KLASIFIKASI JENIS TUMOR OTAK (MENINGIOMA, GLIOMA, DAN PITUITARY) DENGAN METODE CONVOLUTIONAL NEURAL NETWORK BERDASARKAN HASIL SCAN MRI

Fajar, Adnan Nur (2021) KLASIFIKASI JENIS TUMOR OTAK (MENINGIOMA, GLIOMA, DAN PITUITARY) DENGAN METODE CONVOLUTIONAL NEURAL NETWORK BERDASARKAN HASIL SCAN MRI. Other thesis, UPN “VETERAN” YOGYAKARTA.

[thumbnail of ABSTRAK.pdf]
Preview
Text
ABSTRAK.pdf

Download (49kB) | Preview
[thumbnail of Cover.pdf]
Preview
Text
Cover.pdf

Download (168kB) | Preview
[thumbnail of DAFTAR ISI.pdf]
Preview
Text
DAFTAR ISI.pdf

Download (218kB) | Preview
[thumbnail of Daftar Pustaka.pdf]
Preview
Text
Daftar Pustaka.pdf

Download (470kB) | Preview
[thumbnail of Skripsi-123160089.pdf] Text
Skripsi-123160089.pdf
Restricted to Repository staff only

Download (5MB)

Abstract

ABSTRAK
Penelitian ini akan membantu model klasifikasi multiclass jenis tumor pada otak . Pada
penelitian ini akan menggunakan 4 kelas klasifikasi yaitu 3 jenis tumor otak yaitu meningioma,
glioma dan pituitary serta otak normal. Dengan penelitian ini akan mengetahui keberhasilan dari
suatu model CNN yang dibangun untuk klasifikasi jenis tumor otak berdasarkan dari hasil scan
MRI dengan penambahan preprocessing teknik Cropping. Pengujian menunjukan metode CNN
dapat melakukan klasifikasi jenis tumor otak dengan cukup baik .Hasil penelitian menunjukan
bahwa dengan ditambahkan cropping pada citra mampu meningkatkan akurasi pada pengujian
Confusion Matrix.
Pengujian confusion matrix dengan data test sebanyak 630 menunjukan dengan preprocessing
cropping didapatkan rata-rata akurasi 83 %, presisi 83 %, recall 82 %,dan F1-score 82% sedangkan
tanpa cropping didapatkan rata-rata hasil akurasi 61%, presisi 72 %, dan recall 58 %,dan F1-score
58%.
Kata Kunci : klasifikasi, tumor otak ,CNN,cropping ,confusion matrix

ABSTRACT
This research will help model the multiclass classification of brain tumors . In this study, we
will use 4 classification classes, namely 3 types of brain tumors, namely meningioma, glioma and
pituitary and normal brain. This research will determine the success of a CNN model that was built
for the classification of brain tumor types based on the results of MRI scans with the addition of
preprocessing Cropping techniques. The test shows that the CNN method can classify brain tumor
types quite well. The results show that by adding cropping to the image, it can increase the
accuracy of the Confusion Matrix test.
The confusion matrix test with 630 test data shows that with preprocessing cropping the
average accuracy is 83%, precision is 83%, recall is 82%, and F1-score is 82%, while without
cropping, the average accuracy is 61%, precision is 72%, and 58% recall, and 58% F1-score.
Keywords: classification, brain tumor, CNN, cropping, confusion matrix

Item Type: Thesis (Other)
Uncontrolled Keywords: klasifikasi, tumor otak ,CNN,cropping ,confusion matrix
Subjects: Z Bibliography. Library Science. Information Resources > ZA Information resources
Divisions: Faculty of Engineering, Science and Mathematics > School of Engineering Sciences
Depositing User: Eko Yuli
Date Deposited: 07 Dec 2021 03:58
Last Modified: 23 Aug 2022 07:01
URI: http://eprints.upnyk.ac.id/id/eprint/27229

Actions (login required)

View Item View Item