TITLE: Managing Uncertainty in Computer Aided Diagnosis of Diffuse Lung Diseases

PRESENTER: Mamatha Rudrapatna, http://www.cse.unsw.edu.au/~mamathar/, mamathar@cse.unsw.edu.au

AFFILIATION:School of Computer Science and Engineering, UNSW, http://www.cse.unsw.edu.au/

DATE: Friday 24th March 2006

TIME: 14:00:00

PLACE: Level 4 Meeting room K17

ABSTRACT:

The task of pattern / object recognition on images usually
involves sequential application of a number of vision operators for
segmentation, feature extraction and classification. The overall quality
of a vision system depends on each operator used since inaccuracies are
introduced at every stage. Vision systems used for mission critical
applications such as clinical support simply require high accuracy and
the usual sensitivity-specificity trade-off is not possible since such
applications demand very few false negatives without too many false
positives. In this thesis, a hierarchical fusion framework is proposed
that helps analyse error propagation through various stages and improve
overall accuracy. Preliminary results are presented.

BIOGRAPHY OF SPEAKER:

Mamatha Rudrapatna is PhD Student at the School of Computer Science and Engineering, UNSW. Her PhD topic is on Computer Aided Diagnosis for
Diffuse Lung Diseases.

Host:

Arcot Sowmya

Seminar Convenor:

Van Hai Ho