Search:
Lehrstuhl  |  Institut  |  Fakultät  |  LMU
print

Theseus

Period: 2008 - 2012
Funding: Federal Ministry of Economy and Technology (BMWi)
Project parts: MEDICO, CTC
Partners: official list of partners
Official Website: http://theseus-programm.de

Use Case MEDICO

The vision of the application scenario MEDICO is to establish syntax and semantics based on image vocabularies to build an intelligent scalable robust search engine for the medical domain, with applications in Clinical Decision Support and Computer Aided Diagnosis Systems.

A principal contributions to the MEDICO project are methods allowing fast image-based similarity queries. As image similarity is not clearly defined, we develop adaptive similarity measures for various kinds and combinations of image descriptors. One example for similarity queries on medical images is the localization of the current body height of a given single query image using k-nearest neighbor regression as in the image below.

Slice localization using k-nearest neighbor regression

An even more challenging application is the transfer of similarity queries to actual 3-dimensional query regions. Our system enables a physician to outline a region of interest in an image volume as query resulting in a list of similar scans for patients comparable in manually annotated and automatically induced diagnosis. (See also an illustrating video on the Theseus Medico YouTube Channel.)

The latest news on THESEUS MEDICO are posted on our twitter channel.

Core technology Cluster (CTC)

Some of the basic technologies being developed by the research partners are functions for the automatic creation of metadata for audio, video, 2D and 3D picture files and their application-specific combination, as well as mechanisms for the semantic processing of multimedia documents and their associated services.

The research partners are working on scalable technology for shaping and structuring ontologies. In the course of this work, they will utilise innovative processes based on machine learning principles; these will help them to discover logical-deterministic and probabilistic relationships in text, audio, image and video data. This, in turn, will allow statistical Machine Learning methods to be developed that will make the extraction of knowledge from relational data possible; this data can then be used to create or extend existing ontologies.

One of our last projects within the CTC was the creation of supplementary tools for the Use Case MEDICO by processing computer tomography (CT) scans in order to apply the possibility of similarity search upon image data. The image below illustrates the concept of a possible workflow.

Supervised Student Works

2012

  • Visualisierung von Metadaten als Video-Overlay in HTML5
    Ekaterina Selivanova (Projektarbeit in cooperation with Siemens)
  • Identifkation von Fitnessübungen aus Bewegungsdaten mittels Symbolic Aggregate approXimation
    Sebastian Steuer (Projektarbeit)

2011

  • Visualisierung und Zusammenfassung von 3D Ähnlichkeitsrankings auf CT Daten
    Manuel Klette (Bachelorarbeit)
  • CT Retrieval in Rasterdatenbanken
    Christoph Wagner (Bachelor Thesis)
  • 3D-Bilddeskriptoren für Ähnlichkeitssuche auf CT-Bildern
    Michael Stockerl (Bachelorarbeit)
  • Local Sensitive Hashing for image similarity search on a modern non-SQL data storage
    Maximilian Kwapil in cooperation with Mayflower GmbH (Diplomarbeit)
  • Vollautomatische Erkennung der Wirbelsäule in 2D CT-Bildern
    Robert Greil (Bachelorarbeit)
  • A validation framework for the efficient retrieval of CT subvolumes
    Bernhard Meyer (Bachelorarbeit)
  • Mustererkennung in Zeitreihen
    Alexander Stautner (Diplomarbeit)
  • Indexing hochdimensionaler Daten
    Christiane Gargitter (Diplomarbeit)
  • Klassifikation von Wirbelkörpern in CT-Bildern
    Michael Strukelj (Diplomarbeit)
  • Metadaten-basierte Video-Recherche im Surveillance Umfeld
    Adrian Fiedler (Bachelorarbeit in cooperation with Siemens)
  • Rule-based Search for Events in Annotated Video Sequences
    Daniel Buschek (Bachelorarbeit in cooperation with Siemens)
  • Data Mining auf Bewegungsdaten
    Christian Mönnig (Diplomarbeit)
  • Middleware-gestützte Suche in heterogenen (Video)-Datenbanken
    Wolfgang Ostermeier (Projektarbeit in cooperation with Siemens)

2010

  • Extraktion relevanter Bildregionen mit geringer Schärfentiefe
    Michael Weiler (Diplomarbeit)
  • Implementation and Evaluation of a Fuzzy Algorithm for the Segmentation of Regions of Interest in Low Depth of Field Images
    Philipp Grosselfinger (Bachelorarbeit)
  • Implementation and Evaluation of an Unsupervised Videosegmentation Algorithm of Regions of Interest in Low Depth of Field Images
    Sirithana Tirnardvanich (Bachelorarbeit)
  • Effiziente Suche in hochdimensionalen Daten
    Christiane Gargitter (Projektarbeit)
  • Effiziente Suche in hochdimensionalen Daten
    Christian Mönnig (Projektarbeit)
  • Locality relations in patch-based image descriptors
    Florian Schlegel (Programming Assignment)
  • Segmentation of liver lesions in CT box annotations
    Michael Shekelyan (Programming Assignment)
  • An examination of tag visual-word co-occurrence in image repositories
    Simon Mang (Bachelorarbeit)
  • Automatische Detektion der Wirbelsäule auf 2D Bildern
    Michael Strukelj (Projektarbeit)
  • 2D Wirbelsäulendetektion durch interessante Punkte
    "Annika Tonch" (Bachelorarbeit)
  • Automatische Annotation von medizinischen Volumendaten
    Justin Heesemann (Diplomarbeit)
  • Positionsbestimmung in medizinischen Volumendatensätzen
    Aleksandar Havalyov (Diplomarbeit)
  • Entwicklung eines Frameworks zur Analyse und Darstellung von Beschleunigungssensordaten
    Alexander Stautner (Projektarbeit)

2009

  • Modellbasiertes Matching von lokalen Bilddeskriptoren
    Marcin Czyzewski (Diplomarbeit)
  • Evaluation von Gaussian Mixture Models und der X-Tree Datenstruktur
    Andreas Zitzelsberger (Diplomarbeit)
  • Image Retrieval mit Verteilungsfunktionen
    Stephan Kahle (Diplomarbeit)


Publications

2012
27A. Cavallaro, H.-P. Kriegel, M. Petri, M. Schubert
Semantic Localization driven Partial Image Retrieval in CT Series
Methods of Information in Medicine, 2012.
26T. Bernecker, F. Graf, H.-P. Kriegel, N. Seiler, C. Tuermer, D. Dill
Knowing: A Generic Data Analysis Application
In Proceedings of the 15th International Conference on Extending Database Technology (EDBT), Berlin, Germany, 2012.
25H.-P. Kriegel, M. Petri, M. Schubert, M. Shekelyan, M. Stockerl
Efficient Similarity Search on 3D Bounding Box Annotations
In Proceedings of the SPIE Medical Imaging Conference 2012: Advanced PACS-based Imaging Informatics and Therapeutic Applications (SPIE), San Diego, CA: 8319-10, 2012.
24F. Graf, R. Greil, H.-P. Kriegel, M. Schubert, A. Cavallaro
Enhanced Detection of the Vertebrae in 2D CT-Images
In Proceedings of the SPIE Medical Imaging Conference 2012: Image Processing (SPIE), San Diego, CA, 2012.
2011
23F. Graf, H.-P. Kriegel, M. Schubert, S. Poelsterl, A. Cavallaro
2D Image Registration in CT Images using Radial Image Descriptors
In Medical Image Computing and Computer-Assisted Intervention (MICCAI), Toronto, Canada: 607–614, 2011.
22A. Cavallaro, F. Graf, H.-P. Kriegel, M. Schubert, M. Thoma
Region of Interest Queries in CT Scans
In Proceedings of the 12th International Symposium on Spatial and Temporal Databases (SSTD), Minneapolis, MN: 56–73, 2011.
21F. Graf, H.-P. Kriegel, M. Renz, M. Schubert
MARiO: Multi Attribute Routing in Open Street Map
In Proceedings of the 12th International Symposium on Spatial and Temporal Databases (SSTD), Minneapolis, MN: 486–490, 2011.
20F. Graf, H.-P. Kriegel, S. Pölsterl, M. Schubert, A. Cavallaro
Position Prediction in CT Volume Scans
In Proceedings of the 28th International Conference on Machine Learning (ICML) Workshop on Learning for Global Challenges, Bellevue, Washington, WA, 2011.
19T. Bernecker, F. Graf, H.-P. Kriegel, C. Moennig, A. Zimek
BeyOND – Unleashing BOND
In Proceedings of the 37th International Conference on Very Large Data Bases (VLDB) Workshop on Ranking in Databases (DBRank), Seattle, WA: 34–39, 2011.
18F. Stegmaier, M. Döller, K. Schlegel, H. Kosch, S. Seifert, M. Kramer, A. Hutter, M. Thoma, H.-P. Kriegel, A. Cavallaro
Generische Datenintegration zur semantischen Diagnoseunterstützung im Projekt THESEUS MEDICO
In Proceedings of the 41st Jahrestagung der Gesellschaft für Informatik: Datenmanagement und Interoperabilität im Gesundheitswesen, Berlin, Germany: 41–55, 2011.
17F. Graf, H.-P. Kriegel, M. Weiler
Robust Segmentation of Relevant Regions in Low Depth of Field Images
In Proceedings of the IEEE International Conference on Image Processing (ICIP), Brussels, Belgium, 2011.
16S. Seifert, M. Thoma, F. Stegmaier, M. Hammon, M. Kramer, M. Huber, H.-P. Kriegel, A. Cavallaro, D. Comaniciu
Combined semantic and similarity search in medical image databases
In Proceedings of the SPIE Medical Imaging Conference 2011: Advanced PACS-based Imaging Informatics and Therapeutic Applications, Lake Buena Vista, FL, 7967: 7967–2, 2011.
15F. Graf, H.-P. Kriegel, M. Schubert, M. Strukelj, A. Cavallaro
Fully automatic detection of the vertebrae in 2D CT images
In Proceedings of the SPIE Medical Imaging Conference 2011: Image Processing (SPIE), Lake Buena Vista, FL, 7962: 79620C, 2011.
14M. Petri
Similarity Search Applications in Medical Images
PhD Thesis, Ludwig-Maximilians-Universität München, 2011.
13F. Graf, H.-P. Kriegel, M. Schubert
Maximum Gain Round Trips with Cost Constraints
Technical Report, Ludwig-Maximilians-Universität München, Munich, Germany, 2011.
2010
12M. Thoma, H. Cheng, A. Gretton, J. Han, H.-P. Kriegel, A. Smola, L. Song, P. S. Yu, X. Yan, K. M. Borgwardt
Discriminative Frequent Subgraph Mining with Optimality Guarantees
Statistical Analysis and Data Mining, 3(5): 302–318, 2010.
11F. Graf, H.-P. Kriegel, M. Renz, M. Schubert
Memory-Efficient A*-Search using Sparse Embeddings
In Proceedings of the 18th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM GIS), San Jose, CA: 462–465, 2010.
10T. Emrich, F. Graf, H.-P. Kriegel, M. Schubert, M. Thoma
Optimizing All-Nearest-Neighbor Queries with Trigonometric Pruning
In Proceedings of the 22nd International Conference on Scientific and Statistical Database Management (SSDBM), Heidelberg, Germany: 501–518, 2010.
9T. Emrich, F. Graf, H.-P. Kriegel, M. Schubert, M. Thoma
Similarity Estimation using Bayes Ensembles
In Proceedings of the 22nd International Conference on Scientific and Statistical Database Management (SSDBM), Heidelberg, Germany: 537–554, 2010.
8T. Bernecker, T. Emrich, F. Graf, H.-P. Kriegel, P. Kröger, M. Renz, E. Schubert, A. Zimek
Subspace Similarity Search: Efficient k-NN Queries in Arbitrary Subspaces
In Proceedings of the 22nd International Conference on Scientific and Statistical Database Management (SSDBM), Heidelberg, Germany: 555–564, 2010.
7T. Bernecker, T. Emrich, F. Graf, H.-P. Kriegel, P. Kröger, M. Renz, E. Schubert, A. Zimek
Subspace Similarity Search Using the Ideas of Ranking and Top-k Retrieval
In Proceedings of the 26th International Conference on Data Engineering (ICDE) Workshop on Ranking in Databases (DBRank), Long Beach, CA: 4–9, 2010.
6T. Emrich, F. Graf, H.-P. Kriegel, M. Schubert, M. Thoma
On the Impact of Flash SSDs on Spatial Indexing
In Proceedings of the 6th International Workshop on Data Management on New Hardware (DaMoN), Indianapolis, IN: 3–8, 2010.
5T. Emrich, H.-P. Kriegel, P. Kröger, M. Renz, A. Züfle
Boosting Spatial Pruning: On Optimal Pruning of MBRs
In Proceedings of the ACM International Conference on Management of Data (SIGMOD), Indianapolis, IN: 39–50, 2010.
4F. Graf, H.-P. Kriegel, M. Renz, M. Schubert
PAROS: Pareto Optimal Route Selection
In Proceedings of the ACM International Conference on Management of Data (SIGMOD), Indianapolis, IN: 1199–1202, 2010.
3T. Emrich, F. Graf, H.-P. Kriegel, M. Schubert, M. Thoma, A. Cavallaro
CT Slice Localization via Instance-Based Regression
In Proceedings of the SPIE Medical Imaging Conference 2010: Image Processing (SPIE), San Diego, CA, 7623: 762320, 2010.
2009
2M. Thoma, H. Cheng, A. Gretton, J. Han, H.-P. Kriegel, A. Smola, L. Song, P. S. Yu, X. Yan, K. M. Borgwardt
Near-optimal supervised feature selection among frequent subgraphs
In Proceedings of the 9th SIAM International Conference on Data Mining (SDM), Sparks, NV: 1075–1087, 2009.
2008
1H.-P. Kriegel, P. Kunath, A. Pryakhin, M. Schubert
MUSE: Multi-Represented Similarity Estimation
In Proceedings of the 24th International Conference on Data Engineering (ICDE), Cancun, Mexico: 1340–1342, 2008.

Team

Scientific Head: Prof. Dr. Hans-Peter Kriegel
Project Leader: PD Dr. Matthias Schubert
Current Members:
blank