Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Science Midwifery

Analysis of axial T2 TSE images using deep learning reconstruction in MRI of brain tumors Muzdalifah, Nadifah Pratiwi; Utami, Hernastiti Sedya; Hidayat, Fathur Rachman; Wibowo, Kusnanto Mukti; Jadmika, Muhammad Riefki; Samudra, Alan
Science Midwifery Vol 13 No 1 (2025): April: Health Sciences and related fields
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/midwifery.v13i1.1867

Abstract

Magnetic Resonance Imaging (MRI) Brain examinations often encounter uncooperative patients, necessitating rapid scanning techniques that yield optimal results. To address this challenge, advanced technologies such as deep learning can be leveraged to accelerate scan time, reduce noise, and enhance image precision. This study aims to evaluate the disparity in MRI Brain image quality with and without deep learning in tumor cases to achieve superior diagnostic imaging. Employing a quantitative experimental approach, this research analyzed a sample of 30 patients collected from January to February 2025. Three Radiologist Specialists assessed the images using a questionnaire based on the Visual Grading Analysis (VGA) method. The obtained responses were statistically examined through Cohen’s Kappa consistency test and Wilcoxon Signed-Rank Test. Findings revealed a statistically significant difference in image information between deep learning-assisted and conventional MRI scans. In T2 TSE sequences, deep learning reconstruction demonstrated superior anatomical visualization of the Gray Matter, White Matter, Lateral Ventricles, Basal Ganglia, and Parafalx Cerebri. However, in brain tumor pathology visualization, conventional MRI exhibited sharper and more distinct tumor delineation. Although deep learning-enhanced T2 TSE sequences reduced scan duration and improved overall image quality, they provided suboptimal diagnostic information in tumor cases.
Comparison of image quality between flexible coil and special purpose coil sagittal section on mri manus digit 2-3 Verina, Alya; Utami, Hernastiti Sedya; Susanto, Fani; Hidayat, Fathur Rachmat; Syafi’ie, Mochammad; Samudra, Alan
Science Midwifery Vol 13 No 1 (2025): April: Health Sciences and related fields
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/midwifery.v13i1.1912

Abstract

MRI examination of the manus has a challenge in obtaining optimal image quality. Good image quality depends on four characteristics, one of which is the Signal to Noise Ratio (SNR). SNR can be affected by the use of appropriate radiofrequency (RF) coils so that it can increase the SNR value. This study aims to analyze the impact of using different coils on SNR and CNR values, especially in sagittal sections of Proton Density Fat Sat scans. This study is a quantitative study using 10 volunteer samples in the period of February 2025. The results of the study show that the SNR value in the overall anatomy has a p-value <0.05 which means there is a significant difference using flexible coils and special purpose coils. While the SNR value in per-anatomy manus digits 2-3 has a significant difference except for the volar plate anatomy with a p-value = 0.121 meaning there is no significant difference. And compared to flexible coil, the use of special purpose coil in this study has a higher CNR value, especially in the anatomy of the Distal Interphalangeal Joint - Middle Phalanges with an average value of 1221.200. This study provides evidence that the selection of the right coil greatly affects the quality of the resulting image, and the use of special purpose coil is considered to produce better image quality. Therefore, the author recommends the use of special purpose coils in MRI examinations of the manus digits 2-3.