Pengaruh Tegangan Tabung terhadap Kualitas Citra pada Pemeriksaan Computed Tomography Kepala menggunakan Iterative Reconstruction

Abstract

Iterative reconstruction can optimize radiation dose and improve image quality on CT scan. This research method is quantitative analytic with the analysis of the results of the head CT examination parameters associated with image quality to changes in variations of 80 kV, 100 kV, 120 kV with the use of iterative reconstruction. Image quality measurements are the Hounsfield Unit (HU) value, standard deviation, and Signal to Noise Ratio (SNR) using Radiant Viewers. Effective dose measurement using the Dose Length Product (DLP). Then perform the Kruskal Wallis test to find out whether there is an effect of tube voltage and Iterative Reconstruction on the SNR value using IBM SPSS version 24. The results image quality of the HU value increase due to changes in the kV value, but the value does not change significantly when the iDose changes, for the standard The deviation has decreased due to changes in kV, but the value of the value does not experience a significant change at the time of change in iDose, while SNR increases due to changes in kV value and changes in iDose. The percentage ratio of the effective dose in the use of standard kV with 80 kV decreased radiation dose by 62%, while at 100 kV there was a decrease of 25%, and the use of 120 kV experienced an increase of 25%. The results of the Kruskal Wallis test p-value <0.001, therefore it can be concluded that there is a difference in the SNR image quality at each change in iDose and kV parameters.

Analysis and Improvement of CT-Scan/MRI Medical Image Quality in Stroke Patients with Hybrid Thresholding Method: A Case Study in Padang BMC Hospital

“…The automatic segmentation method can be done by depicting the bleeding stroke through CT Scan or MRI. Some studies only discuss the problems encountered in segmenting CT scans and MRI images and the relative benefits and limitations of currently available methods for segmenting medical images [15], [16], [17]. Deep learning-based medical image segmentation is still experiencing difficulties in several studies.…”