2D image post-processing.
①Multiplanar reconstruction (MPR)
MPR is a two-dimensional image processing method to obtain arbitrary coronal, sagittal, transverse, and oblique images of human tissues and organs from the original transverse-axis images after post-processing, which is very similar to MR images and shows the morphological changes of various systems and organs of the whole body, especially in determining the nature, extent of invasion, and adjacent relationships of anatomical structures and organs such as skull base, neck, pulmonary hilar, mediastinum, abdomen, pelvis, and large vessels. (2) Surface reconstruction (CPR)
Surface reconstruction (CPR): It is a special method of MPR and is suitable for the display of some curved structures and organs of the human body, such as jaws, tortuous blood vessels and bronchioles. The objective jaw accuracy of the curved reconstruction image is very closely related to the accuracy of the operator’s dotted line. c) Computational volume reconstruction (CVR) CVR is another special way of MPR. It is used to increase the layer thickness of coronal, sagittal, transaxial and oblique images in order to show the morphology of the tissue and organ structures that travel parallel to the plane, such as blood vessels and bronchi. It can also increase the signal-to-noise ratio of the image.
Data acquisition requirements: 1. Positive body position; 2. Layer thickness ≤ 1.0mm/per layer for head and neck organs and bones; layer thickness ≤ 3.0mm/per layer for thoracic and abdominal organs, with 50% overlap reconstruction; 3. FC 10 (soft tissue)/FC 30 (bone) for reconstruction function; 4. 5, the sternoclavicular joint, shoulder and hip joints and other parts of the reconstruction image must be selected RASP to remove artifact interference.
Key points of 2D image post-processing techniques: 1. Adjust the window width and window position appropriately; 2. Generate axial preview images at small intervals (<2mm) to determine the location and extent of the lesion; 3. Adjust the interval, layer thickness and image frame number to generate MPR images for the determined lesion extent; 4. If the patient is not in proper position, adjust the image with oblique reconstruction to obtain symmetrical images.
3-D image post-processing:
(a) 3D volume reconstruction
Volume reconstruction (VR).
VR is currently one of the most commonly used techniques in post-processing multilevel spiral CT 3D images. VR images are mainly suitable for displaying lesions of the following organs and systems
(1) Bone
VR images can show the morphology of physiological protrusions (e.g., spines, rudiments, nodes and crests, etc.), depressions (e.g., fossae, grooves and indentations, etc.), cavities (e.g., cavities, sinuses, canals, tracts, holes, etc.) and enlargements (e.g., head, neck and condyles, etc.) of normal skull, trunk and extremity bones, as well as bony structures of joints (e.g., joint heads and pelvises, etc.) in a three-dimensional, visual and clear manner. For long, short, flat and irregular bones, especially for fractures of wrist, ankle, elbow, shoulder, hip and spine and their accessories, joint dislocations, deformities and bone tumors, the location, extent, scope and adjacent relationship with surrounding tissues and organs, and for orthopedic and plastic surgery to develop surgical plans, predict the possibility of surgery and assess the healing of surgery It has high clinical application value for orthopedics and orthopedic surgery, predicting the possibility of surgery and assessing the healing of surgery.
Requirements for data acquisition: a) Positioning: b) Layer thickness <2.0 mm per layer, overlap reconstruction interval ≤ 0.5 mm; c) Use of skeletal reconstruction function FC30: d) Use of small field of view to enlarge the scan as much as possible under the condition of ensuring sufficient scanning range for the hand, metacarpal and joint; e) Use of RASP parameters for the reconstruction of sternoclavicular joint, shoulder joint and hip joint to (e) the sternoclavicular joint, shoulder joint and hip joint should be reconstructed with RASP parameters to remove artifact interference; f) the patient should be in the open mouth position (or bite pad) during the maxillofacial scan.
Key points of image post-processing techniques: a) accurately select the upper and lower limits of the preset CT values, especially for the reconstruction of thin flat bones (e.g., scapula) with special care to avoid artificial bone defects or the illusion of destruction; b) use CIipping, Cutting and other tools to remove the interference of scanning brackets, fixed casts and other images and clearly reveal the lesion, if necessary: c) use the Seed technique to perform electronic joint separation for bone joints. Seed technology can be used to perform electronic joint separation for a clearer view of the joint head and cap; d) Appropriate adjustment of the intensity of pseudo-color and masking light to make the images clearer and more realistic: e) Accurate positioning with the help of MPR images is required when judging complex anatomical structures or small fracture gaps and free fragments; f) Multi-angle rotation of the image can show the lesion site and its three-dimensional spatial relationship with adjacent structures as clearly and completely as possible. The three-dimensional spatial relationship with adjacent structures.
(2) Vascular system
VR as the main post-processing technology of MS-CTA in the vascular system, especially for the arterial vascular system lesions to clearly and accurately display the complete morphology, alignment and lesions of large and complex vessels, with a strong sense of stereoscopic image, can intuitively display the three-dimensional spatial anatomical relationship between lesions and vessels, vessels and vessels and other organs around them from multiple angles, and its diagnostic value has been recognized by clinicians. The diagnosis of large arterial vascular lesions such as: aneurysm, arteriovenous malformation, stenosis, infarction, occlusion, entrapment and calcification of the vessel wall has basically replaced DSA examination. The diagnosis of cerebral aneurysm has been confirmed by domestic and international studies, and 3D-CTA has high accuracy, sensitivity and specificity, and can accurately detect cerebral aneurysm with a diameter of <3mm. As a rapid and non-invasive examination, it can accurately show the location, shape and size of the aneurysm, evaluate the spatial relationship between the aneurysm neck and the aneurysm body, the aneurysm-carrying artery and the surrounding vessels, and simulate the surgical access to provide a visual and reliable basis for selecting the appropriate surgical treatment plan. In recent years, there are many reports in the literature advocating the use of 3D-CTA to replace or partially replace DSA for the diagnosis of cerebral aneurysms.
Cerebral artery CTA data acquisition requirements: a) acquisition layer thickness ≤ 3.0mm/per layer; b) overlapping reconstruction interval ≤ 2.0mm; c) selection of soft tissue reconstruction function, such as FC=10/43; d) contrast agent dosage 1.0-2.0ml/kg; e) injection rate 2.5-3.0ml/sec; f) delay time 15-20sec. and contrast agent available if necessary tracking technique (Sure-Start); g) scanning direction from the bottom up; h) scanning range for Willis circumflex aneurysm from the first cervical vertebra upward 10 cm, and try to use magnification scanning technique.
Key points of its image post-processing techniques.
a. Accurately select the upper and lower limits of the preset CT values, too high or too low will image the clarity and authenticity of lesion display. However, appropriately increasing the lower limit value can identify whether the posterior communicating artery is an aneurysm or a funnel-like dilatation. After gradually changing the domain value, the aneurysm remains domed, while the funnel-like dilatation becomes conical; b) remove the interference of images of the inferior sagittal sinus, straight sinus and large cerebral vein as well as the skull with tools such as Clipping or Cutting; c) from anterior-posterior, posterior-anterior, left-right lateral and cephalad and pedalad (d) adjust the intensity of pseudo-color and masking light to make the images clearer and more realistic; (e) use Fly-around technology to assist in the determination of suspected aneurysms <2.0 mm in diameter; (f) multi-angle rotation of the image may clearly and completely show the three-dimensional spatial relationship between the aneurysm neck and the aneurysm body, the aneurysm-carrying artery and the surrounding vessels (g) For posterior communicating aneurysms, 3D-MRA is also feasible to better reveal the full picture of the aneurysm without the interference of the skull base bone.
The main factors affecting the quality of cerebral artery CTA post-processing images are
a) Data acquisition layer thickness: thin layer (<3mmb) data acquisition can improve its resolution. b) Contrast agent dose: appropriate contrast agent dose (about 100ml) can ensure higher contrast concentration in blood vessels, making the image of blood vessels, especially fine vessels, clearer and more realistic. c) Contrast agent injection rate: injection rate should be >3.0ml/s to avoid contrast agent in blood vessels during sweeping The contrast agent should be diluted by blood flow during the scan to keep its concentration at a high peak. d) Delay time: It is the key to the success or failure of data acquisition. If the scan is started too early, the contrast agent in the vessel has not yet reached its peak and is not adequately mixed with the blood; conversely, the contrast agent is diluted by the blood flow and enters the veins and perivascular tissues excessively, thus affecting the imaging quality of the target vessel. e) Cardiac output per beat and circulation time: There are individual differences in cardiac function and circulation time, and the optimal delay time may also vary. Therefore, the patient’s cardiac function status should be understood before the scan plan is made so that the delay time can be adjusted according to the specific situation. f) Shoulder bone artifacts: The supra-arch branch vessels are more affected by shoulder bone artifacts. Therefore, the RASP parameters should be selected in the scan plan to remove the interference of bone artifacts.
Urinary system
VR images can clearly show the complete morphology of the kidney, calyces and pelvis enhanced by contrast, as well as the course and obstruction of the entire ureter, the site and degree of stenosis, and can visualize the anatomical relationship between the kidney, ureter and surrounding vessels and bone in multiple angles.
VR can show the renal pelvis, ureter, and bladder by debridement, shearing, and rotation in the examination of urological diseases, as well as preserve the spine and pelvis, and also distinguish the organs and bones of the urinary system with different colors. The tumor application VR Multi-Threshold values Curve technology can obtain images of bone, blood vessels and soft tissues simultaneously on the same 3D image of tumors of various systems and organs enhanced by contrast, which can accurately locate the tumor and show the state of the lesion itself as well as the adjacent relationship with surrounding tissues, organs and blood vessels and the invaded and extruded displacement. The processed images can be rotated at any angle for multi-directional observation and analysis of the lesion. In order to clearly display the hidden part of the lesion, the images can be cropped, cut, drilled and made into automatic movies to provide clinicians with richer imaging information for making correct judgments about the disease.
Acquisition data requirements: a) acquisition layer thickness should be appropriately selected according to different parts and lesion size (general layer thickness should be less than 3.0mm/per layer); b) delayed scan time should be determined according to tumor blood supply; c) reconstruction function should be selected as FC10/43; d) overlapping reconstruction should be used. Key points of image post-processing techniques: a) accurate adjustment of multi-curves; b) setting pseudo-color for CT values of different tissues C) MPR images should be referred to for complex anatomical structures or small lesions.
Density volume reconstruction (IVR)
IVR images use the depth and transmittance information of all body elements for imaging and are mainly suitable for observing tissues and organs with small differences in CT values in the abdomen and lungs. The acquisition data requirements and the technical points of image post-processing are the same as those of SVR; the technical points of image post-processing: 1) accurate adjustment of multi-curves; 2) appropriate adjustment of window width and window position. ivr images show the relationship between bronchi and lung tumor, and in this case, although no enhancement was performed, the bronchi and tumor tissues in the lung were still clearly shown through image post-processing, and the tumor bronchi could be seen to be The relationship is close.
Maximum density projection (MIP)
MIP is one of the projection techniques that uses all the image values of the volume data with the highest density in the line of sight direction. Because the imaging data is derived from the 3D volume data, the direction of projection can be changed at will; because the imaging data is taken from the most dense image element value in the 3D volume data, its main advantage is that it can reflect the density difference of tissues more realistically, and clearly and precisely show the shape, alignment, abnormal changes and calcification of the vessel wall and the distribution range of the blood vessels enhanced by contrast, and the normal dynamics of long bones, short bones, flat bones, etc., as well as fractures, tumors, and bones. It is also very sensitive to the normal dynamics of long bones, short bones, flat bones, and changes in bone density caused by fractures, tumors, osteoporosis and other lesions. In addition, it is particularly useful for the display and localization of abnormal high-density foreign bodies in vivo. The disadvantages of MIP are that it is not possible to obtain valuable images of complex anatomical sites with close densities and overlapping structures; the images lack a sense of spatial depth, and it is difficult to show the three-dimensional spatial relationships between arteries and veins and between them and the skull with complex intracranial travel. The main way to overcome these shortcomings is to use techniques such as Clipping, Cutting, Seed or Segmentation to remove the interference of tissue images other than the target organ and to rotate the image at an appropriate angle. In the same case, the VR image shows the stone less clearly than the MIP image, and the MIP image shows the iliac artery calcification more clearly than the VR image.
Minimal density projection (Min-IP) Min-IP is a projection technique that uses the least dense image element value in the direction of the line of sight in volumetric data for imaging. Since the CT value of the airway and the specially treated (cleaned and inflated) gastrointestinal tract, etc. is the lowest (-1000 HU) among the tissues and organs in the human body, Min-IP is mainly used to show lesions in hollow organs such as the large airway, bronchial tree and gastrointestinal tract. Key points of image post-processing techniques: appropriate cutting of the image with Clipping in order to remove the overlapping interference of bone and soft tissue images around the target organ; 2) appropriate adjustment of the window width and window position to clearly show the lesions in the hollow organs and the contrast relationship with the surrounding tissues.