Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI) Seminar Nasional Aplikasi Teknologi Informasi (SNATI) TELKOMNIKA (Telecommunication Computing Electronics and Control) International Journal of Advances in Intelligent Informatics International Journal of Industrial Research and Applied Engineering
Daswara Djajasasmita
Universitas Jenderal Achmad Yani
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Mechanical Engineering

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Significant variables extraction of post-stroke EEG signal using wavelet and SOM kohonen Esmeralda C. Djamal; Deka P. Gustiawan; Daswara Djajasasmita
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 17, No 3: June 2019
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v17i3.11776

Abstract

Stroke patients require a long recovery. One success of the treatment given is the evaluation and monitoring during recovery. One device for monitoring the development of post-stroke patients is Electroencephalogram (EEG). This research proposed a method for extracting variables of EEG signals for post-stroke patient analysis using Wavelet and Self-Organizing Map Kohonen clustering. EEG signal was extracted by Wavelet to obtain Alpha, beta, theta, gamma, and Mu waves. These waves, the amplitude and asymmetric of the symmetric channel pairs are features in Self Organizing Map Kohonen Clustering. Clustering results were compared with actual clusters of post-stroke and no-stroke subjects to extract significant variable. These results showed that the configuration of Alpha, Beta, and Mu waves, amplitude together with the difference between the variable of symmetric channel pairs are significant in the analysis of post-stroke patients. The results gave using symmetric channel pairs provided 54-74% accuracy.
Identifikasi Variabel-Variabel dari Sinyal Elektroensephalogram Pasien Rehabilitasi Stroke Menggunakan Wavelet dan Self-Organizing Map Deka P Gustiawan; Esmeralda C Djamal; Agus Komarudin; Daswara Djajasasmita
Seminar Nasional Aplikasi Teknologi Informasi (SNATI) 2018
Publisher : Jurusan Teknik Informatika, Fakultas Teknologi Industri, Universitas Islam Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Evaluasi terhadap pasien paska stroke yang terukur sangat dibutuhkan untuk mengetahui perkembangan aktivitas di otak dalam masa rehabilitasi. Salah satunya instrumen yang dapat menangkap aktivitas listrik di otak adalah Elektroensephalogram (EEG). Pengamatan visual yang dilakukan dokter dari rekaman EEG adalah kerapatan, amplitudo, bentuk gelombang, dan perbandingan sinyal pada kanal yang simetrik, namun tidaklah mudah. Penelitian ini melakukan ekstraksi dari sinyal EEG untuk memperoleh variabel-variabel signifikan dari pasien paska stroke. Sinyal EEG diperoleh dari 25 pasien paska stroke dan 25 orang sehat dari 14 kanal. Setiap sinyal selama 180 detik diekstraksi menggunakan Wavelet untuk memperoleh gelombang Alfa, Beta, Teta, Gama, dan Mu. Clustering dilakukan menggunakan Self Organizing Map (SOM) Kohonen dengan fitur masukan kelima gelombang, amplitudo, dan asimetrik dari kanal simetrik. SOM melakukan clustring berdasarkan fitur-fitur pembeda pola, sehingga hasil clustring dibandingkan dengan cluster dari data sebenarnya. Cara ini dilakukan untuk menentukan variabel-variabel sinyal EEG beserta kanal-kanalnya yang memberikan akurasi terbaik. Hasil penelitian menunjukkan penggunaan keenam fitur dari 14 kanal dan fitur sinkronisasi dari 7 pasang kanal memberikan ketepatan klustering sebesar 54-68%. Akurasi fitur tertinggi diperoleh dari variabel perubahan amplitudo. Sistem identifikasi telah diimplementasikan dalam perangkat lunak dan diintegrasikan dengan wireless EEG Emotiv. Waktu komputasi dari sistem identifikasi sekitar empat menit, cukup realistis yang dapat digunakan untuk membantu analisis dokter.
Post-Stroke identification of EEG signals using recurrent neural networks and long short-term memory Wanodya Sansiagi; Esmeralda Contessa Djamal; Daswara Djajasasmita; Arlisa Wulandari
International Journal of Advances in Intelligent Informatics Vol 7, No 2 (2021): July 2021
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v7i2.512

Abstract

Stroke often causes disability, so patients need rehabilitation for recovery. Therefore, it is necessary to measure its effectiveness. An Electroencephalogram (EEG) can capture the improvement of activity in the brain in stroke rehabilitation. Therefore, the focus is on the identification of several post-rehabilitation conditions. This paper proposed identifying post-stroke EEG signals using Recurrent Neural Networks (RNN) to process sequential data. Memory control in the use of RNN adopted Long Short-Term Memory. Identification was provided out on two classes based on patient condition, particularly "No Stroke" and "Stroke". EEG signals are filtered using Wavelet to get the waves that characterize a stroke. The four waves and the average amplitude are features of the identification model. The experiment also varied the weight correction, i.e., Adaptive Moment Optimization (Adam) and Stochastic Gradient Descent (SGD). This research showed the highest accuracy using Wavelet without amplitude features of 94.80% for new data with Adam optimization model. Meanwhile, the feature configuration tested effect shows that the use of the amplitude feature slightly reduces the accuracy to 91.38%. The results also show that the effect of the optimization model, namely Adam has a higher accuracy of 94.8% compared to SGD, only 74.14%. The number of hidden layers showed that three hidden layers could slightly increase the accuracy from 93.10% to 94.8%. Therefore, wavelets as extraction are more significant than other configurations, which slightly differ in performance. Adam's model achieved convergence in earlier times, but the speed of each iteration is slower than the SGD model. Experiments also showed that the optimization model, number of epochs, configuration, and duration of the EEG signal provide the best accuracy settings.
Miniaturized Infusion Monitoring System with Weight Sensor (Load Cell) Based on AT-MEGA 328 Microcontroller Daswara Djajasasmita; M. Reza Hidayat; Susanto Sambasri
International Journal of Industrial Research and Applied Engineering Vol 5, No 1: APRIL 2020
Publisher : Institute of Research and Community Outreach - Petra Christian University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.9744/jirae.5.1.14-18

Abstract

The development of medical equipment with advanced technology can provide convenience in providing services to the community. One of the equipment that is in the hospital and most often used is an IV. In its use, manual infusion is used to determine the volume of the infusion and must be monitored every hour or even minute by the nurse. This is considered quite difficult, especially in the era of the Covid-19 pandemic where minimal interaction is required from 2 individuals. This is done by utilizing sensor technology to monitor the patient's condition so that the frequency of the nurses checking the condition of the infusion is getting less. Therefore, in this research, manufactured of an infusion monitoring system using aweight sensor (Load Cell) based on the ATMEGA 328 microcontroller was carried out. The sensor of this monitoring system uses a Load Cell Weight Sensor with the HX711 module which is integrated into the ARDUINO UNO MCU. The output of the system is displayed on a 16 x 2 mm LCD as well as a Macro Excel which will display the percentage value of intravenous fluids in the PC and real-time automatic data logging into the macro excel. The infusion used uses Nacl fluid infusion. The test results of the system as a whole show that the data for measuring levels (%) of intravenous fluids can be sent and displayed on the LCD and PC. The incoming data is converted into a table at certain time intervals according to the user's choice into the operator. The experiment was carried out 10 times by looking at changes in the contents of the infusion fluid over time in 11 stages where the LED lights up when the weight of the infusion reaches 40 - 46 grams. Then, the buzzer and LED have turned on when the weight of the infusion is less than 5 grams. From the experiment, it can be concluded the 10% setpoint alarm works well, i.e. when the infusion load is less than the 1% set point, the buzzer and LED will light up until the intravenous fluid is replaced with a new one.
Emotion recognition of electroencephalogram using hybrid convolutional neural networks and vision transformer Esmeralda Contessa Djamal; Revan Vio Endriansyah; Daswara Djajasasmita
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 4: August 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i4.pp3934-3943

Abstract

Electroencephalogram (EEG) signal-based emotion classification faces challenges in spatio-temporal complexity and inter-channel redundancy. This paper proposes a 2D convolutional neural network (CNN) and vision transformer (ViT) approach. Relevant emotions require signal extraction in the 4-40 Hz frequency band using the discrete wavelet transform (DWT). This study uses the SEED dataset from 15 subjects, with 62 channels reduced to 12. Every 5-second segment is decomposed using DWT into four frequency bands, which are then mapped to a 2D spatial representation for CNNs and ViTs. Experiment results show that the DWT-2D CNN-ViT achieves the best accuracy of 87% with a final loss of 0.087. In the meantime, the CNN-only model (85%), the ViT-only model (77%), and the DWT-only model (47%). Testing with several optimizers also shows that AdamW provides the highest performance with the fastest training time of 12.34 minutes. These results demonstrate that this integration can produce more efficient and stable learning. These findings indicate that the combination of DWT and the CNN-ViT hybrid architecture is effective for accurate, stable EEG-based emotion recognition and is potentially applicable to real-time emotion-monitoring systems.
Co-Authors Arlisa Wulandari Deka P Gustiawan Deka P. Gustiawan Esmeralda C Djamal Esmeralda C. Djamal Esmeralda Contessa Djamal Komarudin, Agus M. Reza Hidayat Revan Vio Endriansyah Susanto Sambasri Wanodya Sansiagi