Image of PERANCANGAN USER INTERFACE BERBASIS WEBSITE DENGAN MENGIMPLEMENTASIKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK MENDETEKSI PENYAKIT TUBERKULOSIS MENGGUNAKAN CITRA CHEST X-RAY.

Skripsi

PERANCANGAN USER INTERFACE BERBASIS WEBSITE DENGAN MENGIMPLEMENTASIKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK MENDETEKSI PENYAKIT TUBERKULOSIS MENGGUNAKAN CITRA CHEST X-RAY.

Penilaian

0,0

dari 5
Penilaian anda saat ini :  

Convolutional Neural Network (CNN) is a part of deep learning commonly used for image classification. There are many studies that utilize CNN for classifying tuberculosis and covid-19, as well as normal conditions, using chest x-ray images. However, these studies are still rarely implemented on Indonesian data. Furthermore, the CNN models built have not been deployed in the form of a user interface that can be used by health workers. In this study, three CNN architectures, namely AlexNet, LeNet, and a modified architecture, are used to classify tuberculosis, covid-19, and normal conditions by training them on a dataset that combines Indonesia and Kaggle datasets. The results show that the AlexNet architecture is the best architecture with the highest accuracy of 97.52% on the Kaggle dataset, 64.45% for the RSUP dr. Rivai Abdullah dataset, and 92.43% for the combined dataset. This model is then used for deployment in a user interface. During testing using new data from RSUP dr. Rivai Abdullah, the model embedded in the website was able to detect 7 out of 10 new data with an accuracy percentage of 70%. The web-based user interface, built using the Gradio library, is capable of providing an initial diagnosis for patients to assist medical staff.

Availability
Inventory Code Barcode Call Number Location Status
2307002703T116859T1168592023Central Library (Referens)Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1168592023
Publisher
Indralaya : Prodi Teknik Elektro, Fakultas Teknik Universitas Sriwijaya ;.,
Collation
vii, 68 hlm.; ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
621.381 07
Content Type
Text
Media Type
unmediated
Carrier Type
-
Edition
-
Subject(s)
Teknik Elektronika
Prodi Teknik Elektro
Specific Detail Info
-
Statement of Responsibility
Other version/related

No other version available

File Attachment
Comments
You must be logged in to post a comment