinteren Rachenwand gehen nach aktuellen Literaturdaten mit schlechten onkologischen Ergebnissen einher. Costantino et al. haben daher ein neues Behandlungsprotokoll auf Basis einer neoadjuvanten Chemotherapie (NCT) und transoralen Roboterchirurgie (TORS) entwickelt und kommen in ihrer vorlaufigen Auswertung von 20 Fallen zu einem vielversprechenden Ergebnis.

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%.

We propose an efficient hyperspectral imaging (HSI) Deep Learning system for laryngeal cancer prediction. An in-vivo data set with 13 hyperspectral (HS) cubes of malignant laryngeal tumors has been collected directly from 13 different patients at Klinikum Braunschweig using a HS camera. In such medical applications, the data set is usually very sparse, limited, complex and noisy. Therefore, an efficient HSI Deep Learning system is highly needed. First, a relevant wavelength analysis has been done in order to detect the most informative channels in the HS cubes to speed up the prediction and reduce the noise. Based on the results, a new UNet, called Efficient Exception UNet (EFX-UNet), is devised and the two channels No. 13 and No. 14 from each cube are used for training and prediction. The EFX-UNet predicts the probability of the patient having a malignant tumor and yields an initial segmentation. After that a deeper analysis is carried on for the detected patients using the 8 channels No. 9 to No. 16 with our modified Deep UNet to precisely segment the tumor. The system is tested on real patient data. With EFXUNet we could reduce the prediction time from 60 s with Deep UNet to only 5 s. Although the data set is very limited and very complex, we could achieve a dice score test of 68.6% with EFX-UNet for initial segmentation and 88% with the Deep UNet in our system.

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)".

ê Die vorliegende Studie ist statistisch sauber aufgebaut, um unabhängige Variablen zu identifizieren, und ihre Schlussfolgerungen sind auf den ersten Blick stimmig. Bei genauerer Betrachtung zeigen sich jedoch meines Erachtens gravierende Einschränkungen, die mir diese Arbeit als geradezu gefährlich erscheinen lassen, da die Autoren recht absolut formulierte und weitreichende Folgerungen ziehen.

Vor allem in den Facharztpraxen und der Notaufnahme von HNO-Kliniken ist die traumatische Trommelfellperforation ein häufiges Krankheitsbild. Ein einheitliches Vorgehen (aktive Intervention vs. beobachtendes Abwarten) gibt es jedoch noch nicht. In der folgenden Studie soll vor allem der spontane Heilungsverlauf analysiert und dargestellt werden. Untersucht wurde ein Kollektiv aus 272 Patienten im Zeitraum von 2002 bis 2019, die sich mit einer traumatischen Trommelfellperforation in zwei HNO-ärztlichen Facharztpraxen vorstellten. Die Therapie erfolgte in allen Fällen konservativ, der Verlauf wurde prospektiv beobachtet. Die Auswertung des Patientenkollektivs erfolgte retrospektiv anhand der Patientenakten. Von den 272 Patienten waren 185 männlich und 87 weiblich. Das Durchschnittsalter betrug 30 Jahre. Die Perforationen lagen am häufigsten im vorderen oberen und hinteren unteren Quadranten. Gemäß der Gradeinteilung nach Griffin entsprach die Perforationsgröße am häufigsten Grad I. Die drei häufigsten Ursachen, die zur traumatischen Trommelfellperforation führten, waren der Schlag auf das Ohr, das Barotrauma und Fremdkörper. Unter dem abwartend-konservativen Regime konnte bei der Patientengruppe, die zu Nachkontrollen erschien, bei 95 % eine komplette Heilung dokumentiert werden. Ein abwartend-konservatives Regime kann bei traumatischen Trommelfellperforation als angemessen beurteilt werden, sofern eine HNO-fachärztliche Nachkontrolle gesichert ist. Eine Ausnahme bildet das Explosionstrauma, das jedoch gegenwärtig in Mitteleuropa seltener auftritt als früher, bei dem es zur Entwicklung sekundärer Cholesteatome kommen kann und somit eine aktive Therapie mit operativer Exploration des Mittelohrs samt Unterfütterung der Perforation indiziert ist.

Zusammenfassung Hintergrund Das Schlucken ist einer der komplexesten Bewegungsabläufe. Die Integrität der epithelialen Auskleidung ist dabei essenziell. Fragestellung Welche chirurgischen Ansätze wurden in den Anfängen der modernen Chirurgie entwickelt und welche stehen uns heute zur Verfügung, um die Integrität der Schluckstraße zu erhalten bzw. wiederherzustellen? Material und Methoden Selektive Literaturrecherche durch Studium der früheren Operationslehren sowie Online-Archive und Rückgriff auf Fälle aus eigener Praxis. Ergebnisse Bis in die 1950er-Jahre standen nur sehr unbefriedigende Techniken zur Verfügung, um die epitheliale Auskleidung wiederherzustellen. Seitdem jedoch kann die rekonstruktive Chirurgie insbesondere mittels mikrovaskulär reanastomosierter Transplantate sowohl im Rahmen der primären Chirurgie als auch bei sekundären Eingriffen die Integrität der Schluckstraße wiederherstellen. Schlussfolgerungen Das Erfordernis einer anatomischen und damit funktionellen Rehabilitation mittels Chirurgie ist so aktuell wie immer. Insbesondere im onkologischen Kontext könnte mittels verbesserter Früherkennung und lokalen Therapiemodalitäten die therapieassoziierte Schädigung des Schluckakts minimiert werden.

Background Traumatic eardrum perforation is a common presentation in otorhinolaryngologic practices and emergency clinics. A consistent management strategy (active intervention vs. watchful waiting) is, however, still lacking.Objective In the following study, the outcome of watchful waiting is analyzed and presented.Materials and methods A collective of 272 patients presenting at two different specialist ENT practices within days of traumatic tympanic membrane perforation from June 2002 to March 2019 were analyzed. Treatment was non-surgical, with prospective monitoring. Whereas antibiotics were not given at all in one practice, they were given only upon signs of infection in the other practice. The outcome was evaluated retrospectively on the basis of patient files.Results The collective consisted of 185 males and 87 females. Mean age was 30 years (range: 7 months to 82 years). The perforations were most commonly located in the upper anterior and lower posterior quadrants. According to Griffin grading, the size was grade I in 97%. The three most common causes were impact to the ear, barotrauma, and foreign bodies. Under a watchful waiting regimen, 95% of the patients presenting for follow-up checks showed complete closure.Conclusion Watchful waiting can be assessed as appropriate in traumatic eardrum perforation, provided otorhinolaryngologic follow-up is ensured. An exception is blast injury, which is now much less common in Central Europe, as this is associated with a risk of secondary cholesteatomas. In these rare cases, active treatment with surgical exploration of the middle ear including relining the perforation is indicated.

Background. Swallowing is one of the most complex movement patterns. The integrity of the epithelial lining is essential. Objective. Which surgical techniques were developed at the beginnings of modern surgery and what methods are now available to maintain/reconstitute the integrity of the swallowing apparatus? Materials and methods. This study comprises a selective literature search in early operation manuals and online archives, with incorporation of the authors' own experience. Results. Up until the 1950s, only very limited techniques were available to reconstruct the epithelial lining. Microvascular reanastomosed grafts were the game changer for reconstructive surgery, enabling reconstitution of the swallowing apparatus in primary surgery but also in challenging secondary interventions after insufficient or complicated primary therapy. Conclusion. The need for anatomical and functional rehabilitation by reconstructive surgery is as pertinent as ever. Particularly in the oncological context, improved early detection and novel local treatment modalities could minimize treatment-associated damage to swallowing.

Early detection of cancerous tissue is crucial for long-term patient survival. In the head and neck region, a typical diagnostic procedure is an endoscopic intervention where a medical expert manually assesses tissue using RGB camera images. While healthy and tumor regions are generally easier to distinguish, differentiating benign and malignant tumors is very challenging. This requires an invasive biopsy, followed by histological evaluation for diagnosis. Also, during tumor resection, tumor margins need to be verified by histological analysis. To avoid unnecessary tissue resection, a non-invasive, image-based diagnostic tool would be very valuable. Recently, hyperspectral imaging paired with deep learning has been proposed for this task, demonstrating promising results on ex-vivo specimens. In this work, we demonstrate the feasibility of in-vivo tumor type classification using hyperspectral imaging and deep learning. We analyze the value of using multiple hyperspectral bands compared to conventional RGB images and we study several machine learning models' ability to make use of the additional spectral information. Based on our insights, we address spectral and spatial processing using recurrent-convolutional models for effective spectral aggregating and spatial feature learning. Our best model achieves an AUC of 76.3%, significantly outperforming previous conventional and deep learning methods.

Early detection of head and neck tumors is crucial for patient survival. Often, diagnoses are made based on endoscopic examination of the larynx followed by biopsy and histological analysis, leading to a high interobserver variability due to subjective assessment. In this regard, early non-invasive diagnostics independent of the clinician would be a valuable tool. A recent study has shown that hyperspectral imaging (HSI) can be used for non-invasive detection of head and neck tumors, as precancerous or cancerous lesions show specific spectral signatures that distinguish them from healthy tissue. However, HSI data processing is challenging due to high spectral variations, various image interferences, and the high dimensionality of the data. Therefore, performance of automatic HSI analysis has been limited and so far, mostly ex-vivo studies have been presented with deep learning. In this work, we analyze deep learning techniques for in-vivo hyperspectral laryngeal cancer detection. For this purpose we design and evaluate convolutional neural networks (CNNs) with 2D spatial or 3D spatiospectral convolutions combined with a state-of-the-art Densenet architecture. For evaluation, we use an in-vivo data set with HSI of the oral cavity or oropharynx. Overall, we present multiple deep learning techniques for in-vivo laryngeal cancer detection based on HSI and we show that jointly learning from the spatial and spectral domain improves classification accuracy notably. Our 3D spatio-spectral Densenet achieves an average accuracy of 81%.

Here we present a study where we used in vivo hyperspectral imaging (HSI) for the detection of upper aerodigestive tract (UADT) cancer. Hyperspectral datasets were recorded in 100 patients before surgery in vivo. We established an automated data interpretation pathway that can classify the tissue into healthy and tumorous using, different deep learning techniques. Our method is based on convolutional neural networks (CNNs) with 2D spatial or 3D spatio-spectral convolutions combined with a state-of-the-art Densenet architecture. Using both the spatial and spectral domain improves classification accuracy notably. Our 3D spatio-spectral Densenet classification method achieves an average accuracy of over 80%.

Faramarzi M et al. Comparing Cold/Liquid Diet vs Regular Diet on Posttonsillectomy Pain and Bleeding. Otolaryngol Head Neck Surg 2018; 159: 755–760 Die Tonsillektomie ist eine häufige Operation. Hinsichtlich Nahrungsmittelauswahl und Art der Ernährung nach der Tonsillektomie gehen die Meinungen aber auseinander. Ärzte des HNO-Forschungszentrums in Schiras im Iran verglichen die Auswirkungen einer kalt/flüssigen Diät und einer normalen Ernährung auf Schmerzen und Blutungen im Bereich der Tonsillektomie.

BACKGROUND:The immune checkpoint, indoleamine 2,3-dioxygenase 1, is under investigation as target of novel immunotherapies for cancers, including head and neck squamous cell carcinomas (HNSCC). The aim of our study was to analyze DNA methylation of the encoding gene (IDO1) in HNSCC. METHODS:Methylation of three CpG sites within the promoter, promoter flank, and gene body was investigated and correlated with mRNA expression, immune cell infiltration, mutational burden, human papillomavirus (HPV)-status, and overall survival in a cohort of N = 528 HNSCC patients obtained from The Cancer Genome Atlas. In addition, IDO1 immunohistochemistry and DNA methylation analysis was performed in an independent cohort of N = 138 HNSCC samples. FINDINGS:Significant inverse correlations of IDO1 methylation and IDO1 mRNA expression were found in the promoter and promoter flank region (Spearman's ρ = -0.163 and ρ = -0.377, respectively) while a positive correlation was present in the gene body (ρ = 0.502; all P < 0.001). IDO1 DNA methylation significantly correlated with IDO1 protein expressing immune cells as well as tumor cells. IDO1 promoter flank hypermethylation was significantly associated with poor overall survival (P < 0.001). In addition, we discovered significant correlations between IDO1 methylation and expression with RNA signatures of immune cell infiltrates and with HPV-status, mutational load (methylation only), and interferon γ signature. INTERPRETATION:Our results suggest IDO1 expression levels are epigenetically regulated by DNA methylation. This study provides rationale to test IDO1 methylation as potential biomarker for prediction of response to IDO1 immune checkpoint inhibitors in HNSCC.

Hyperspectral imaging (HSI) is a technology with high potential in the field of non-invasive detection of cancer. However, in complex imaging situations like HSI of the larynx with a rigid endoscope, various image interferences can disable a proper classification of cancerous tissue. We identified three main problems: i) misregistration of single images in a HS cube due to patient heartbeat ii) image noise and iii) specular reflections (SR). Consequently, an image pre-processor is developed in the current paper to overcome these image interferences. It encompasses i) image registration ii) noise removal by minimum noise fraction (MNF) transformation and iii) a novel SR detection method. The results reveal that the pre-processor improves classification performance, while the newly developed SR detection method outperforms global thresholding technique hitherto used by 46%. The novel pre-processor will be used for future studies towards the development of an operational scheme for HS-based larynx cancer detection. RGB image of the larynx derived from the hyperspectral cube and corresponding specular reflections (a) manually segmented and (b) detected by a novel specular reflection detection method.

Früherkennung von kanzerösen Läsionen ist nach wie vor entscheidend für die Prognose des Patienten. Dabei stellt sich der diagnostische Zugang zu Mundhöhle und Oropharynx vergleichsweise unproblematisch dar. Dennoch ist die Inzidenz unverändert hoch, da die Läsionen rein visuell in vielen Fällen zu spät als maligne erkannt werden. Das hyperspektrale Imaging wurde bereits im Larynx als mögliche Untersuchungsmethode zur Früherkennung präkanzeröser und kanzeröser Veränderungen beschrieben. Hier evaluieren wir die Methode für Mundhöhle und Oropharynx.