Fully automated accurate patient positioning in computed tomography using anterior–posterior localizer images and a deep neural network: a dual-center study - Takeaways - MDSpire

Fully automated accurate patient positioning in computed tomography using anterior–posterior localizer images and a deep neural network: a dual-center study

  • By

  • Yazdan Salimi

  • Isaac Shiri

  • Azadeh Akavanallaf

  • Zahra Mansouri

  • Hossein Arabi

  • Habib Zaidi

  • January 27, 2023

  • 0 min

Share

  • 1

    Deep learning algorithms can automate patient-specific positioning in chest CT scans using only anterior-posterior localizer images.

  • 2

    A study evaluated 7295 chest CT images, focusing on optimizing patient positioning to enhance diagnostic value and minimize radiation risks.

  • 3

    Mis-centering in CT scans can lead to significant increases in radiation doses and degraded image quality, affecting patient safety.

  • 4

    Previous methods for automatic patient positioning often relied on 3D cameras, which can introduce errors and require precise calibration.

  • 5

    The study employed a modified U-NET deep neural network to improve the accuracy of patient positioning in computed tomography.

Original Source(s)

Fully automated accurate patient positioning in computed tomography using anterior–posterior localizer images and a deep neural network: a dual-center study

  • by Yazdan Salimi, Isaac Shiri, Azadeh Akavanallaf, Zahra Mansouri, Hossein Arabi, Habib Zaidi

  • January 27, 2023

Related Content

Infections Lead Thoracic CT Findings in SLE

Researchers characterize spectrum of pulmonary and extrathoracic involvement in systemic lupus erythematosus.

CT imaging features of pulmonary SMARCA4-deficient undifferentiated carcinoma: a retrospective case series

Combining radiomics and machine learning for enhanced localization of premature ventricular contractions