Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal ELKHA : Jurnal Teknik Elektro METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi
Dirgayussa, I Gde Eka
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Energy Industrial & Manufacturing Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Perancangan, Implementasi Monitoring dan Kontrol Alat Pemanggang Kopi Tampubolon, Friyogi; Pratama, Yohanssen; Dirgayussa, I Gde Eka
ELKHA : Jurnal Teknik Elektro Vol. 12 No. 2 October 2020
Publisher : Faculty of Engineering, Universitas Tanjungpura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/elkha.v12i2.41188

Abstract

Coffee roasting is the process of removing the water that is exist in the coffee beans. Roasted coffee beans will change weight and give a nice aroma and taste. The longer the coffee beans are roasted, the color of the coffee beans will continue getting darker blackish brown. The roasting level of coffee beans is determined from the change in color of coffee beans starting from light, medium and dark. Roasting coffee beans that develop on a home industry scale is still manual, that is, using more human labor in its operation. Therefore, this research will be made an automatic coffee roasting machine using a heater to heating the coffee beans, a DC motor to stir roasted coffee beans and a webcam to monitor changes in the color of coffee beans when roasted. Components of heating elements and motors controlled by Arduino Mega 2560 microcontroller while the webcam is connected with Raspberry Pi 3. As a component of performance that has been met with sensors as data collectors, microcontrollers as data processors and actuators as control systems. In this researh 3D modeling for a roasting container is done using SketchUp 3D design software. The results of the coffee roasting machine can meet the requirements of the system designed in accordance with the roasting level desired by the user and the thermocoupel give a better result in reading the temperature parameter compared to infrared thermometer. In 4,5 minutes the difference reading in temperature reach 27,50C between two sensors.
FEW-SHOT LEARNING FOR AML CELL CLASSIFICATION USING PROTOTYPICAL NETWORKS Dirgayussa, I Gde Eka; Herman, Kevin Elfancyus; Nugroho, Doni Bowo; Sekar Asri Tresnaningtyas; Meita Mahardianti; Nurul Maulidiyah; Rafli Filano; Rudi Setiawan; Muhammad Artha Jabatsudewa Maras; Yohanssen Pratama
METHODIKA: Jurnal Teknik Informatika dan Sistem Informasi Vol. 11 No. 2 (2025): Volume 11 Nomor 2 Tahun 2025
Publisher : Universitas Methodist Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46880/mtk.v11i2.4650

Abstract

Accurate blood cell classification is crucial for diagnosing Acute Myeloid Leukemia (AML) but limited medical data poses challenges for traditional machine learning models. This study presents a Few-Shot Learning (FSL) framework utilizing a Prototypical Network architecture with a ResNet-34 backbone to classify AML blood cell types from microscopic images. In this study, we utilize datasets consisting of 15 morphologically distinct cell classes. A 15-way, 5-shot, 5-query episodic setup was adopted to simulate data-scarce conditions. Evaluation via 5-fold cross-validation yielded strong performance, with an average accuracy of 97.76%, precision of 98.78%, recall of 96.55%, and F1-score of 97.76%. FSL training times were consistent (4.22–4.26 minutes per fold), and t-SNE along with confusion matrices confirmed the model’s ability to distinguish similar cell types. To validate the approach, its performance was compared with a conventional supervised CNN using the same ResNet-34 backbone. The FSL model outperformed the CNN across all metrics such as accuracy (98.32% vs. 77.25%), precision (98.55% vs. 76.87%), recall (98.31% vs. 78.66%), and F1-score (98.33% vs. 75.26%), while also requiring far less training time (~4.24 min/fold vs. ~420 min total). These results highlight the promise of FSL based methods for accurate, efficient, and scalable hematologic diagnostics in data limited settings.
Co-Authors Doni Bowo Nugroho Herman, Kevin Elfancyus Meita Mahardianti Muhammad Artha Jabatsudewa Maras Nurul Maulidiyah Pratama, Yohanssen Rafli Filano Rudi Setiawan Sekar Asri Tresnaningtyas Tampubolon, Friyogi Yohanssen Pratama