Deep Learning Wins MICCAI 2013 Grand Challenge
and ICPR 2012 Contest on Mitosis Detection
(first Deep Learner to win a contest on object detection in large images)
Nagoya, Japan, 22 September 2013, at the MICCAI 2013 conference organised by the Society for Medical Image Computing and Computer Assisted Intervention:
Our Deep Learning Neural Networks (deep NN) won the international MICCAI 2013 Grand Challenge on Mitosis Detection. This was made possible through the efforts of my co-workers Dr. Dan Claudiu Ciresan and Dr. Alessandro Giusti.
Mitosis detection is important for cancer prognosis, but difficult even for trained experts.
Last year, our deep NN already won the ICPR 2012 Contest on Mitosis Detection in Breast Cancer Histological Images, using the same technique - the paper on this  was also published at MICCAI 2013. Title defended!
To our knowledge, this was the first Deep Learner to win a contest on object detection in large images.
The data set of the MICCAI 2013 Grand Challenge , however, was even much larger and more challenging than the one of ICPR 2012 [10,10a]: a real-world dataset including many ambiguous cases and frequently encountered problems such as imperfect slide staining.
More than 89 research groups (universities and companies) registered; 14 submitted results. The gap between best and second best result is large.
Pure supervised gradient descent (40-year-old efficient reverse mode backpropagation, e.g., [2a,2]) was applied [5,5a] to our special neural architecture [3,4,1] consisting of deep and wide GPU-based Multi-Column Max-Pooling Convolutional Neural Networks (MC GPU-MPCNN or simply deep NN) with alternating weight-sharing convolutional layers [8,5] and max-pooling layers [9,9a,5a,6] topped by fully connected layers  (over two decades, LeCun's lab has invented many improvements of such CNN). This architecture is biologically rather plausible, inspired by early neuroscience-related work [7,8]. (No unsupervised pre-training!)
Our MC GPU-MPCNN [3,4] also were the first systems to achieve human-competitive or even superhuman pattern recognition performance on various other benchmarks [4,13-16]. Many leading IT companies and research labs are now using them, too, e.g., . Besides mitosis detection, our deep NN also have many other obvious biomedical applications, such as automatic detection of melanoma, detection of plaque in CT heart scans, segmentation of all kinds of biomedical images - you name it.
The world spends over 10% of GDP on healthcare (over 6 trillion USD per year), much of it on medical diagnosis through expensive experts. Partial automation of this could not only save billions of dollars, but also make expert diagnostics accessible to many who currently cannot afford it.
When we started Deep Learning research over 20 years ago [13,16], limited computing power forced us to focus on toy applications. How things have changed! It is gratifying to observe that today our deep NN may actually help to improve healthcare and save human lives.
 Dan Claudiu Ciresan, Alessandro Giusti, Luca Maria Gambardella, Jürgen Schmidhuber. Mitosis Detection in Breast Cancer Histology Images using Deep Neural Networks. MICCAI 2013.
 Paul J. Werbos. Applications of advances in nonlinear sensitivity analysis. In R. Drenick, F. Kozin, (eds): System Modeling and Optimization: Proc. IFIP (1981), Springer, 1982.
[2a] S. Linnainmaa. The representation of the cumulative rounding error of an algorithm as a Taylor expansion of the local rounding errors. Master's Thesis (in Finnish), Univ. Helsinki, 1970. See chapters 6-7 and FORTRAN code on pages 58-60.
 D. C. Ciresan, U. Meier, J. Masci, L. M. Gambardella, J. Schmidhuber. Flexible, High Performance Convolutional Neural Networks for Image Classification. International Joint Conference on Artificial Intelligence (IJCAI-2011, Barcelona), 2011. PDF. ArXiv preprint.
 D. C. Ciresan, U. Meier, J. Schmidhuber. Multi-column Deep Neural Networks for Image Classification. Proc. IEEE Conf. on Computer Vision and Pattern Recognition CVPR 2012, p 3642-3649, 2012. PDF. Longer TR: arXiv:1202.2745v1 [cs.CV]
 Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, L. D. Jackel: Backpropagation Applied to Handwritten Zip Code Recognition, Neural Computation, 1(4):541-551, 1989.
M. Ranzato, F.J. Huang, Y. Boureau, Y. LeCun. Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition. Proc. CVPR 2007, Minneapolis, 2007.
 D. Scherer, A. Mueller, S. Behnke. Evaluation of pooling operations in convolutional architectures for object recognition. In Proc. ICANN 2010.
 Hubel, D. H., T. N. Wiesel. Receptive Fields, Binocular Interaction And Functional Architecture In The Cat's Visual Cortex. Journal of Physiology, 1962.
 K. Fukushima. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 36(4): 193-202, 1980.
 Weng, J., Ahuja, N., and Huang, T. S. (1992). Cresceptron: a self-organizing neural network which grows adaptively. In International Joint Conference on Neural Networks (IJCNN), vol 1, p 576-581.
[9a] M. Riesenhuber, T. Poggio. Hierarchical models of object recognition in cortex. Nature Neuroscience 11, p 1019-1025, 1999.
 ICPR 2012 Contest on Mitosis Detection in Breast Cancer Histological Images (MITOS dataset)
[10a] L. Roux, D. Racoceanu, N. Lomenie, M. Kulikova, H. Irshad, J. Klossa, F. Capron, C. Genestie, G. Le Naour, M. N. Gurcan. Mitosis detection in breast cancer histological images - An ICPR 2012 contest. J Pathol. Inform. 4:8, 2013.
 MICCAI 2013 Grand Challenge on Mitosis Detection, organised by M. Veta, M.A. Viergever, J.P.W. Pluim, N. Stathonikos, P. J. van Diest of University Medical Center Utrecht
 Krizhevsky, A., Sutskever, I. and Hinton, G. E. ImageNet Classification with Deep Convolutional Neural Networks. NIPS 25, MIT Press, 2012.
 J. Schmidhuber. My first Deep Learner of 1991 + Deep Learning timeline 1962-2013
 J. Schmidhuber. 2011: First Superhuman Visual Pattern Recognition
 J. Schmidhuber. First Human-Competitive Handwriting Recognition Through Deep Learning
 J. Schmidhuber. Deep Learning since 1991