Comparison the Accuracy of Fetal Brain Extraction from T2-Half-Fourier Acquisition Single-Shot Turbo Spin-Echo (HASTE) MR Image with T2-True Fast Imaging with Steady State Free Precession (TRUFI) MR Image by Level Set Algorithm

Zabihzadeh, Mansour; Pishghadam, Morteza; Kazemi, Kamran; Nekooi, Sirous; Tahmasebi Birgani, Mohamad Javad; Seilanian-Toosi, Farrokh

doi:10.22038/ijp.2017.22292.1864

Comparison the Accuracy of Fetal Brain Extraction from T2-Half-Fourier Acquisition Single-Shot Turbo Spin-Echo (HASTE) MR Image with T2-True Fast Imaging with Steady State Free Precession (TRUFI) MR Image by Level Set Algorithm

Authors

Mansour Zabihzadeh ¹

Morteza Pishghadam ²

Kamran Kazemi ³

Sirous Nekooi ⁴

Mohamad Javad Tahmasebi Birgani ²

Farrokh Seilanian-Toosi ⁴

¹ Department of Medical Physics and Department of Clinical Oncology, Faculty of Medicine, Golestan Hospital, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran.

² Department of Medical Physics, Faculty of Medicine, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran.

³ Department of Electrical and Electronics Engineering, Shiraz University of Technology, Shiraz, Iran.

⁴ Department of Radiology, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran.

10.22038/ijp.2017.22292.1864

Abstract

Background
Access to appropriate images of fetal brain can greatly assist to diagnose of probable abnormalities. The aim of this study was to compare the suitability of T2-True Fast Imaging with Steady State Free Precession (T2-TRUFI), and T2-Half-Fourier Acquisition Single-Shot Turbo Spin-Echo (T2- HASTE( magnetic resonance imaging (MRI) to extract the fetal brain using the level set algorithm.
Materials and Method: T2-TRUFI and T2-HASTE MRI of the uterus were performed. The fetal brain was cropped from the image manually, with an adequate margin of maternal tissues; and then the fetal brain was extracted using level set. The outcome was statistically analyzed to compare its success, error (Sensitivity and Specificity), and similarities (Dice and Jaccard), with those of images obtained by radiologist.
Results: The mean values of statistical tests to evaluate the similarity (Dice and Jaccard) and the success and error (Sensitivity and specificity) between two T2-TRUFI and T2-HASTE were calculated as 97.35%, 94.98%, 95.88%, 95.88%, 99.45%, 91.10%,83.82%, 86.44% and 99.11%, respectively. However, the results from two images showed high scores to extract the fetal brain, but images from the T2-HASTE technique resulted in better visually output.
Conclusion
Based on our results, the T2-HASTE MR images are preferred to fetal brain extraction by level set algorithm compared to the T2-TRUFI MR images.

Keywords

Level set algorithm

Fetal brain extraction

MR images

sensitivity

specificity