Abstract

Vertebral morphometry is a commonclinically-used method for vertebral fracture detection andclassification, based on height measurements of vertebralbodies in radiographic images. This method is quantitativeand does not require specific operator skills, but its actualaccuracy is affected by errors made during the timeconsumingmanual or semi-automatic measurements. In thispaper, we propose an innovative fully automatic approach tovertebral morphometry. A novel algorithm, based on a localphase symmetry measure and an "Active Shape Model",was implemented and tested on lateral X-ray radiographs of50 patients. Thoracic and lumbar vertebral bodies in eachimage were independently segmented and measured by boththe automatic algorithm and an experienced radiologist,whose manually-obtained results were assumed as theground truth. The algorithm showed reasonably low errorrates regarding both vertebral localization and morphometricmeasurements with a sensitivity of 86.5% and a perfectspecificity of 100%, because no false positive were present.Furthermore, its performance did not appreciably worsen onpoor quality images, emphasizing a significant potential fora prompt translation into clinical routine.

Autore Pugliese

• Casciaro Ernesto , Conversano Francesco , Franchini Roberto , Casciaro Sergio

Tutti gli autori

• R.Franchini; F. Conversano; E. Casciaro; E. Quarta; A. Grimaldi; M. Muratore ; S. Casciaro

Titolo volume/Rivista

Non Disponibile

Anno di pubblicazione

2014

ISSN

Non Disponibile

ISBN

978-92-990073-5-8

Numero di citazioni Wos

Nessuna citazione

Ultimo Aggiornamento Citazioni

Non Disponibile

Numero di citazioni Scopus

Non Disponibile

Ultimo Aggiornamento Citazioni

Non Disponibile

Settori ERC

Non Disponibile

Codici ASJC

Non Disponibile