Skripsi

SEGMENTASI POLIP KOLONOSKOPI REAL-TIME UNTUK DETEKSI DINI KANKER KOLOREKTAL DENGAN METODE YOLOV11-SEG

---

Colorectal cancer is a leading cause of cancer-related deaths globally, predominately developing from adenomatous polyps. While colonoscopy is the gold standard for polyp detection, it suffers from a miss rate of 26% due to operator fatigue and visual variability. This study aims to develop a real-time Computer-Aided Diagnosis (CADx) system for polyp segmentation using the efficient YOLOv11-Seg method. The research evaluated 12 model configurations across different scales (Nano, Small, Medium, Large) and input resolutions on the Kvasir-SEG dataset, utilizing the Rational Unified Process (RUP) methodology. Evaluation results identified the YOLOv11s-Seg (Small variant) with an input resolution of 640x640 pixels as the optimal configuration. This model achieved high segmentation quality with a mask mAP 0.5 of 0.97, Dice Coefficient of 0.90, and IoU of 0.84. In addition to precise accuracy, the system attained an average inference speed of 53.89 FPS (18.55 ms/frame) in a GPU environment, significantly exceeding the standard clinical endoscopy video rate (25–30 FPS). The system was implemented into a web-based interface using Streamlit, demonstrating its feasibility as an accurate and responsive real-time early diagnostic tool.

---

Availability

Detail Information

Series Title

-

Call Number

T1897302025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xiv, 90 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

616.994 07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Deteksi Kanker Kolorektal

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment