Computer-aided Diagnosis in Orthopedic Radiology: Quantitative Evaluation of Vertebral Morphometry

Event Date

Apr 20 2012 - 3:00pm - 4:00pm

Location

ICT 518

Description

The analysis of radiographic images is fundamental to the assessment of deformities of the musculoskeletal system and facilitates quantitative assessment of processes related to the emergence and development of deformities, such as those due to scoliois and compression fractures. Most of the methods for quantitative assessment of vertebral fractures proposed in the literature require the marking of reference points on each vertebral body (e.g., points that enable the measurement of the height of the anterior, medial, and posterior portions of a vertebra). However, due to the extensive work involved with such analysis performed manually, the usability of the methods is not clinically practical on a large scale.

In this context, the development of techniques based on computer vision and image processing, which can minimize or obviate the time and effort needed of the expert to analyze vertebral morphometry, is desirable. This seminar will present digital image processing and pattern recognition methods to obtain quantitative measures of vertebral bodies in lateral lumbar radiographic images. The methods are expected to enable efficient approaches to quantitative vertebral morphometry in studies of vertebral deformities and fractures.

For further information contact: Dr. Raj Rangayyan at: ranga@ucalgary.ca

Presenter/Speaker

Paulo Mazzoncini de Azevedo Marques, PhD Associate Professor in Biomedical Informatics and Medical Physics School of Medicine of Ribeirão Preto University of São Paulo, Brazil