Skip to main content

Personal tools

You are here: Home / News & Events / [Research Talk] The Localized Feature Selection (LFS) Method and its Application to the Prediction of Emergence of Coma Patients

[Research Talk] The Localized Feature Selection (LFS) Method and its Application to the Prediction of Emergence of Coma Patients

By Prof. Jim Reilly, March 1, 2019 at 12 noon

Prof. Jim Reilly will be presenting at Vector Institute this Friday (March 1, 2019) at 12 noon on "The Localized Feature Selection (LFS) Method and its Application to the Prediction of Emergence of Coma Patients". This is based on the work done by Dr. Narges ArmanFard who was co-supervised by Drs. Jim Reilly and John Connolly. Both Drs. Reilly and Connolly are among the 63 selected Faculty Affiliates from across 9 institutions in Ontario.

The presenter, Dr. Jim Reilly, 
received his PhD. degree from McMaster University in 1980 in electrical engineering. He was employed in the telecommunications industry for a total of 7 years and was then appointed to the Department of Electrical and Computer Engineering at McMaster University in 1985. His research interests lie in the intersection of signal processing, Bayesian methods, machine learning, neuroscience and neuro-technology, with specific interest in the treatment and diagnosis of disorders of the brain. He is a member of the IEEE Signal Processing Society MLSP Technical Affairs Committee and is a registered professional engineer in the Province of Ontario.

Abstract:
Typical feature selection methods choose an optimal global feature subset that is applied over all regions of the sample space. In contrast, in this talk we propose the novel localized feature selection (LFS) approach whereby each region of the sample space is associated with its own distinct optimal feature subset, which may vary both in membership and size across the sample space. This allows the feature set to optimally adapt to local variations in the sample space. An associated method for measuring the similarities of a query datum to each of the respective classes is also proposed. The team demonstrate the method is robust against the over-fitting problem and that it is effective when the number of candidate features far exceeds the number of training samples.

The team apply the LFS method to the problem of predicting emergence of coma patients. Previous work has shown that the presence of the mismatch negativity (MMN) component of the event-related potential (with respect to the EEG) has strong correlation with emergence. The difficulty with current approaches is that the MMN is very difficult to detect, rendering low sensitivity values. The result show some clinical evidence that suggests the LFS method is useful in this context.