Komparasi Algoritma Machine Learning Untuk Mendeteksi Citra Histopatologi Kanker

Zahro, Savina (2024) Komparasi Algoritma Machine Learning Untuk Mendeteksi Citra Histopatologi Kanker. Undergraduate thesis, Politeknik Negeri Jember.

	Text (Abstract) Abstract.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (8kB)
	Text (Bab 1 Pendahuluan) Bab 1 Pendahuluan.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (89kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (155kB)
	Text (Laporan Lengkap) Laporan Lengkap.pdf - Submitted Version Restricted to Registered users only Download (13MB) | Request a copy

Abstract

The study compares the performance of machine learning algorithms for cancer classification in histopathology images. The algorithms compared Gray Level Co-Occurrence Matrix (GLCM) + Learning Vector Quantization (LVQ) with two feature selection is Principal Component Analysis (PCA) and regression analysis, as well as Convolution Neural Network (CNN) with a modified ResNet50V2 architecture. Evaluation was carried out on breast and cervical cancer datasets (train 90%, test 10% for GLCM+LVQ, and train 70%, valid 20%, test 10% for CNN). The results showed that the algorithm with the ResNet50V2 CNN model was superior in cancer histopathology classification compare to GLCM+LVQ. CNN ResNet50V2 achieved 87% accuracy for histopathology images of breast and 95% for cervical cancer, while GLCM+LVQ with PCA feature selection only achieved 57,5% for breast and 65% for cervix. ResNet50V2 CNN can be a more effective choice for cancer classification in histopathology images

Actions (login required)

View Item