Multiple Instance Learning for Computer Aided Diagnosis
Glenn Fung, Murat Dündar, Balaji Krishnapuram, R. Bharat Rao · The MIT Press eBooks · 2007
Many computer aided diagnosis (CAD) problems can be best modelled as a multiple-instance learning (MIL) problem with unbalanced data: i.e., the training data typically consists of a few positive bags, and a very large number of nega-tive instances. Existing MIL algorithms are much too computationally expensive for these datasets. We describe CH, a framework for learning a Convex Hull representation of multiple instances that is significantly faster than existing MIL algorithms. Our CH framework applies to any standard hyperplane-based learning algorithm, and for some algorithms, is guaranteed to find the global optimal solu-tion. Experimental studies on two different CAD applications further demonstrate that the proposed algorithm significantly improves diagnostic accuracy when com-pared to both MIL and traditional classifiers. Although not designed for standard MIL problems (which have both positive and negative bags and relatively bal-anced datasets), comparisons against other MIL methods on benchmark problems also indicate that the proposed method is competitive with the state-of-the-art. 1