Muhammad Hafid Giofanny
Poltekkes Kemenkes Semarang

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

ANALISIS INFORMASI CITRA DIAGNOSTIK TEKNIK PARALLEL IMAGING GRAPPA DENGAN DAN TANPA ARTIFICIAL INTELLIGENT (AI) DEEP RESOLVED: Studi pada MRI Lumbal T2 TSE Potongan Sagital pada Kasus HNP Muhammad Hafid Giofanny; Sugiyanto Sugiyanto
Jurnal Imejing Diagnostik (JImeD) Vol. 12 No. 2 (2026): JULY 2026
Publisher : Poltekkes Kemenkes Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31983/jimed.v12i2.15195

Abstract

Background: Hernia Nucleus Pulposus (HNP) is commonly evaluated using lumbar MRI, with sagittal T2 TSE sequences serving as a standard imaging protocol. However, reducing scan time with GRAPPA parallel imaging may decrease image quality. Artificial Intelligence (AI) Deep Resolved has been developed to improve image resolution and signal quality without increasing acquisition time. Therefore, this study evaluated the effect of AI Deep Resolved combined with GRAPPA on the diagnostic image quality of lumbar MRI in HNP cases. Methods: This quantitative experimental study compared the image quality of Lumbar MRI T2 TSE sagittal sequences acquired using conventional GRAPPA and GRAPPA combined with AI Deep Resolved in detecting HNP. Sixteen patients at RS Mardi Rahayu Kudus underwent both imaging protocols using identical parameters. Two radiologists evaluated the images using a 3-point scoring system. Data were analyzed using Cohen’s Kappa test for inter-observer agreement and the Wilcoxon test to compare image quality between the two techniques. Results: The results showed that GRAPPA combined with AI Deep Resolved produced significantly better diagnostic image quality than conventional GRAPPA alone (p < 0.05). All evaluated anatomical structures, including HNP, demonstrated higher image quality scores with AI assistance. The mean rank score was higher for the AI-assisted technique (91.50) than for conventional GRAPPA (0.00), indicating superior image quality. Scan time remained unchanged (1 minute 24 seconds), and inter-observer agreement was excellent (Cohen’s Kappa = 0.886). Conclusions: The integration of AI Deep Resolved with the GRAPPA parallel imaging technique improves the quality of Lumbar MRI T2 TSE sagittal images by providing clearer visualization of anatomical structures and HNP lesions. This method produces better and more detailed images than conventional GRAPPA without increasing scan time, thereby supporting more accurate diagnosis. Keywords: MRI Lumbal ; T2 TSE ; GRAPPA ; Artificial Intelligent (AI) Deep Resolved ; Diagnostic Image Information