Computational Radiology and Crohn's Disease

@inproceedings{schuffler_semi-automatic_2014,
    title = {Semi-{Automatic} {Crohn}'s {Disease} {Severity} {Estimation} on {MR} {Imaging}},
    url = {http://link.springer.com/chapter/10.1007%2F978-3-319-13692-9_12},
    abstract = {Crohn’s disease (CD) is a chronic inflammatory bowel disease which can be visualized by magnetic resonance imaging (MRI). For CD grading, several non-invasive MRI based severity scores are known, most prominent the MaRIA and AIS. As these scores rely on manual MRI readings for individual bowel segments by trained radiologists, automated MRI assessment has been more and more focused in recent research. We show on a dataset of 27 CD patients that semi-automatically measured bowel wall thickness (ABWT) and dynamic contrast enhancement (DCE) completely outperform manual scorings: the segmental correlation to the Crohn’s Disease Endoscopic Index of Severity (CDEIS) of ABWT and DCE is significantly higher (r = .78) than that of MaRIA (r = .45) or AIS (r = .51). Also on a per-patient basis, the models with ABWT and DCE show significantly higher correlation (r = .69) to global CDEIS than MaRIA (r = .46).},
    author = {Sch\"uffler, P. J. and Mahapatra, D. and Naziroglu, R. E. and Li, Z. and Puylaert, C. A. J. and Andriantsimiavona, R. and Vos, F. M. and Pends\'e, D. A. and Nio, C. Yung and Stoker, J. and Taylor, S. A. and Buhmann, J. M.},
    year = {2014},
}

@misc{schuffler_computer_2014,
    address = {London},
    type = {Poster},
    title = {Computer {Aided} {Crohn}'s {Disease} {Severity} {Assessment} in {MRI}},
    url = {https://www.researchgate.net/publication/262198303_Computer_Aided_Crohns_Disease_Severity_Assessment_in_MRI},
    author = {Sch\"uffler, Peter J. and Mahapatra, Dwarikanath and Vos, Franciscus M. and Buhmann, Joachim M.},
    year = {2014},
}

@incollection{schuffler_model_2013,
    series = {Lecture {Notes} in {Computer} {Science}},
    title = {A {Model} {Development} {Pipeline} for {Crohn}'s {Disease} {Severity} {Assessment} from {Magnetic} {Resonance} {Images}},
    volume = {8198},
    isbn = {978-3-642-41082-6},
    url = {http://link.springer.com/chapter/10.1007%2F978-3-642-41083-3_1},
    abstract = {Crohn's Disease affects the intestinal tract of a patient and can have varying severity which influences treatment strategy. The clinical severity score CDEIS (Crohn’s Disease Endoscopic Index of severity) ranges from 0 to 44 and is measured by endoscopy. In this paper we investigate the potential of non-invasive magnetic resonance imaging to assess this severity, together with the underlying question which features are most relevant for this estimation task. We propose a new general and modular pipeline that uses machine learning techniques to quantify disease severity from MR images and show its value on Crohn’s Disease severity assessment on 30 patients scored by 4 medical experts. With the pipeline, we can obtain a magnetic resonance imaging score which outperforms two existing reference scores MaRIA and AIS.},
    booktitle = {Abdominal {Imaging}. {Computation} and {Clinical} {Applications}},
    publisher = {Springer Berlin Heidelberg},
    author = {Sch\"uffler, Peter J. and Mahapatra, Dwarikanath and Tielbeek, Jeroen A. W. and Vos, Franciscus M. and Makanyanga, Jesica and Pends\'e, Doug A. and Nio, C. Yung and Stoker, Jaap and Taylor, Stuart A. and Buhmann, Joachim M.},
    editor = {Yoshida, Hiroyuki and Warfield, Simon and Vannier, Michael},
    year = {2013},
    keywords = {AIS, CDEIS, Crohn’s Disease, MaRIA, abdominal MRI},
    pages = {1--10},
}

@article{mahapatra_crohns_2015,
    title = {Crohn's {Disease} {Segmentation} from {MRI} {Using} {Learned} {Image} {Priors}},
    url = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=7163951},
    abstract = {We use a Field of Experts (FoE) model to segment abdominal regions from MRI affected with Crohns Disease (CD). FoE learns a prior model of diseased and normal bowel, and background non-bowel tissues from manually annotated training images. Unlike current approaches, FoE does not rely on hand designed features but learns the most discriminative features (in the form of filters) for different classes. FoE filter responses are integrated into a Random forest (RF) model that outputs probability maps for the test image and finally segments the diseased region. Experimental results show our method achieves significantly better performance than existing methods.},
    journal = {Proceedings IEEE ISBI 2015},
    author = {Mahapatra, D. and Sch\"uffler, P. J. and Vos, F. M. and Buhmann, J. M.},
    year = {2015},
    pages = {625--628},
}

@article{mahapatra_automatic_2013,
    title = {Automatic {Detection} and {Segmentation} of {Crohn}'s {Disease} {Tissues} from {Abdominal} {MRI}},
    volume = {32},
    issn = {0278-0062},
    url = {http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6600949},
    doi = {10.1109/TMI.2013.2282124},
    abstract = {We propose an information processing pipeline for segmenting parts of the bowel in abdominal magnetic resonance images that are affected with Crohn's disease. Given a magnetic resonance imaging test volume, it is first oversegmented into supervoxels and each supervoxel is analyzed to detect presence of Crohn's disease using random forest (RF) classifiers. The supervoxels identified as containing diseased tissues define the volume of interest (VOI). All voxels within the VOI are further investigated to segment the diseased region. Probability maps are generated for each voxel using a second set of RF classifiers which give the probabilities of each voxel being diseased, normal or background. The negative log-likelihood of these maps are used as penalty costs in a graph cut segmentation framework. Low level features like intensity statistics, texture anisotropy and curvature asymmetry, and high level context features are used at different stages. Smoothness constraints are imposed based on semantic information (importance of each feature to the classification task) derived from the second set of learned RF classifiers. Experimental results show that our method achieves high segmentation accuracy with Dice metric values of 0.90 ± 0.04 and Hausdorff distance of 7.3 ± 0.8 mm. Semantic information and context features are an integral part of our method and are robust to different levels of added noise.},
    journal = {IEEE Transactions on Medical Imaging},
    author = {Mahapatra, D. and Sch\"uffler, P. J. and Tielbeek, J. A. W. and Makanyanga, J. C. and Stoker, J. and Taylor, S. A. and Vos, F. M. and Buhmann, J. M.},
    year = {2013},
    keywords = {Anisotropic magnetoresistance, Context, Crohn\&\#x2019, Diseases, Entropy, Image segmentation, Radio frequency, content, graph cut, image features, probability maps, random forests, s disease, segmentation, semantic information, shape, supervoxels},
    pages = {2332--2347},
}

@article{saur_effect_2010,
    title = {Effect of {Reader} {Experience} on {Variability}, {Evaluation} {Time} and {Accuracy} of {Coronary} {Plaque} {Detection} with {Computed} {Tomography} {Coronary} {Angiography}},
    volume = {20},
    issn = {0938-7994},
    url = {http://dx.doi.org/10.1007/s00330-009-1709-7},
    doi = {10.1007/s00330-009-1709-7},
    number = {7},
    journal = {European Radiology},
    author = {Saur, Stefan and Alkadhi, Hatem and Stolzmann, Paul and Baumueller, Stephan and Leschka, Sebastian and Scheffel, Hans and Desbiolles, Lotus and Fuchs, Thomas J. and Szekely, Gabor and Cattin, Philippe},
    year = {2010},
    keywords = {Medicine},
    pages = {1599--1606},
}

@article{saur_guided_2009,
    title = {Guided {Review} by {Frequent} {Itemset} {Mining}: {Additional} {Evidence} for {Plaque} {Detection}},
    volume = {4},
    url = {http://dx.doi.org/10.1007/s11548-009-0290-5},
    number = {3},
    journal = {International Journal of Computer Assisted Radiology and Surgery},
    author = {Saur, Stefan C. and Alkadhi, Hatem and Desbiolles, Lotus and Fuchs, Thomas J. and Szekely, Gabor and Cattin, Philippe C.},
    year = {2009},
    pages = {263--271},
}

@article{saur_prediction_2009,
    title = {Prediction {Rules} for the {Detection} of {Coronary} {Artery} {Plaques}: {Evidence} from {Cardiac} {CT}},
    volume = {44},
    url = {http://journals.lww.com/investigativeradiology/Abstract/2009/08000/Prediction_Rules_for_the_Detection_of_Coronary.8.aspx},
    doi = {http://dx.doi.org/10.1097/RLI.0b013e3181a8afc4},
    abstract = {Objectives: To evaluate spatial plaque distribution patterns in coronary arteries based on computed tomography coronary angiography data sets and to express the learned patterns in prediction rules. An application is proposed to use these prediction rules for the detection of initially missed plaques.
    Material and Methods: Two hundred fifty two consecutive patients with chronic coronary artery disease underwent contrast-enhanced dual-source computed tomography coronary angiography for clinical indications. Coronary artery plaques were manually labeled on a 16-segment coronary model and their position (ie, segments and bifurcations) and composition (ie, calcified, mixed, or noncalcified) were noted. The frequent itemset mining algorithm was used to statistically search for plaque distribution patterns. The patterns were expressed as prediction rules: given plaques at certain locations as conditions, a prediction rule gave evidence—with a certain confidence value—for a plaque at another location within the coronary artery tree. Prediction rules with the highest confidence values were evaluated and described. Furthermore, to improve manual plaque detection, all prediction rules were applied on the patient data to search for segments with potentially missed plaques. These segments were then reviewed in a second, guided reading for the existence of plaques. The same number of segments was also determined by a weighted random approach to evaluate the quality of prediction resulting from frequent itemset mining.
    Results: In 200 of 252 (79.4\%) patients, at least one coronary plaque (range, 1–22 plaques) was found. In total 1229 plaques (990 calcified, 80.6\%; 227 mixed, 18.5\%; 12 noncalcified, 1\%) distributed, over 916 coronary segments and 507 vessels were manually labeled. Four plaque distribution patterns were identified: 20.6\% of the patients had no plaques at all; 31.7\% had plaques in the left coronary artery tree; 46.4\% had plaques both in left and right coronary arteries, whereas 1.2\% of the patients had plaques solely in the right coronary artery (RCA). General rules were found predicting plaques in the left anterior descending artery (LAD), given plaques in segments of the RCA or in the left main artery. Further general rules predicted plaques in the LAD, given plaques in the circumflex artery. In the guided review, the segment selection based on the prediction rules from frequent itemset mining performed significantly better (P {\textless} 0.001) than the weighted random approach by revealing 48 initially missed plaques.
    Conclusions: This study demonstrates spatial plaque distribution patterns in coronary arteries as determined with cardiac CT. Use of the frequent itemset mining algorithm yielded rules that predicted plaques at certain sites given plaques at other sites of the coronary artery tree. Use of these prediction rules improved the manual labeling of coronary plaques as initially missed plaques could be predicted with the guided review.},
    number = {8},
    journal = {Investigative Radiology},
    author = {Saur, Stefan C. and Cattin, Philippe C. and Desbiolles, Lotus and Fuchs, Thomas J. and Szekely, Gabor and Alkadhi, Hatem},
    year = {2009},
    pages = {483--490},
}

@inproceedings{fuernstahl_automatic_2008,
    address = {Paris, France},
    title = {Automatic and {Robust} {Forearm} {Segmentation} {Using} {Graph} {Cuts}},
    isbn = {978-1-4244-2002-5},
    url = {http://dx.doi.org/10.1109/ISBI.2008.4540936},
    doi = {10.1109/ISBI.2008.4540936},
    abstract = {The segmentation of bones in computed tomography (CT) images is an important step for the simulation of forearm bone motion, since it allows to include patient specific anatomy in a kinematic model. While the identification of the bone diaphysis is straightforward, the segmentation of bone joints with weak, thin, and diffusive boundaries is still a challenge. We propose a graph cut segmentation approach that is particularly suited to robustly segment joints in 3-d CT images. We incorporate knowledge about intensity, bone shape and local structures into a novel energy function. Our presented framework performs a simultaneous segmentation of both forearm bones without any user interaction.},
    booktitle = {5th {IEEE} {International} {Symposium} on {Biomedical} {Imaging}. {ISBI} 2008},
    publisher = {IEEE},
    author = {Fuernstahl, Philipp and Fuchs, Thomas J. and Schweizer, Andreas and Nagy, Ladislav and Szekely, Gabor and Harders, Matthias},
    year = {2008},
    pages = {77--80},
}

@inproceedings{kaynig_neuron_2010,
    address = {Los Alamitos, CA, USA},
    title = {Neuron {Geometry} {Extraction} by {Perceptual} {Grouping} in {ssTEM} {Images}},
    volume = {0},
    isbn = {978-1-4244-6984-0},
    url = {http://www.computer.org/portal/web/csdl/doi/10.1109/CVPR.2010.5540029},
    doi = {http://doi.ieeecomputersociety.org/10.1109/CVPR.2010.5540029},
    booktitle = {{IEEE} {Computer} {Society} {Conference} on {Computer} {Vision} and {Pattern} {Recognition}},
    publisher = {IEEE Computer Society},
    author = {Kaynig, Verena and Fuchs, Thomas J. and Buhmann, Joachim M.},
    year = {2010},
    pages = {2902--2909},
}

@inproceedings{kaynig_geometrical_2010,
    address = {Beijing, China},
    series = {{MICCAI}'10},
    title = {Geometrical {Consistent} {3D} {Tracing} of {Neuronal} {Processes} in {ssTEM} {Data}},
    isbn = {3-642-15744-0 978-3-642-15744-8},
    url = {http://portal.acm.org/citation.cfm?id=1928047.1928075},
    booktitle = {Proceedings of the 13th international conference on {Medical} image computing and computer-assisted intervention: {Part} {II}},
    publisher = {Springer-Verlag, Berlin, Heidelberg},
    author = {Kaynig, Verena and Fuchs, Thomas J. and Buhmann, Joachim M.},
    year = {2010},
    pages = {209--216},
}