DE | EN
Startseite
Über uns
Übersicht
Zahlen und Fakten
Organisation
WissenschaftlerInnen
Kontakt
Anfahrt
Stellenangebote
Forschung
Übersicht
Anwendungsfelder
Projekte
Publikationen
WissenschaftlerInnen
Preprints
Institutionelle Kooperation
Archiv 02-14
Transfer
Übersicht
Branchen
Referenzen
MODAL-AG
Spin Offs
Software
Patente
Schule
Übersicht
MathInside
MATHEATHLON
Matheon-Kalender
What'sMath
Lehrerfortbildung
Sommerschulen
Termine
Presse
Übersicht
Pressemitteilungen
Neuigkeiten
Übersicht
Matheon Köpfe
Zahl der Woche
Neuigkeiten 2002-2014
Veranstaltungen
Übersicht
Workshops
15 Jahre Matheon
Mediathek
Übersicht
Fotos
Videos
Audios
Broschüren
Bücher
Aufgelesen

Since 2019, Matheon's application-oriented mathematical research activities are being continued in the framework of the Cluster of Excellence MATH+
www.mathplus.de
The Matheon websites will not be updated anymore.

Dr. Hans Lamecker

Projektleiter

Zuse Institut Berlin
Takustr. 7
14195 Berlin
+49 (0) 30 +49 30 841 85 177
lamecker@zib.de
Webseite


Forschungsschwerpunkte

Bild- und Geometrieverarbeitung, statistische Formmodellierung, medizinische Therapieplanung

Projekte als Projektleiter

  • GV-AP17

    Machine Learning Approaches for Enhanced, Shape Model Based 3D Image Segmentation

    Dr. Hans Lamecker / Dr.-Ing. Stefan Zachow

    Projektleiter: Dr. Hans Lamecker / Dr.-Ing. Stefan Zachow
    Projekt Mitglieder: Dr. Anirban Mukhopadhyay
    Laufzeit: 01.10.2014 - 30.09.2019
    Status: beendet
    Standort: Konrad-Zuse-Zentrum für Informationstechnik Berlin

    Beschreibung

    Fully automatic segmentation of arbitrary anatomical structures from 3D medical image data is a challenging, yet unsolved problem. Though fully automatic segmentation is essential for further clinical analysis, complexity of anatomical structures across population makes a generalized segmentation scheme extremely challenging. Moreover, specific challenges of different imaging modalities have so far hindered the possibility of a general purpose fully automatic 3D segmentation framework. Statistical 3D shape models have proven to be valuable shape priors that are to be deformed within their range of normal variation in shape to match the respective image information. Within the project, we are aiming to combine Machine Learning along with the statistical shape priors for getting a step closer to a general 3D image segmentation approach. In particular, Machine Learning techniques for image matching based on intensity will be developed in order to improve both the model building as well as the segmentation process. Image-based Cost Functions: Principal Component Analysis (PCA) on local intensity profiles has not proven to beneficially act as a robust cost function. Random Forest Regression Voting (RFRV), though a powerful method for 2D image data, turned out to be impractical for 3D data, due to huge memory consumption and computational time. Dictionary Learning (DL) does not require any heuristics and is general enough to be applied across anatomies and modalities. DL operations are matrix operations, thus being efficiently evaluated. Joint Dictionary Learning: Given 3D image data and accordingly segmented anatomical structures of interest, rotational invariant histograms of oriented gradients (HoG) are sampled at the structures’ boundaries. These feature samples are used as input for learning a dictionary. A second dictionary is learnt for background image information. A combined dictionary of foreground and background features has been established, acting as a cost function for image segmentation. Cost Function for a test patch: Sum of residuals from representations by the two dictionaries.

    http://www.zib.de/projects/machine-learning-approaches-enhanced-shape-model-based-3d-image-segmentation