Implementasi Automatic Window Level Pada Citra T2 MRI Menggunakan Contrast Limited Adaptive Histogram Equalization

Romziah, Nur Jazilatur (2026) Implementasi Automatic Window Level Pada Citra T2 MRI Menggunakan Contrast Limited Adaptive Histogram Equalization. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5001221004-Undergraduate_thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Citra Magnetic Resonance Imaging (MRI) T₂-weighted seringkali memberikan kontras yang kurang optimal karena keterbatasan sistem akuisisi dan inhomogenitas intensitas, sehingga perlu peningkatan kualitas citra. Penelitian ini mengkaji pengaruh metode automatic window level (AWL) dan contrast limited adaptive histogram equalization (CLAHE) terhadap kualitas citra T₂-weighted menggunakan fantom gel yang diakuisisi pada MRI GE SIGNA Pioneer 3 Tesla dengan variasi echo time (TE) dari 75 ms hingga 200 ms pada dua nilai repetition time (TR), yaitu 1000 ms dan 1500 ms. Evaluasi kualitas citra dilakukan secara kuantitatif menggunakan signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), structural similarity index measure (SSIM), dan entropi Shannon. Hasil penelitian menunjukkan bahwa nilai SNR dan CNR tertinggi diperoleh pada citra asli (raw), diikuti AWL, dan terendah pada CLAHE. Nilai SSIM AWL terhadap citra asli (raw) lebih tinggi dibandingkan CLAHE yang menunjukkan bahwa CLAHE mempengaruhi perubahan struktural yang lebih besar terhadap referensi. Nilai entropi mengikuti pola yang sama, dengan raw, AWL, dan CLAHE. Tren nilai SNR dan CNR terhadap peningkatan TE secara umum konsisten dengan teori peluruhan sinyal T₂ pada variasi TE dari 75 ms hingga 250 ms, meskipun ditemukan anomali pada beberapa nilai TE yang dipengaruhi oleh beberapa hal. Secara visual, CLAHE menghasilkan citra dengan kontras lokal paling tajam dan detail objek paling terlihat, meskipun disertai peningkatan noise pada area latar belakang (background).
======================================================================================================================================
T₂-weighted Magnetic Resonance Imaging (MRI) images often exhibit suboptimal contrast due to limitations of the acquisition system and intensity inhomogeneity, making image enhancement necessary. This study investigated the effect of combining automatic window level (AWL) and contrast limited adaptive histogram equalization (CLAHE) on the quality of T₂-weighted MRI images acquired using a gel phantom on a 3 Tesla GE SIGNA Pioneer MRI system. The images were acquired with echo time (TE) variations ranging from 75 ms to 200 ms at two repetition time (TR) values, namely 1000 ms and 1500 ms. Image quality was quantitatively evaluated using signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), structural similarity index measure (SSIM), and Shannon entropy. The results showed that the highest SNR and CNR values were obtained from the original (raw) images, followed by the AWL processed images, while the lowest values were observed in images processed with the CLAHE method. The SSIM values of AWL relative to the original images were higher than those of the CLAHE images, indicating that CLAHE introduced greater structural changes with respect to the reference image. Shannon entropy exhibited a similar pattern, with the highest values observed in the original images, followed by AWL and CLAHE images. The trends of SNR and CNR with increasing TE were generally consistent with the theoretical T₂ signal decay at TE variations from 75 ms to 200 ms, although anomalies were observed at several TE values due to multiple contributing factors. Visually, the CLAHE method produced images with the highest local contrast and the most distinct structural details, although accompanied by increased background noise.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Automatic Window Level, Contrast Limited Adaptive Histogram Equalization, T₂ MRI, Fantom gel. Automatic Window Level, Contrast Limited Adaptive Histogram Equalization, T₂ MRI, Gel Phantom.
Subjects:	Q Science > QC Physics R Medicine > RC Internal medicine > RC78.7.N83 Magnetic resonance imaging.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Nur Jazilatur Romziah
Date Deposited:	03 Aug 2026 03:24
Last Modified:	03 Aug 2026 03:24
URI:	http://repository.its.ac.id/id/eprint/141697

Actions (login required)

View Item