Deteksi Multiview Kanker Payudara Dari Citra Mamografi Menggunakan Multiview Convolutional Neural Network

Anggraini, Sisilia (2023) Deteksi Multiview Kanker Payudara Dari Citra Mamografi Menggunakan Multiview Convolutional Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember Surabaya.

Text 07311940000006-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (6MB) | Request a copy

Abstract

Deteksi dini kanker payudara melalui citra mammografi sangat penting untuk meningkatkan tingkat kesembuhan dan mengurangi angka kematian. Penelitian sebelumnya telah mengembangkan deteksi kanker payudara dari citra tampilan Craniocaudal (CC) atau Mediolateral Oblique (MLO), namun masih dilakukan secara terpisah. Oleh karena itu, penelitian ini bertujuan untuk mengembangkan sistem deteksi kanker payudara berbasis komputer menggunakan citra mammografi dari tampilan CC dan MLO secara bersamaan. Dengan mengintegrasikan informasi dari kedua tampilan, diharapkan dapat meningkatkan akurasi deteksi. Metode deteksi yang digunakan adalah Multiview Convolutional Neural Network (MVCNN) yang dirancang khusus untuk menerima tampilan CC dan MLO. Sistem yang dikembangkan untuk mengenali pola dan fitur dari kedua tampilan sehingga dapat memberikan informasi berupa hasil deteksi normal atau abnormal. Untuk meningkatkan efektivitas sistem deteksi, dilakukan prapemrosesan citra yang meliputi tahap background removal, image enhancement, dan pectoral muscle removal. Pengujian sistem deteksi dilakukan dengan membandingkan pembelajaran pada tiga jenis dataset yang telah dibentuk. Hasil penelitian menunjukkan bahwa pembelajaran pada dataset kedua, tanpa menggunakan pectoral muscle removal menghasilkan hasil yang optimal yaitu akurasi deteksi sebesar 98,63%, presisi sebesar 97,29%, sensitivitas 100%, dan spesifisitas 97,29%. Dengan menggabungkan metode MVCNN dengan dataset yang efektif, sistem deteksi yang dikembangkan dapat membantu meningkatkan efisiensi dan akurasi dalam proses diagnosis kanker payudara. Sehingga berpotensi untuk meningkatkan kesembuhan dan mengurangi angka kematian akibat kanker payudara melalui deteksi dini yang lebih efektif.
===============================================================================================================================
Early detection of breast cancer through mammography images is crucial to improve survival rates and reduce mortality. Previous studies have developed breast cancer detection from Craniocaudal (CC) or Mediolateral Oblique (MLO) view images, but they have been performed separately. Therefore, this study aims to develop a computer-based breast cancer detection system using mammography images from both CC and MLO views simultaneously. By integrating information from both views, it is expected to improve detection accuracy. The detection method used is the Multiview Convolutional Neural Network (MVCNN) specifically designed to handle CC and MLO views. The developed system recognizes patterns and features from both views to provide information regarding normal or abnormal detection outcomes. To enhance the effectiveness of the detection system, image pre-processing is conducted, including background removal, image enhancement, and pectoral muscle removal stages. The detection system is tested by comparing the learning performance on three types of formed datasets. The research results show that learning on the second dataset, without using pectoral muscle removal, yields optimal outcomes with a detection accuracy of 98.63%, precision of 97.29%, sensitivity of 100%, and specificity of 97.29%. By combining the MVCNN method with an effective dataset, the developed detection system can help improve efficiency and accuracy in breast cancer diagnosis processes. Thus, it has the potential to enhance survival rates and reduce breast cancer-related mortality through more effective early detection.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Deteksi Multiview, Kanker Payudara, Prapemrosesan Citra Mamografi, Multiview Convolutional Neural Network, CC dan MLO, Multiview Detection, Breast Cancer, Mammography Image Pre-processing, Multiview Convolutional Neural Network, CC and MLO
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Sisilia Anggraini
Date Deposited:	29 Aug 2023 04:07
Last Modified:	29 Aug 2023 04:07
URI:	http://repository.its.ac.id/id/eprint/103425

Actions (login required)

View Item