Skripsi

DETEKSI STRUKTUR JANTUNG PADA ANAK DENGAN VIDEO ULTRASOUND MENGGUNAKAN DEEP LEARNING  

---

Medical imaging, such as ultrasound, is widely used in various diagnoses and medical treatments due to its advantages, including being a comfortable, non-invasive procedure, radiation-free, affordable, and capable of real-time observation. Therefore, the application of deep learning is expected to be used for object detection in children's hearts. This study employs YOLOv5, YOLOv7, and YOLOv8 models based on CNN architecture to detect cardiac structures in children. In this research, YOLOv7 proved to have the best overall performance. On Dataset 1, YOLOv7 recorded the highest precision (0.98), recall (0.914), and mAP50 (0.943), although its mAP50-95 was slightly lower compared to other models. On Dataset 2, YOLOv7 again excelled with precision (0.931), recall (0.92), and mAP50 (0.922), although its mAP50-95 remained slightly lower. Meanwhile, YOLOv8 outperformed in recall and precision, but its mAP50 and mAP50-95 were lower compared to YOLOv7. YOLOv5 showed better performance in precision and mAP50-95 compare

---

Availability

Detail Information

Series Title

-

Call Number

T1582912024

Publisher

Indralaya : Program Magister Ilmu Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xvii, 103 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

616.075 707

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Program Magister Ilmu Komputer
Ultrasonografi — Diagnostik jantung anak

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment