Label Stability in Multiple Instance Learning

We address the problem of instance label stability in multiple instance learning MIL classifiers. These classifiers are trained only on globally annotated images bags, but often can provide fine-grained annotations for image pixels or patches instances. This is interesting for computer aided diagnosis CAD and other medical image analysis tasks for which only a coarse labeling is provided. Unfortunately, the instance labels may be unstable. This means that a slight change in training data could potentially lead to abnormalities being detected in different parts of the image, which is undesirable from a CAD point of view. Despite MIL gaining popularity in the CAD literature, this issue has not yet been addressed. We investigate the stability of instance labels provided by several MIL classifiers on 5 different datasets, of which 3 are medical image datasets breast histopathology, diabetic retinopathy and computed tomography lung images. We propose an unsupervised measure to evaluate instance stability, and demonstrate that a performance-stability trade-off can be made when comparing MIL classifiers.

[1]  Gwénolé Quellec,et al.  A multiple-instance learning framework for diabetic retinopathy screening , 2012, Medical Image Anal..

[2]  Paul A. Viola,et al.  Multiple Instance Boosting for Object Detection , 2005, NIPS.

[3]  Marco Loog,et al.  Multiple instance learning with bag dissimilarities , 2013, Pattern Recognit..

[4]  Ronald M. Summers,et al.  Seeing Is Believing: Video Classification for Computed Tomographic Colonography Using Multiple-Instance Learning , 2012, IEEE Transactions on Medical Imaging.

[5]  Bram van Ginneken,et al.  A Novel Multiple-Instance Learning-Based Approach to Computer-Aided Detection of Tuberculosis on Chest X-Rays , 2015, IEEE Transactions on Medical Imaging.

[6]  Zhuowen Tu,et al.  Weakly supervised histopathology cancer image segmentation and classification , 2014, Medical Image Anal..

[7]  Jinbo Bi,et al.  Computer Aided Detection of Pulmonary Embolism with Tobogganing and Mutiple Instance Classification in CT Pulmonary Angiography , 2007, IPMI.

[8]  A. Dirksen,et al.  The Danish Randomized Lung Cancer CT Screening Trial—Overall Design and Results of the Prevalence Round , 2009, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[9]  Lauge Sørensen,et al.  Texture-Based Analysis of COPD: A Data-Driven Approach , 2012, IEEE Transactions on Medical Imaging.

[10]  Thomas Hofmann,et al.  Support Vector Machines for Multiple-Instance Learning , 2002, NIPS.

[11]  Melih Kandemir,et al.  Empowering Multiple Instance Histopathology Cancer Diagnosis by Cell Graphs , 2014, MICCAI.

[12]  Murat Dundar,et al.  Multiple-Instance Learning Algorithms for Computer-Aided Detection , 2008, IEEE Transactions on Biomedical Engineering.

[13]  Joselene Marques.  Osteoarthritis Imaging by Quantification of Tibial Trabecular Bone , 2012 .

[14]  Jinbo Bi,et al.  Multiple Instance Learning of Pulmonary Embolism Detection with Geodesic Distance along Vascular Structure , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[15]  T. Poggio,et al.  General conditions for predictivity in learning theory , 2004, Nature.

[16]  Kim L. Boyer,et al.  A min-max framework of cascaded classifier with multiple instance learning for computer aided diagnosis , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[17]  Yixin Chen,et al.  MILES: Multiple-Instance Learning via Embedded Instance Selection , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Mohammad H. Poursaeidi,et al.  Robust support vector machines for multiple instance learning , 2012, Annals of Operations Research.

[19]  Melih Kandemir,et al.  Computer-aided diagnosis from weak supervision: A benchmarking study , 2015, Comput. Medical Imaging Graph..

[20]  Li Sun,et al.  ECG Analysis Using Multiple Instance Learning for Myocardial Infarction Detection , 2012, IEEE Transactions on Biomedical Engineering.

[21]  Lauge Sørensen,et al.  Classification of COPD with Multiple Instance Learning , 2014, 2014 22nd International Conference on Pattern Recognition.

[22]  Laude,et al.  FEEDBACK ON A PUBLICLY DISTRIBUTED IMAGE DATABASE: THE MESSIDOR DATABASE , 2014 .

[23]  Isabelle Guyon,et al.  A Stability Based Method for Discovering Structure in Clustered Data , 2001, Pacific Symposium on Biocomputing.