Affiliation: IT University of Copenhagen
- Segmentation of the posterior ribs in chest radiographs using iterated contextual pixel classificationMarco Loog
Image Sciences Institute, University Medical Center Utrecht, 3508 GA Utrecht, The Netherlands
IEEE Trans Med Imaging 25:602-11. 2006..In a sixfold cross-validation experiment, ICPC achieves a classification accuracy of 0.86 +/- 0.06, as compared to 0.94 +/- 0.02 for the second human observer...
- Filter learning: application to suppression of bony structures from chest radiographsM Loog
The Image Group, IT University of Copenhagen, Rued Langgaards Vej 7, 2300 Copenhagen S, Denmark
Med Image Anal 10:826-40. 2006....
- Image dissimilarity-based quantification of lung disease from CTLauge Sørensen
The Image Group, Department of Computer Science, University of Copenhagen, Denmark
Med Image Comput Comput Assist Interv 13:37-44. 2010..817. This is significantly better compared to combining individual region classifications into an overall image classification, and compared to common computerized quantitative measures in pulmonary CT...
- A computer-aided diagnosis system for detection of lung nodules in chest radiographs with an evaluation on a public databaseArnold M R Schilham
Image Sciences Institute, University Medical Center Utrecht, The Netherlands
Med Image Anal 10:247-58. 2006..For four false positives, this increases to 67%. This is close to the previously reported 70% detection rate of the radiologists...
- Linear dimensionality reduction via a heteroscedastic extension of LDA: the Chernoff criterionMarco Loog
Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands
IEEE Trans Pattern Anal Mach Intell 26:732-9. 2004..This criterion combines separation information present in the class mean as well as the class covariance matrices. Extensive experiments and a comparison with similar dimension reduction techniques are presented...
- Segmentation of anatomical structures in chest radiographs using supervised methods: a comparative study on a public databaseBram van Ginneken
Image Sciences Institute, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands
Med Image Anal 10:19-40. 2006..All results, including the manual segmentations, have been made publicly available to facilitate future comparative studies...
- On combining computer-aided detection systemsMeindert Niemeijer
University Medical Center Utrecht, Image Sciences Institute, 3584 CX Utrecht, The Netherlands
IEEE Trans Med Imaging 30:215-23. 2011..For both applications, combination results in a large and significant increase in performance when compared to the best individual CAD system...
- A family of principal component analyses for dealing with outliersJ Eugenio Iglesias
Department of Computer Science, University of Copenhagen, Denmark
Med Image Comput Comput Assist Interv 10:178-85. 2007..The results show that the algorithm can implement both an outlier-enhancing and a robust PCA. The former improves the segmentation performance in fractured vertebrae, while the latter does so in the unfractured ones...
- On distributional assumptions and whitened cosine similaritiesMarco Loog
Pattern Recognition Group, Faculty of Elctrical Engineering, Mathematics and Computer Science, Delft Unversity of Technology, Delf, The Netherlands
IEEE Trans Pattern Anal Mach Intell 30:1114-5. 2008..29, no. 6, pp. 1086-1090, June 2007. This communication makes the observation that some of the distributional assumptions made to derive this measure are very restrictive and, considered simultaneously, even inconsistent...
- Localized maximum entropy shape modellingMarco Loog
Department of Computer Science, Nordic Bioscience A S, University of Copenhagen, Herlev, Copenhagen, Denmark
Inf Process Med Imaging 20:619-29. 2007..The experiments on point distributions demonstrate that improved shape models can be obtained using this localized maximum entropy modelling...
- Automated effect-specific mammographic pattern measuresJakob Raundahl
Department of Computer Science DIKU, University of Copenhagen, 2100 Copenhagen, Denmark
IEEE Trans Med Imaging 27:1054-60. 2008..Age effects are significantly detected by our method where standard methodologies fail. The separation of HRT subpopulations using our approach is comparable to the best methodology, which is interactive...
- Efficient segmentation by sparse pixel classificationErik B Dam
Nordic Bioscience, 2730 Herlev, Denmark
IEEE Trans Med Imaging 27:1525-34. 2008..We show that each algorithm is optimal for specific tasks, and that both algorithms allow a speedup of one or more orders of magnitude on typical segmentation tasks...
- Quantifying effect-specific mammographic densityJakob Raundahl
University of Copenhagen, Department of Computer Science, Denmark
Med Image Comput Comput Assist Interv 10:580-7. 2007..Moreover, the automated method is capable of detecting age effects where standard methodologies completely fail...