Nashrullah, Raditya Farel (2023) Smooth Support Vector Machine (SSVM) untuk Deteksi Penderita Major Depressive Disorder (MDD) Berdasarkan Sinyal Elektroensefalogram (EEG). Other thesis, Institut Teknologi Sepuluh Nopember.
Text
06211940000113-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (4MB) | Request a copy |
Abstract
Major Depressive Disorder (MDD) merupakan gangguan suasana hati yang berlangsung secara intens dan menyebabkan perubahan pola aktivitas penderitanya, yang menduduki peringkat pertama penyakit mental yang dialami oleh penduduk di Indonesia. Saat ini, diagnosis penderita MDD masih menggunakan evaluasi intensitas gejala melalui kuesioner yang dianggap bersifat subjektif. Penelitian ini bertujuan untuk mengembangkan metode diagnosis Major Depressive Disorder (MDD) yang bersifat lebih objektif melalui penggunaan sinyal Elektroensefalogram (EEG). Penelitian ini menerapkan penggunaan metode Smooth Support Vector Machine (SSVM) pada tahap klasifikasi yang dipercaya memiliki tingkat efisien yang lebih baik untuk melakukan klasifikasi pada data dengan dimensi besar. Rekaman EEG mengalami proses filterisasi menggunakan Finite Impulse Response (FIR) Filter untuk menghilangkan noise dan mendapatkan sinyal pada masing-masing sub-band frekuensi. Sinyal yang telah difilter dipotong menjadi epoch berdurasi sepuluh detik dengan lima detik overlap. Tahap selanjutnya, dilakukan ekstraksi fitur dari potongan sinyal EEG tersebut untuk mendapatkan fitur-fitur yang akan digunakan sebagai variabel prediktor dalam tahap klasifikasi. Hasil klasifikasi menggunakan metode SSVM menunjukkan model dengan nilai AUC sebesar 0,988 menggunakan fitur pada sub-band delta, theta, alpha, beta, dan gamma, menandakan kemampuan model dalam mendiagnosis MDD yang sangat baik. Penelitian ini juga menemukan bahwa sub-band gamma merupakan sub-band terbaik dalam diagnosis MDD dengan nilai AUC tertinggi, yaitu 0,961, menggunakan metode klasifikasi SSVM dibandingkan dengan sub-band frekuensi lainnya. Hasil penelitian ini diharapkan dapat memberikan bantuan serta alternatif metode objektif bagi psikiater dalam mendiagnosis MDD yang bersifat objektif.
===============================================================================================================================
Major Depressive Disorder (MDD) is an intense mood disorder that causes significant changes in the activity patterns of individuals and ranks as the leading mental illness experienced by the population in Indonesia. Currently, the diagnosis of MDD relies on subjective evaluations of symptoms obtained through questionnaires and interviews. Therefore, this study aims to develop a more objective method for diagnosing Major Depressive Disorder (MDD) using Electroencephalogram (EEG) tests. The study employs the Smooth Support Vector Machine (SSVM) method for classification, as it is believed to have better efficiency in classifying high-dimensional data. The EEG recordings undergo a filtering process using the Finite Impulse Response (FIR) Filter to eliminate noise and obtain signals in specific frequency sub-bands. The filtered signals are segmented into epochs of ten seconds, with a five-second overlap, to obtain more stationary signal segments. Feature extraction is then performed on these EEG segments to obtain predictor variables for the classification stage. The classification results using the SSVM method demonstrate a model with an AUC value of 0.988 using features extracted from delta, theta, alpha, beta, and gamma sub-band frequency, indicating excellent diagnostic capabilities for MDD. The study also finds that the gamma sub-band is the most effective in diagnosing MDD, with the highest AUC value of 0.961 among other frequency sub-bands when using the SSVM classification method. The findings of this research are expected to provide assistance and alternative objective methods for psychiatrists in diagnosing MDD objectively.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Diagnosis, Electroencephalogram, Finite Impulse Response Filter, Major Depressive Disorder, Smooth Support Vector Machine, Diagnosis, Elektroensefalogram, Finite Impulse Response Filter, Major Depressive Disorder, Smooth Support Vector Machine |
Subjects: | H Social Sciences > HA Statistics Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. R Medicine > RC Internal medicine > RC386.5 Electroencephalography. T Technology > T Technology (General) > T57.5 Data Processing |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
Depositing User: | Raditya Farel Nashrullah |
Date Deposited: | 25 Aug 2023 06:51 |
Last Modified: | 25 Aug 2023 06:51 |
URI: | http://repository.its.ac.id/id/eprint/104312 |
Actions (login required)
View Item |