Getting high-resolution computed tomography (CT) images is essential for accurate diagnoses. Pathological images often require post-processing techniques to enhance their visualization and optimize the identification of pathologies. This work aims to evaluate the added value of combining post-processing techniques in CT vascular imaging, focusing on optimizing medical pathology visualization for radiologists’ needs. Utilizing reconstruction algorithms and a dedicated workstation, studies of angiography CT were manipulated to establish ideal visualization in areas of interest such as the head, neck, abdomen, thorax, and lower limbs. Post-processing techniques (Multiplanar Reconstruction, MPR; Volume Rendering, VR; Curved Planar Reformation, CPR; and Maximum Intensity Projection, MIP) were employed. The physics team developed the post-processing protocols and 18 radiologists evaluated them using a Likert scale. Participants rated all items very positively, with scores above 4 for all protocols with a very small standard deviation, showing inter-observer agreement. Image quality, visualization of essential anatomies, ease of observing pathologies, and contribution to accurate diagnoses were evaluated very close to or at the maximum level of satisfaction, indicating a significant and highly satisfactory contribution of the post-processing techniques in the clinical practice. The lowest rates revealed some concerns from participants regarding post-processing time. Radiologists considered the time required for post-processing vascular images satisfactory for the head and neck. However, the time required for post-processing the thoracic, abdominal, and lower limb vessels requires more optimization, giving lower rate values.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Evaluation of Diagnostic Value of Post-processing Techniques for Vascular Imaging in Computed Tomography

  • Kaine Schuch Peglow,
  • L. F. S. Toschi,
  • M. L. Vezzosi,
  • A. M. Marques da Silva

摘要

Getting high-resolution computed tomography (CT) images is essential for accurate diagnoses. Pathological images often require post-processing techniques to enhance their visualization and optimize the identification of pathologies. This work aims to evaluate the added value of combining post-processing techniques in CT vascular imaging, focusing on optimizing medical pathology visualization for radiologists’ needs. Utilizing reconstruction algorithms and a dedicated workstation, studies of angiography CT were manipulated to establish ideal visualization in areas of interest such as the head, neck, abdomen, thorax, and lower limbs. Post-processing techniques (Multiplanar Reconstruction, MPR; Volume Rendering, VR; Curved Planar Reformation, CPR; and Maximum Intensity Projection, MIP) were employed. The physics team developed the post-processing protocols and 18 radiologists evaluated them using a Likert scale. Participants rated all items very positively, with scores above 4 for all protocols with a very small standard deviation, showing inter-observer agreement. Image quality, visualization of essential anatomies, ease of observing pathologies, and contribution to accurate diagnoses were evaluated very close to or at the maximum level of satisfaction, indicating a significant and highly satisfactory contribution of the post-processing techniques in the clinical practice. The lowest rates revealed some concerns from participants regarding post-processing time. Radiologists considered the time required for post-processing vascular images satisfactory for the head and neck. However, the time required for post-processing the thoracic, abdominal, and lower limb vessels requires more optimization, giving lower rate values.