• Publikationen
Show publication details

Wirtz, Andreas; Wambach, Johannes; Wesarg, Stefan

Automatic Teeth Segmentation in Cephalometric X-Ray Images Using a Coupled Shape Model

2018

OR 2.0 Context-Aware Operating Theaters, Computer Assisted Robotic Endoscopy, Clinical Image-Based Procedures, and Skin Image Analysis

International Workshop on OR 2.0 Context-Aware Operating Theaters (OR 2.0) <1, 2018, Granada, Spain>

Lecture Notes in Computer Science (LNCS), 11041

Cephalometric analysis is an important tool used by dentists for diagnosis and treatment of patients. Tools that could automate this time consuming task would be of great assistance. In order to provide the dentist with such tools, a robust and accurate identification of the necessary landmarks is required. However, poor image quality of lateral cephalograms like low contrast or noise as well as duplicate structures resulting from the way these images are acquired make this task difficult. In this paper, a fully automatic approach for teeth segmentation is presented that aims to support the identification of dental landmarks. A 2-D coupled shape model is used to capture the statistical knowledge about the teeth’s shape variation and spatial relation to enable a robust segmentation despite poor image quality. 14 individual teeth are segmented and labeled using gradient image features and the quality of the generated results is compared to manually created gold-standard segmentations. Experimental results on a set of 14 test images show promising results with a DICE overlap of 77.2% and precision and recall values of 82.3% and 75.4%, respectively.

978-3-030-01200-7

Show publication details

Wirtz, Andreas; Mirashi, Sudesh Ganapati; Wesarg, Stefan

Automatic Teeth Segmentation in Panoramic X-Ray Images Using a Coupled Shape Model in Combination with a Neural Network

2018

Medical Image Computing and Computer Assisted Intervention – MICCAI 2018: Part IV

International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) <21, 2018, Granada, Spain>

Lecture Notes in Computer Science (LNCS), 11073

Dental panoramic radiographs depict the full set of teeth in a single image and are used by dentists as a popular first tool for diagnosis. In order to provide the dentist with automatic diagnostic support, a robust and accurate segmentation of the individual teeth is required. However, poor image quality of panoramic x-ray images like low contrast or noise as well as teeth variations in between patients make this task difficult. In this paper, a fully automatic approach is presented that uses a coupled shape model in conjunction with a neural network to overcome these challenges. The network provides a preliminary segmentation of the teeth region which is used to initialize the coupled shape model in terms of position and scale. Then the 28 individual teeth (excluding wisdom teeth) are segmented and labeled using gradient image features in combination with the model’s statistical knowledge about their shape variation and spatial relation. The segmentation quality of the approach is assessed by comparing the generated results to manually created goldstandard segmentations of the individual teeth. Experimental results on a set of 14 test images show average precision and recall values of 0.790 and 0.827, respectively and a DICE overlap of 0.744.

978-3-030-00936-6

Show publication details

Mirashi, Sudesh Ganapati; Glesner, Manfred [1. Gutachten]; Wirtz, Andreas [Betreuer]

Model-based Segmentation of the Teeth in Panoramic Radiograph Images

2018

Darmstadt, TU, Master Thesis, 2018

In this thesis, a fully automatic approach for teeth segmentation in dental panoramic radiograhpic images is presented. The approach uses an exsiting coupled shape model framework, which was developed at Fraunhofer IGD, in conjunction with a convolutional neural network (CNN). The CNN provides a preliminary segmentation of the teeth region which is used to initialize the coupled shape model in terms of position and scale. Then, the 28 individual teeth (excluding wisdom teeth) are segmented and labeled using gradient image features in combination with the statistical knowledge about their shape variation and spatial relation. A combination of separate adaptation steps is used to ensure a robust segmentation. The segmentation quality of the approach is assessed by comparing the generated results to manually created gold-standard segmentations of the individual teeth.

Show publication details

Wambach, Johannes; Hergenröther, Elke [Referent]; Rapp, Stefan [Korreferent]; Wirtz, Andreas [Betreuer]

Vollautomatische Segmentierung der Zähne in Fernröntgenseitenbildern unter Verwendung statistischer Formmodelle

2018

Darmstadt, Hochschule, Master Thesis, 2018

Fernröntgenseitenbilder (FRS) sind in der Zahnmedizin und der Kieferorthopädie ein wichtiges Hilfsmittel für die Behandlungsplanung. Auf einem FRS wird der Kopf direkt von der Seite aufgenommen. Ein FRS wird unter anderem für die Analyse des Gesichtsschädelaufbaus, der Erfassung der vertikalen und sagittalen bzw. horizontalen Kieferlagebeziehungen und der dentalen Beziehungen verwendet. Für die Analyse eines FRS werden Punkte und Linien markiert und mit diesen Winkel und Strecken berechnet. Die manuelle Analyse von FRS-Aufnahmen ist sehr zeitaufwendig. Da es sich hierbei um die Erkennung von Bildmerkmalen auf 2D-Grauwertbildern handelt, ist es möglich die Analyse von FRS-Aufnahmen zu automatisieren. In dieser Arbeit wird ein System zur vollautomatischen Segmentierung der Zähne in Fernröntgenseitenbilder unter Verwendung statistischer Formmodelle vorgestellt. Aus manuell segmentierten FRS-Aufnahmen wird ein statistisches Modell mit artikuliertem Atlas trainiert. Es wird jeweils ein Modell für Oberkiefer und ein Modell für Unterkiefer gelernt. Es wird dann eine Initialisierung des mittleren Modells auf einem Testbild, anhand von erkannten Bildmerkmalen, durchgeführt. So wird die initiale Position, Rotation und Skalierung des mittleren Modells gefunden. Danach kann das mittlere Modell an ein Testbild angepasst werden. Die Anpassung verwendet die ermittelten Gradienten im Testbild und das gelernte Modellwissen. Die so erhaltenen Segmentierungen werden mit manuellen Segmentierungen verglichen. Der berechnete Mittelwert der Überlappung der automatischen und manuellen Segmentierungen des Ober- und Unterkiefers liegt bei knapp über 80 Prozent.