The automated measurement of CT number linearity using an ACR accreditation phantom.

We developed a software to automatically measure the linearity between the CT numbers and densities of objects using an ACR 464 CT phantom, and investigated the CT number linearity of 16 different CT scanners. The software included a segmentation-rotation method. After segmenting five objects within the phantom image, the software computed the mean CT number of each object and plotted a graph between the CT numbers and densities of the objects. Linear regression and coefficients of regression, R2, were automatically calculated. The software was used to investigate the CT number linearity of 16 CT scanners from Toshiba, Siemens, Hitachi, and GE installed at 16 hospitals in Indonesia. The linearity of the CT number obtained on most of the scanners showed a strong linear correlation (R2> 0.99) between the CT numbers and densities of the five phantom materials. Two scanners (Siemens Emotion 16) had the strongest linear correlation withR2= 0.999, and two Hitachi Eclos scanners had the weakest linear correlation withR2< 0.99.

Comparative estimation of pleural effusion volume based on lateral decubitus position of chest x-ray and CT scan imaging.

Previous study using thoracic phantom for estimating fluid volume has been obtained which represents the case of pleural effusion based on the size of the x-ray radiograph. The models are obtained in the form of three equations, the pleural effusion volume as a function of height, length times the height, and area of the radiograph image. The three models of estimation have high linearity with ratio value more than 0.988, higher than the modelling measurement using ultrasonography modality. The modelling is expected to give a contribution on developing method for helping clinicians estimate the pleural effusion volume as a basic for performing fluid aspiration and to monitor the therapy. However, because modelling is developed using phantoms, then to be applied clinically, further research is needed for its application to patients. The height function model yields correlation value of 0.966 and paired T-test value of 0.892. The height times length function model yields correlation value of 0.982 and paired T-test value of 0.611. The area function model yields correlation value of 0.997 and paired T-test value of 0.647. From the three equations, measurement of estimated pleural effusion volume using area function on chest x-ray lateral decubitus position is the most appropriate equation. Corresponding to the results of the measurement of gold standard using a CT scan. Height measurement is the measurement that is the fastest and easiest in the application. Limitation of the study is it only can be done in right lateral decubitus position of the patient, and also cannot be applied to patients with condition such as post lung surgery, massive subpulmonic/ supradiaphragmatic pleural effusion, empyema, an atypical pleural effusion such as septated, encapsulated, loculated pleural effusion and anatomical deformity, scoliosis, or abnormalities of thoracic cavity.

An Improved Method of Automated Noise Measurement System in CT Images.

BACKGROUND: It is necessary to have an automated noise measurement system working accurately to optimize dose in computerized tomography (CT) examinations. OBJECTIVE: This study aims to develop an algorithm to automate noise measurement that can be implemented in CT images of all body regions. MATERIALS AND METHODS: In this retrospective study, our automated noise measurement method consists of three steps as follows: the first is segmenting the image of the patient. The second is developing a standard deviation (SD) map by calculating the SD value for each pixel with a sliding window operation. The third step is estimating the noise as the smallest SD from the SD map. The proposed method was applied to the images of a homogenous phantom and a full body adult anthropomorphic phantom, and retrospectively applied to 27 abdominal images of patients. RESULTS: For a homogeneous phantom, the noises calculated using our proposed and previous algorithms have a linear correlation with R2 = 0.997. It is found that the noise magnitude closely follows the magnitude of the water equivalent diameter (Dw) in all body regions. The proposed algorithm is able to distinguish the noise magnitude due to variations in tube currents and different noise suppression techniques such as strong, standard, mild, and weak ones in a reconstructed image using the AIDR 3D algorithm. CONCLUSION: An automated noise calculation has been proposed and successfully implemented in all body regions. It is not only accurate and easy to implement but also not influenced by the subjectivity of user.