Surface Based Medical Image Analysis

skull Traditionally medical image analysis has tackled problems in 2D images coming from for example X-ray machines or 3D images from computed tomography or magnetic resonance machines. Many algorithms working on volumetric data have been developed. An example is image registration where pre- and post-operative scannings are aligned and compared for the purpose of evaluating surgery outcome. Recently, fast and accurate range scanners have been introduced in the planning and evaluation of for example craniofacial surgery. A typical range-scanner is based on structured light and a stereo-camera setup and the acquisition time is therefore less than a second and no harmful radiation is involved. The natural output of range scanners are 3D surfaces. Currently, my research is based on surface based medical image analysis.

An introduction (in Danish) to some of my research can be seen on Danskernes Akademi.

Statistical Shape Analysis of the Human Ear Canal

ears Hearing aids come in a number of different styles. The smallest of these styles is called CIC (completely in the canal) and it has a number of attractive properties. First of all the small size is cosmetically appealing because a well-produced CIC is as good as invisible in-situ. Secondly, the CIC has some acoustic advantages. The shell of a CIC must have size and shape that allows it to contain a microphone, an amplifier, a loudspeaker, and battery. On the other hand it is desirable to design these components in a way that makes it possible to minimise the size of the final hearing aid.

A statistical shape model of the human ear canal was computed using a set of scanned ear impressions. Eventhough, the training set was rather small it was demonstrated that an anatomically plausible model could be generated and that the major variation in the data could be explained by less than seven modes of variation. The shape model of the ear has later been used as input in the design of in-the-ear devices. It provides information about the maximal variations of the shape of the ear in a standard population.

Statistical Shape and Appearance Analysis

In general, we are doing research within the statistics of shape and appearance and the associated methods. An example is modelling of the human face for early diagnostics of genetically associated disaeases or syndromes. Another example is the shape modelling of the human cochlear for surgical planning as part of the large EU project HEAR-EU. We have also investigated the connection between eye color and structure and the human genetics.

An open database of 3D scans of the human head, ear, and torso

Rasmus 3D head The aim of this project is to generate a database of high-resolution 3D scans of the head and torso of humans. The data will be presented on a web-portal where software to view, manipulate and process the data is also available. The intended users of the data and tools are students and researches working with acoustical modelling. Specifically, the data can for example be used to optimise spatial perception in hearing aids, boom design for headsets, and simulating individual head related transfer functions.

Mathematical modelling of the sound field surrounding the head is an emerging discipline that has shown promise to alleviate some of the difficulties in for example designing and testing the performance of new hearing aid designs. However, current state-of-the arts methods are mostly based on synthetic data and the results are therefore somewhat misleading. The lack of data is mainly due to the difficulty in acquiring real 3D data of the human head and torso. Especially, the 3D geometry of the human ear is very difficult to obtain using traditional 3D acquisition techniques like CT, MR, and laser scanning. Recently, the 3D Laboratory at the school of dentistry at the University of Copenhagen obtained a 3dMD cranial scanner by a donation from the Oticon Foundation. This scanner can be used to capture high quality 3D scans of the head, ear (pinna and part of concha), and torso of humans. The aim of this project is to use the scanner at the 3D laboratory to capture the torso and head geometry of a group of test persons. Furthermore, ear impressions should be taken and scanned so the final and merged data is a precise 3D presentation of torso, head, and the ear canal.

Some possible applications:

  • Simulating the influence of microphone placement in hearing devices
  • Boom design for headsets
  • Simulating head related transfer functions
  • Comparing simulated and measured HRTFs
  • Statistical analysis of the relation between torso+head+ear shape and HRTFs
  • Wind noise simulation for hearing devices
  • Optimal placement of electromagnetic coils in hearing devices
  • Visualization of hearing aid designs for development and for Marketing

The data is published in the OpenHear database.

Publications

The full and updated list of publications (the list below is not updated).

Ph.D. Thesis

  • R. R. Paulsen, Statistical Shape Analysis of the Human Ear Canal with Application to In-the-Ear Hearing Aid Design, Informatics and Mathematical Modelling, Technical University of Denmark, Building 321, Richard Petersens Plads, DK-2800 Kgs. Lyngby, 2004 [full] [bibtex][pdf]

Peer Reviewed Journal Papers

Peer Reviewed Conference Papers

  • J. Wilm, O. V. Olesen, R. R. Paulsen, L. Højgaard, B. Roed, R. Larsen. Real Time Surface Registration for PET Motion Tracking. SCIA 2011.
  • J. Fagertun, D. D. Gomez, M. F. Hansen1, R. R. Paulsen. Sparse Similarity-based Fisherfaces. SCIA 2011.
  • R. R. Paulsen and R. Larsen. Anatomically Plausible Surface Alignment and Reconstruction. Proc. Eurographics UK conference on Theory and Practice of Computer Graphics. pp. 249-254, 2010. (Best paper prize). Software page.
  • O. V. Olesen, R. R. Paulsen, L. Højgaard, B. Roed, and R. Larsen. Motion tracking in narrow spaces: a structured light approach. MICCAI 2010.
  • M. Durey, R. R. Paulsen, M. Giulliano, and T. Dabelsteen. Semi-automated tracking of behaviour of Betta splendens. Proc. The 8th French-Danish Workshop on Spatial Statistics and Image Analysis in Biology. 2010.
  • S. S. Thorup, H. Olafsdottir, T. A. Darvann, N. V. Hermann, P. Larsen, R. R. Paulsen, C. A. Perlyn, MK M. Gillian, S. Kreiborg, R. Larsen. Multivariate Analysis of Variance: Finding significant growth in mice with craniofacial dysmorphology caused by the Crouzon mutation. Proc. The 8th French-Danish Workshop on Spatial Statistics and Image Analysis in Biology. 2010.
  • S. S. Thorup, T. A. Darvann, N. V. Hermann, P. Larsen, H. Olafsdottir, R. R. Paulsen, A. A. Kane, D. Govier, L.-J. Lo, S. Kreiborg, R. Larsen. Dealing with difficult deformations: Construction of a knowledge-based deformation atlas. SPIE 2010
  • O. V. Olesen, R. R. Paulsen, L. Højgaard, B. Roed, R. Larsen. External motion tracking for brain imaging: structured light tracking with invisible light. Proc. IEEE Nuclear Science Symposium and Medical Imaging. 2010.
  • O. V. Olesen, L. Højgaard, M. R. Jørgensen, R. R. Paulsen, B. Roed, R. Larsen. Structured light 3D tracking system for measuring of movements in PET brain imaging. SPIE 2010
  • R. R. Jensen, R. R. Paulsen, and R. Larsen, Analysis of gait using a treadmill and a Time-of-flight camera. Workshop on Dynamic 3D Imaging. 2009.
  • C. Seiler, P. Büchler, N-P. Nolte, M. Reyes, R. R. Paulsen. Hierarchical Markov Random Fields Applied to Model Soft Tissue Deformations on Graphic Hardware. Workshop on 3D Physiological Human. P. 133-148, 2009, Springer Verlag
  • R. R. Jensen, R. R. Paulsen, and R. Larsen, Analyzing gait using a time-of-flight camera. SCIA 2009.
  • R. R. Paulsen, J. A. Bærentzen, and R. Larsen, Regularisation of 3D Signed Distance Fields. SCIA 2009.
  • S. S. Thorup, T. A. Darvann, N. V. Hermann, H. Olafsdottir, P. Larsen, R. Larsen, R. R. Paulsen, C. A. Perlyn, S. Kreiborg, Automatic assessment of craniofacial growth in a mouse model of Crouzon syndrom. ACPA 2008
  • S. Darkner, M. R. Sabuncu, P. Golland, R. R. Paulsen, R. Larsen, Analysis of Surfaces Using Constrained Regression Models. MICCAI 2008.
  • S. Darkner, M. V. Christensen, R. R. Paulsen, R. Larsen Non-rigid surface registration of 2D manifolds in 3D Euclidian space, SPIE -Medical Imaging 2008
  • S. Darkner, D. W. Hansen, R. R. Paulsen, R. Larsen Robust registration for change detection, SPIE -Medical Imaging 2008
  • S. Darkner, R. R. Paulsen, R. Larsen, Analysis of Deformation of the Human Ear and Canal Caused by Mandibular Movement, Medical Image Computing and Computer-Assisted Intervention 2007
  • S. Darkner, M. V. Christensen, R. Larsen, R. R. Paulsen, Evaluating a method for automated rigid registration, SPIE -Medical Imaging 2007
  • S. Darkner, M. Vester-Christensen, R. Larsen, C. Nielsen, R. R. Paulsen, Automated 3D Rigid Registration of Open 2D Manifolds, MICCAI 2006 Workshop "From Statistical Atlases to Personalized Models". [full] [pdf]
  • K. B. Hilger, R. R. Paulsen, R. Larsen, Markov Random Field Restoration of Point Correspondences for Active Shape Modelling, SPIE - Medical Imaging 2004 [full] [bibtex]
  • R. R. Paulsen, C. Nielsen, S. Laugesen, R. Larsen, Using a Shape Model in the Design of Hearing Aids, SPIE - Medical Imaging 2004 [full] [bibtex]
  • R. Larsen, K. B. Hilger, K. Skoglund, S. Darkner, R. R. Paulsen, M. B. Stegmann, B. Lading, H. Thodberg, H. Eiriksson, Some Issues of Biological Shape Modelling with Applications, 13th Scandinavian Conference on Image Analysis (SCIA), Gothenburg, Sweden, vol. 2749, pp. 509-519, Springer, 2003 [bibtex] [pdf]
  • R. R. Paulsen, K. Hilger, Shape Modelling Using Markov Random Field Restoration of Point Correspondences, Information Processing in Medical Imaging, 2003 [full] [bibtex][pdf]
  • R. R. Paulsen, R. Larsen, B. K. Ersbøll, C. Nielsen, S. Laugesen, Testing for Gender Related Size and Shape Differences of the Human Ear canal using Statistical methods, Eleventh International Workshop on Matrices and Statistics, Informatics and Mathematical Modelling, Technical University of Denmark, DTU, Richard Petersens Plads, Building 321, DK-2800 Kgs. Lyngby, 2002 [full][bibtex]
  • R. R. Paulsen, R. Larsen, S. Laugesen, C. Nielsen, B. K. Ersbøll, Building and Testing a Statistical Shape Model of the Human Ear Canal, Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, Tokyo, Japan, Springer, 2002 [full] [bibtex][pdf]

Master's Thesis

  • R. R. Paulsen, On the Estimation of Bone Status,Informatics and Mathematical Modelling, Technical University of Denmark, Building 321, Richard Petersens Plads, DK-2800 Kgs. Lyngby, Denmark, 1998 [full] [bibtex][pdf]

Manuals

  • R. R. Paulsen, Installing VTK 4.1.x on a Windows PC, Informatics and Mathematical Modelling, Technical University of Denmark, 2002 [full][bibtex] [pdf]