Skripsi

DETEKSI RUANG JANTUNG JANIN PADA PANDANGAN 3-VESSEL TRACHEA MENGGUNAKAN ARSITEKTUR FASTER RCNN

---

Penilaian

---

The heart chambers are divided into four different chambers, including the Left Atrium (LA), Right Atrium (RA), Left Ventricle (LV) and Right Ventricle (RV). The heart chamber itself can be seen visually using an Ultrasonography (USG) medical device. On ultrasound, the fetal heart has several points of view, including Four Chamber View (4CV), Left Ventricular Outflow Tract (LVOT), Right Venticular Outflow Tract (RVOT), and Three Vessel of Trachea (TVT). In the Three Vessel of Trachea (TVT) view, the images displayed are Ductus Arteriosus (DUCT), Superior Vena Cava (SVC), Aorta (AoA) and Trachea which are detected by using Deep Learning with Faster-RCNN architecture. Faster R-CNN is one method that is often used in object detection. Faster R-CNN consists of a combination of the Fast R-CNN method and the Region Proposal Network (RPN). In this study, there were two detection cases, namely using selection data and data without selection. The number of models from all cases is 14 for tuning the learning rate parameters, the number of epochs and the best backbone. Of the 14 models tested, the best model was obtained using data without selection and Backbone VGG16 learning rare 0.001 epoch 50. The average mean average precision (MAP) obtained with the best model was 87.55%

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2107002386	T509862	T509862021	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T509862021

Publisher

Inderalaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Uniersitas Sriwijaya., 2021

Collation

xvii, 96 hlm,: ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

005.707

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Komputer
Data Sistem Komputer

Specific Detail Info

-

Statement of Responsibility

MURZ

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment