Automatic segmentation of vertebral interfaces in echographic images

Torna indietro

Abstract

Quantitative ultrasound (QUS) methods forosteoporosis diagnosis potentially provide information aboutthe bone quality and its elastic properties. In this context, anovel ultrasound-based method for spinal and femoraldensitometry was developed by our research group. In orderto maximize its accuracy, it is very important to properlydetect the bone interfaces that will be analyzed as regions ofinterest (ROIs). A fully automatic segmentation algorithmwas developed to select lumbar vertebral interfaces inechographic images and its actual accuracy was assessed inthe present work by means of a visual checking carried outby an expert operator. Abdominal US scans of lumbar spine(from L1 to L4) were performed on 100 female subjects(60.5±3.0 years old) with different ranges of body massindex (BMI) (25.8±4.6 kg/m2). During each US scan, 100frames of radiofrequency (RF) data were stored on a PChard disk for offline analysis. The operator scanned eachvertebra, moving the probe to the next vertebra after 20seconds. For each acquired RF data frame, the implementedalgorithm generated a sectorial echographic image and, if avertebral interface was detected, it was highlighted on thesaved image. The validation procedure was performed by anexpert operator previously trained to detect the "optimal"vertebral interfaces for osteoporosis diagnosis. Resultsshowed that the segmentation algorithm had a highspecificity (93.4%), which reached its maximum on subjectswith BMI < 25 kg/m2 (94.2%), thus avoiding the selectionof false vertebral interfaces and allowing a good accuracy ofosteoporosis diagnosis.

Autore Pugliese

Tutti gli autori

  • M. Aventaggiato; F. Conversano; E. Casciaro; R. Franchini; A. Lay-Ekuakille; 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