Skripsi

PENGGUNAAN CONDITIONAL GENERATIVE ADVERSARIAL NETWORK (CGAN) UNTUK PENINGKATAN KONTRAS CITRA MEDIS ULTRASONOGRAFI DINDING PERUT BERBASIS KETERBATASAN DATA BERLABEL

---

This study implements a Conditional Generative Adversarial Network (CGAN) to enhance the contrast of fetal abdominal ultrasound images. The novelty lies in increasing the Reconstruction Loss weight (λL1) to 1000 to ensure structural stability in medical data. Using U-Net and PatchGAN architectures, the model was trained against CLAHE targets. The results demonstrate significant enhancement with an average PSNR of 40.4 dB and SSIM of 0.9807, vastly outperforming the raw image baseline (PSNR 21.63 dB; SSIM 0.57). This research, implemented via a Gradio-based web application, proves that L1 Loss dominance in CGAN effectively produces sharper ultrasound images while preserving essential structural integrity. Keywords: Conditional GAN, Ultrasound, Contrast Enhancement, PSNR, SSIM, L1 Loss.

---

Availability

Detail Information

Series Title

-

Call Number

T1958272026

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xxiv, 103 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.370 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Pengenalan pola citra

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment