Klasifikasi Penyakit Stroke Menggunakan MobileNetV2 dan Support Vector Machine (SVM) Berdasarkan CItra CT Scan Otak

Devani, Ajeng Listya (2025) Klasifikasi Penyakit Stroke Menggunakan MobileNetV2 dan Support Vector Machine (SVM) Berdasarkan CItra CT Scan Otak. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (COVER) 21081010220.-cover.pdf Download (3MB)
	Text (BAB I) 21081010220.-bab1.pdf Download (330kB)
	Text (BAB II) 21081010220.-bab2.pdf Restricted to Repository staff only until 5 December 2028. Download (1MB)
	Text (BAB III) 21081010220.-bab3.pdf Restricted to Repository staff only until 5 December 2028. Download (1MB)
	Text (BAB IV) 21081010220.-bab4.pdf Restricted to Repository staff only until 5 December 2028. Download (2MB)
	Text (BAB V) 21081010220.-bab5.pdf Download (322kB)
	Text (DAFTAR PUSTAKA) 21081010220.-daftarpustaka.pdf Download (331kB)
	Text (LAMPIRAN) 21081010220.-lampiran.pdf Restricted to Repository staff only until 2028. Download (1MB)

Abstract

Stroke is one of the leading causes of death and disability worldwide, requiring rapid and accurate diagnosis to determine appropriate medical treatment. The process of identifying stroke types through brain CT scan images often faces challenges due to the similarity of tissue structures among normal, ischemic, and hemorrhagic (Bleeding) conditions. In addition, the limited amount of data and variations in image quality from different CT scan devices affect the performance of machine learning–based classification systems. This study developed a brain CT scan image classification system using a combination of MobileNetV2 and Support Vector Machine (SVM). The MobileNetV2 architecture was employed as a feature extractor to obtain efficient image representations, while SVM served as the classification algorithm to improve accuracy and computational efficiency. The dataset was collected from RSUD Haji Surabaya and the public Kaggle repository, with data balance achieved through data augmentation using the Augmentor library. Evaluation was carried out on three different models, namely MobileNetV2-SVM, MobileNetV2, and SVM. The experimental results showed that the MobileNetV2-SVM model achieved the highest accuracy of 97.09%, followed by MobileNetV2 with 84.66%, and SVM with 63.21%. These findings indicate that the combination of MobileNetV2 and SVM significantly enhances classification performance in distinguishing the three stroke categories: normal, ischemic, and hemorrhagic. The resulting model has the potential to be further developed into an image-based diagnostic support system to assist healthcare professionals in rapid, efficient, and accurate stroke detection.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Sari, Anggraini Puspita	NIDN0716088605	anggraini.puspita.if@upnjatim.ac.id
Thesis advisor	Nurlaili, Afina Lina	NIDN0013129303	afina.lina.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ajeng Listya Devani

Date Deposited:

08 Dec 2025 01:26

Last Modified:

08 Dec 2025 01:26

URI:

https://repository.upnjatim.ac.id/id/eprint/48100