Claim Missing Document
Check
Articles

Found 4 Documents
Search

Diagnosa Tumor Otak Berdasarkan Citra MRI (Magnetic Resonance Imaging) Ida Bagus Leo Mahadya Suta; Rukmi Sari Hartati; Yoga Divayana
Jurnal Teknologi Elektro Vol 18 No 2 (2019): (Mei-Agustus) Majalah Ilmiah Teknologi Elektro
Publisher : Universitas Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (373.104 KB) | DOI: 10.24843/MITE.2019.v18i02.P01

Abstract

Brain tumors are one of the most deadly diseases, one of the most common types is glioma, about 6 out of 100,000 patients are glioma sufferers. Digital imagery through Magnetic Resonance Imaging (MRI) is one method to help doctors analyze and classify brain tumor types. However, manual classification requires a long time and has a high risk of errors, so an automatic and accurate method is needed to classify MRI images. Convolutional Neural Network (CNN) is one of the solutions for automatic classification in MRI images. CNN is a deep learning algorithm that has the ability to learn on its own from the previous case. And from the research that has been done, the results obtained that CNN is able to complete the classification of brain tumors with high accuracy. Accuracy enhancements are obtained by developing the CNN algorithm either by determining the kernel value and / or activation function.
Segmentasi Tumor Otak Berdasarkan Citra Magnetic Resonance Imaging Dengan Menggunakan Metode U-NET Ida Bagus Leo Mahadya Suta; Made Sudarma; I Nyoman Satya Kumara
Jurnal Teknologi Elektro Vol 19 No 2 (2020): (Juli - Desember) Majalah Ilmiah Teknologi Elektro
Publisher : Universitas Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/MITE.2020.v19i02.P05

Abstract

Brain tumor is a deadly disease where 3.7% per 100,000 patients have malignant tumors. To analyze brain tumors can be done through magnetic resonance imaging (MRI) image segmentation. Automatic image analysis process is needed to save time and improve accuracy of doctor diagnoses. Automatic segmentation can be done with deep learning. U-NET is one of the methods used to segment medical images because it works at pixel level. By applying the ReLU and Adam Optimizer activation function, this method can solve the problem of segmenting brain tumors. Dataset for the training and validation process using BRATS 2017. Several hyperparameters are applied to this method: learning rate (lr) = 0.0001, batch size (bz) = 5, epoch = 80 and beta (b_1) = 0.9. From a series of processes carried out, accuracy of the U-NET method is calculated by Dice Coefficient formula and results in following accuracy values, during training of 90.22% (Full Tumor), 78.09% (Core Tumor) dan 80.20% (Enhancing Tumor).
Diagnosa Tumor Otak Berdasarkan Citra MRI (Magnetic Resonance Imaging) Ida Bagus Leo Mahadya Suta; Rukmi Sari Hartati; Yoga Divayana
Jurnal Teknologi Elektro Vol 18 No 2 (2019): (Mei-Agustus) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/MITE.2019.v18i02.P01

Abstract

Brain tumors are one of the most deadly diseases, one of the most common types is glioma, about 6 out of 100,000 patients are glioma sufferers. Digital imagery through Magnetic Resonance Imaging (MRI) is one method to help doctors analyze and classify brain tumor types. However, manual classification requires a long time and has a high risk of errors, so an automatic and accurate method is needed to classify MRI images. Convolutional Neural Network (CNN) is one of the solutions for automatic classification in MRI images. CNN is a deep learning algorithm that has the ability to learn on its own from the previous case. And from the research that has been done, the results obtained that CNN is able to complete the classification of brain tumors with high accuracy. Accuracy enhancements are obtained by developing the CNN algorithm either by determining the kernel value and / or activation function.
Segmentasi Tumor Otak Berdasarkan Citra Magnetic Resonance Imaging Dengan Menggunakan Metode U-NET Ida Bagus Leo Mahadya Suta; Made Sudarma; I Nyoman Satya Kumara
Jurnal Teknologi Elektro Vol 19 No 2 (2020): (Juli - Desember) Majalah Ilmiah Teknologi Elektro
Publisher : Program Studi Magister Teknik Elektro Universitas Udayana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/MITE.2020.v19i02.P05

Abstract

Brain tumor is a deadly disease where 3.7% per 100,000 patients have malignant tumors. To analyze brain tumors can be done through magnetic resonance imaging (MRI) image segmentation. Automatic image analysis process is needed to save time and improve accuracy of doctor diagnoses. Automatic segmentation can be done with deep learning. U-NET is one of the methods used to segment medical images because it works at pixel level. By applying the ReLU and Adam Optimizer activation function, this method can solve the problem of segmenting brain tumors. Dataset for the training and validation process using BRATS 2017. Several hyperparameters are applied to this method: learning rate (lr) = 0.0001, batch size (bz) = 5, epoch = 80 and beta (b_1) = 0.9. From a series of processes carried out, accuracy of the U-NET method is calculated by Dice Coefficient formula and results in following accuracy values, during training of 90.22% (Full Tumor), 78.09% (Core Tumor) dan 80.20% (Enhancing Tumor).