<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>6237</id>
    <completedYear/>
    <publishedYear>2017</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>13</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>article</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Biharmonic Density Estimate - a scale space descriptor for 3D deformable surfaces</title>
    <parentTitle language="eng">Pattern Analysis and Application</parentTitle>
    <identifier type="doi">10.1007/s10044-017-0610-2</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <author>Anirban Mukhopadhyay</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Suchendra Bhandarkar</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
  <doc>
    <id>6243</id>
    <completedYear/>
    <publishedYear>2017</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1013</pageFirst>
    <pageLast>1020</pageLast>
    <pageNumber>8</pageNumber>
    <edition/>
    <issue>6</issue>
    <volume>12</volume>
    <type>article</type>
    <publisherName>Springer</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2017-03-29</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Addressing multi-label imbalance problem of Surgical Tool Detection using CNN</title>
    <abstract language="eng">Purpose: A fully automated surgical tool detection framework is proposed for endoscopic video streams. State-of-the-art surgical tool detection methods rely on supervised one-vs-all or multi-class classification techniques, completely ignoring the co-occurrence relationship of the tools and the associated class imbalance.&#13;
&#13;
Methods: In this paper, we formulate tool detection as a multi-label classification task where tool co-occurrences are treated as separate classes. In addition, imbalance on tool co-occurrences is analyzed and stratification techniques are employed to address the imbalance during Convolutional Neural Network (CNN) training. Moreover, temporal smoothing is introduced as an online post-processing step to enhance run time prediction.&#13;
&#13;
Results: Quantitative analysis is performed on the M2CAI16 tool detection dataset to highlight the importance of stratification, temporal smoothing and the overall framework for tool detection. &#13;
&#13;
Conclusion: The analysis on tool imbalance, backed by the empirical results indicates the need and superiority of the proposed framework over state-of-the-art techniques.</abstract>
    <parentTitle language="eng">International Journal of Computer Assisted Radiology and Surgery</parentTitle>
    <identifier type="doi">10.1007/s11548-017-1565-x</identifier>
    <identifier type="url">https://link.springer.com/article/10.1007/s11548-017-1565-x</identifier>
    <note>Selected for final oral presentation</note>
    <enrichment key="PeerReviewed">yes</enrichment>
    <author>Manish Sahu</author>
    <submitter>Manish Sahu</submitter>
    <author>Anirban Mukhopadhyay</author>
    <author>Angelika Szengel</author>
    <author>Stefan Zachow</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="persons" number="zachow">Zachow, Stefan</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="projects" number="BMBF-BiOPAss">BMBF-BiOPAss</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
    <file>https://opus4.kobv.de/opus4-zib/files/6243/preprint.pdf</file>
  </doc>
  <doc>
    <id>6251</id>
    <completedYear/>
    <publishedYear>2017</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber/>
    <edition/>
    <issue>99</issue>
    <volume/>
    <type>article</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Statistical shape modeling of the left ventricle: myocardial infarct classification challenge</title>
    <abstract language="eng">Statistical shape modeling is a powerful tool for visualizing and quantifying geometric and functional patterns of the heart. After myocardial infarction (MI), the left ventricle typically remodels in response to physiological challenges. Several methods have been proposed in the literature to describe statistical shape changes. Which method best characterizes left ventricular remodeling after MI is an open research question. A better descriptor of remodeling is expected to provide a more accurate evaluation of disease status in MI patients. We therefore designed a challenge to test shape characterization in MI given a set of three-dimensional left ventricular surface points. The training set comprised 100 MI patients, and 100 asymptomatic volunteers (AV). The challenge was initiated in 2015 at the Statistical Atlases and Computational Models of the Heart workshop, in conjunction with the MICCAI conference. The training set with labels was provided to participants, who were asked to submit the likelihood of MI from a different (validation) set of 200 cases (100 AV and 100 MI). Sensitivity, specificity, accuracy and area under the receiver operating characteristic curve were used as the outcome measures. The goals of this challenge were to (1) establish a common dataset for evaluating statistical shape modeling algorithms in MI, and (2) test whether statistical shape modeling provides additional information characterizing MI patients over standard clinical measures. Eleven groups with a wide variety of classification and feature extraction approaches participated in this challenge. All methods achieved excellent classification results with accuracy ranges from 0.83 to 0.98. The areas under the receiver operating characteristic curves were all above 0.90. Four methods showed significantly higher performance than standard clinical measures. The dataset and software for evaluation are available from the Cardiac Atlas Project website1.</abstract>
    <parentTitle language="eng">IEEE Journal of Biomedical and Health Informatics</parentTitle>
    <identifier type="doi">10.1109/JBHI.2017.2652449</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <author>Avan Suinesiaputra</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Pierre Albin</author>
    <author>Xenia Alba</author>
    <author>Martino Alessandrini</author>
    <author>Jack Allen</author>
    <author>Wenjia Bai</author>
    <author>Serkan Cimen</author>
    <author>Peter Claes</author>
    <author>Brett Cowan</author>
    <author>Jan D'hooge</author>
    <author>Nicolas Duchateau</author>
    <author>Jan Ehrhardt</author>
    <author>Alejandro Frangi</author>
    <author>Ali Gooya</author>
    <author>Vicente Grau</author>
    <author>Karim Lekadir</author>
    <author>Allen Lu</author>
    <author>Anirban Mukhopadhyay</author>
    <author>Ilkay Oksuz</author>
    <author>Nripesh Parajuli</author>
    <author>Xavier Pennec</author>
    <author>Marco Pereanez</author>
    <author>Catarina Pinto</author>
    <author>Paolo Piras</author>
    <author>Marc-Michael Rohe</author>
    <author>Daniel Rueckert</author>
    <author>Dennis Saring</author>
    <author>Maxime Sermesant</author>
    <author>Kaleem Siddiqi</author>
    <author>Mahdi Tabassian</author>
    <author>Lusiano Teresi</author>
    <author>Sotirios Tsaftaris</author>
    <author>Matthias Wilms</author>
    <author>Alistair Young</author>
    <author>Xingyu Zhang</author>
    <author>Pau Medrano-Gracia</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
  <doc>
    <id>6485</id>
    <completedYear/>
    <publishedYear>2018</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>9</pageLast>
    <pageNumber/>
    <edition/>
    <issue>1</issue>
    <volume>43</volume>
    <type>article</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2017-09-14</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">An Efficient Riemannian Statistical Shape Model using Differential Coordinates</title>
    <abstract language="deu">We propose a novel Riemannian framework for statistical analysis of shapes that is able to account for the nonlinearity in shape variation. By adopting a physical perspective, we introduce a differential representation that puts the local geometric variability into focus. We model these differential coordinates as elements of a Lie group thereby endowing our shape space with a non-Euclidean structure. A key advantage of our framework is that statistics in a manifold shape space becomes numerically tractable improving performance by several orders of magnitude over state-of-the-art. We show that our Riemannian model is well suited for the identification of intra-population variability as well as inter-population differences. In particular, we demonstrate the superiority of the proposed model in experiments on specificity and generalization ability. We further derive a statistical shape descriptor that outperforms the standard Euclidean approach in terms of shape-based classification of morphological disorders.</abstract>
    <parentTitle language="eng">Medical Image Analysis</parentTitle>
    <identifier type="doi">10.1016/j.media.2017.09.004</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <enrichment key="PreprintUrn">urn:nbn:de:0297-zib-61175</enrichment>
    <enrichment key="AcceptedDate">2017-09-12</enrichment>
    <author>Christoph von Tycowicz</author>
    <submitter>Stefan Zachow</submitter>
    <author>Felix Ambellan</author>
    <author>Anirban Mukhopadhyay</author>
    <author>Stefan Zachow</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="persons" number="zachow">Zachow, Stefan</collection>
    <collection role="projects" number="BMBF-TOKMIS">BMBF-TOKMIS</collection>
    <collection role="projects" number="DFG-Knee-Laxity">DFG-Knee-Laxity</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="persons" number="vontycowicz">Tycowicz, Christoph von</collection>
    <collection role="persons" number="ambellan">Ambellan, Felix</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="projects" number="ECMath-CH15">ECMath-CH15</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
    <collection role="institutes" number="GDAP">Geometric Data Analysis and Processing</collection>
  </doc>
  <doc>
    <id>5851</id>
    <completedYear/>
    <publishedYear>2015</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>189</pageFirst>
    <pageLast>197</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>9126</volume>
    <type>conferenceobject</type>
    <publisherName>Springer</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Data-Driven Feature Learning for Myocardial Segmentation of CP-BOLD MRI</title>
    <abstract language="eng">Cardiac Phase-resolved Blood Oxygen-Level-Dependent (CP-&#13;
BOLD) MR is capable of diagnosing an ongoing ischemia by detecting changes in myocardial intensity patterns at rest without any contrast and&#13;
stress agents. Visualizing and detecting these changes require significant post-processing, including myocardial segmentation for isolating the myocardium. But, changes in myocardial intensity pattern and myocardial shape due to the heart’s motion challenge automated standard CINE MR myocardial segmentation techniques resulting in a significant drop of segmentation accuracy. We hypothesize that the main reason behind this phenomenon is the lack of discernible features. In this paper, a multi scale discriminative dictionary learning approach is proposed for supervised learning and sparse representation of the myocardium, to improve the myocardial feature selection. The technique is validated on a challenging dataset of CP-BOLD MR and standard CINE MR acquired in baseline and ischemic condition across 10 canine subjects. The proposed&#13;
method significantly outperforms standard cardiac segmentation techniques, including segmentation via registration, level sets and supervised methods for myocardial segmentation.</abstract>
    <parentTitle language="eng">Functional Imaging and Modeling of the Heart</parentTitle>
    <identifier type="doi">10.1007/978-3-319-20309-6_22</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <enrichment key="Series">Lecture Notes in Computer Science</enrichment>
    <author>Anirban Mukhopadhyay</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Ilkay Oksuz</author>
    <author>Marco Bevilacqua</author>
    <author>Rohan Dharmakumar</author>
    <author>Sotirios Tsaftaris</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
  <doc>
    <id>5852</id>
    <completedYear/>
    <publishedYear>2015</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>12</pageFirst>
    <pageLast>20</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>LNCS 9351</volume>
    <type>conferenceobject</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Unsupervised myocardial segmentation for cardiac MRI</title>
    <abstract language="eng">Though unsupervised segmentation was a de-facto standard for cardiac MRI segmentation early on, recently cardiac MRI segmentation literature has favored fully supervised techniques such as Dictionary Learning and Atlas-based techniques. But, the benefits of unsupervised&#13;
techniques e.g., no need for large amount of training data and better potential of handling variability in anatomy and image contrast, is more evident with emerging cardiac MR modalities. For example, CP-BOLD is a new MRI technique that has been shown to detect ischemia without any&#13;
contrast at stress but also at rest conditions. Although CP-BOLD looks similar to standard CINE, changes in myocardial intensity patterns and shape across cardiac phases, due to the heart’s motion, BOLD effect and artifacts affect the underlying mechanisms of fully supervised segmentation techniques resulting in a significant drop in segmentation  accuracy. In this paper, we present a fully unsupervised technique for segmenting myocardium from the background in both standard CINE MR and CP-BOLD MR. We combine appearance with motion information (obtained via Optical Flow) in a dictionary learning framework to sparsely&#13;
represent important features in a low dimensional space and separate myocardium from background accordingly. Our fully automated method&#13;
learns background-only models and one class classifier provides myocardial segmentation. The advantages of the proposed technique are demonstrated on a dataset containing CP-BOLD MR and standard CINE MR&#13;
image sequences acquired in baseline and ischemic condition across 10&#13;
canine subjects, where our method outperforms state-of-the-art supervised segmentation techniques in CP-BOLD MR and performs at-par for standard CINE MR.</abstract>
    <parentTitle language="eng">Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015</parentTitle>
    <identifier type="doi">10.1007/978-3-319-24574-4_2</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <author>Anirban Mukhopadhyay</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Ilkay Oksuz</author>
    <author>Marco Bevilacqua</author>
    <author>Rohan Dharmakumar</author>
    <author>Sotirios Tsaftaris</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
  <doc>
    <id>5854</id>
    <completedYear/>
    <publishedYear>2015</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>205</pageFirst>
    <pageLast>213</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>9350</volume>
    <type>conferenceobject</type>
    <publisherName>Springer</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Dictionary Learning Based Image Descriptor for Myocardial Registration of CP-BOLD MR</title>
    <abstract language="deu">Cardiac Phase-resolved Blood Oxygen-Level-Dependent (CP-&#13;
BOLD) MRI is a new contrast agent- and stress-free imaging technique&#13;
&#13;
for the assessment of myocardial ischemia at rest. The precise registration &#13;
among the cardiac phases in this cine type acquisition is essential for automating the analysis of images of this technique, since it can potentially lead to better specificity of ischemia detection. However, inconsistency in myocardial intensity patterns and the changes in myocardial shape&#13;
due to the heart’s motion lead to low registration performance for state-&#13;
of-the-art methods. This low accuracy can be explained by the lack of distinguishable features in CP-BOLD and inappropriate metric defini-&#13;
tions in current intensity-based registration frameworks. In this paper,&#13;
the sparse representations, which are defined by a discriminative dictionary learning approach for source and target images, are used to improve myocardial registration. This method combines appearance with Gabor and HOG features in a dictionary learning framework to sparsely represent features in a low dimensional space. The sum of squared differences&#13;
of these distinctive sparse representations are used to define a similarity term in the registration framework. The proposed descriptor is validated on a challenging dataset of CP-BOLD MR and standard CINE MR acquired in baseline and ischemic condition across 10 canines.</abstract>
    <parentTitle language="deu">Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015</parentTitle>
    <identifier type="doi">10.1007/978-3-319-24571-3_25</identifier>
    <enrichment key="PeerReviewed">yes</enrichment>
    <enrichment key="Series">Lecture Notes in Computer Science</enrichment>
    <author>Ilkay Oksuz</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Anirban Mukhopadhyay</author>
    <author>Marco Bevilacqua</author>
    <author>Rohan Dharmakumar</author>
    <author>Sotirios Tsaftaris</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
  <doc>
    <id>5867</id>
    <completedYear/>
    <publishedYear>2016</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>9</pageLast>
    <pageNumber/>
    <edition>IEEE Winter Conference on Applications of Computer Vision (WACV)</edition>
    <issue/>
    <volume/>
    <type>proceedings</type>
    <publisherName>IEEE</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Joint Geometric Graph Embedding for Partial Shape Matching in Images</title>
    <abstract language="eng">A novel multi-criteria optimization framework for matching of partially visible shapes in multiple images using joint geometric graph embedding is proposed. The proposed framework achieves matching of partial shapes in images that exhibit extreme variations in scale, orientation, viewpoint and illumination and also instances of occlusion; conditions which render impractical the use of global contour-based descriptors or local pixel-level features for shape matching. The proposed technique is based on optimization of the embedding distances of geometric features obtained from the eigenspectrum of the joint image graph, coupled with regularization over values of the mean pixel intensity or histogram of oriented gradients. It is shown to obtain successfully the correspondences denoting partial shape similarities as well as correspondences between feature points in the images. A new benchmark dataset is proposed which contains disparate image pairs with extremely challenging variations in viewing conditions when compared to an existing dataset [18]. The proposed technique is shown to significantly outperform several state-of-the-art partial shape matching techniques on both datasets.</abstract>
    <parentTitle language="eng">IEEE Winter Conference on Applications of Computer Vision</parentTitle>
    <enrichment key="PeerReviewed">yes</enrichment>
    <author>Anirban Mukhopadhyay</author>
    <submitter>Anirban Mukhopadhyay</submitter>
    <author>Arun Kumar</author>
    <author>Suchendra Bhandarkar</author>
    <collection role="institutes" number="vis">Visual Data Analysis</collection>
    <collection role="institutes" number="medplan">Therapy Planning</collection>
    <collection role="projects" number="MODAL-MedLab">MODAL-MedLab</collection>
    <collection role="projects" number="Modal-Seg">Modal-Seg</collection>
    <collection role="projects" number="MODAL-Gesamt">MODAL-Gesamt</collection>
    <collection role="institutes" number="VDcC">Visual and Data-centric Computing</collection>
  </doc>
</export-example>
