cover
Contact Name
Triwiyanto
Contact Email
triwiyanto123@gmail.com
Phone
+628155126883
Journal Mail Official
triwiyanto123@gmail.com
Editorial Address
Pucang Jajar Timur No. 10, Surabaya, Indonesia
Location
Kota surabaya,
Jawa timur
INDONESIA
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics
ISSN : -     EISSN : 26568624     DOI : https://doi.org/10.35882/ijeeemi
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics (IJEEEMI) publishes peer-reviewed, original research and review articles in an open-access format. Accepted articles span the full extent of the Electronics, Biomedical, and Medical Informatics. IJEEEMI seeks to be the world’s premier open-access outlet for academic research. As such, unlike traditional journals, IJEEEMI does not limit content due to page budgets or thematic significance. Rather, IJEEEMI evaluates the scientific and research methods of each article for validity and accepts articles solely on the basis of the research. Likewise, by not restricting papers to a narrow discipline, IJEEEMI facilitates the discovery of the connections between papers, whether within or between disciplines. The scope of the IJEEEMI, covers: Electronics: Intelligent Systems, Neural Networks, Machine Learning, Fuzzy Systems, Digital Signal Processing, Image Processing, Electromedical: Biomedical Signal Processing and Control, Artificial intelligence in biomedical imaging, Machine learning and Pattern Recognition in a biomedical signal, Medical Diagnostic Instrumentation, Laboratorium Instrumentation, Medical Calibrator Design. Medical Informatics: Intelligent Biomedical Informatics, Computer-aided medical decision support systems using heuristic, Educational computer-based programs pertaining to medical informatics
Articles 205 Documents
Generating Synthetic B-Mode Fetal Ultrasound Images Using CycleGAN-Based Deep Learning Hermawati, Fajar Astuti; Hardiansyah, Bagus; Andrianto, Andrianto
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 8 No. 1 (2026): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/ijeeemi.v8i1.282

Abstract

B-mode ultrasound (USG) is a key imaging modality for fetal assessment, providing a noninvasive approach to monitor anatomical development and detect congenital anomalies at an early stage. However, portable ultrasound devices commonly used in low-resource healthcare settings often yield low-resolution images with significant speckle noise, reducing diagnostic accuracy. Furthermore, the scarcity of labeled medical data, caused by privacy regulations such as HIPAA and the high cost of expert annotation, poses a significant challenge for developing robust artificial intelligence (AI) diagnostic models. This study proposes a CycleGAN-based deep learning model enhanced with a histogram-guided discriminator (HisDis) to generate realistic synthetic B-mode fetal ultrasound images. A publicly available dataset from the Zenodo repository containing 1,000 grayscale fetal head images was utilized. Preprocessing included normalization, histogram equalization, and image resizing, while the architecture combined two ResNet-based generators and a dual discriminator configuration integrating PatchGAN and histogram-guided evaluation. The model was trained using standard optimization settings to ensure stable convergence. Experimental results demonstrate that the proposed HisDis module accelerates convergence by 18 epochs and reduces the Fréchet Inception Distance (FID) by 23.6 percent from 1580.72 to 1208.49 compared with the baseline CycleGAN. Statistical analysis revealed consistent pixel-intensity distributions between the original and synthetic images, with entropy from 7.16 to 7.40. At the same time, visual assessment confirmed that critical anatomical structures, including the brain midline and lateral ventricles, were well preserved. These results indicate that the CycleGAN-HisDis model produces statistically and visually realistic fetal ultrasound images suitable for medical data augmentation and AI-based diagnostic training. Furthermore, this approach holds potential to enhance diagnostic reliability and clinical education in healthcare settings with limited imaging resources. Future work will focus on clinical validation and generalization across diverse fetal ultrasound datasets.
Telemedicine and AI in Remote Prediabetes Monitoring Among Adolescents Solechah, Siti Aisyah; Saputro, Setyo Wahyu; Adini, Muhammad Hifdzi; Faisal, Mohammad Reza; Kurniawan, Erick; Umiatin, Umiatin
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 8 No. 1 (2026): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/ijeeemi.v8i1.283

Abstract

The escalating prevalence of prediabetes in Indonesia, particularly among children and adolescents, necessitates the development of lightweight, adaptable, and cost-effective telemedicine solutions for the noninvasive monitoring of blood glucose levels. Existing systems predominantly employ machine learning and deep learning approaches that require substantial computational resources and stable internet connectivity, limiting their applicability in regions with constrained digital infrastructure. The objective of this study is to develop an artificial intelligence (AI)–driven telemedicine system that employs an expert system to determine prediabetes status by utilizing commercially available smartwatches as noninvasive optical sensors. The methodological approach includes an examination of smartwatch capabilities to identify Bluetooth Low Energy (BLE) sensors, service architectures, and the Generic Attribute Profile (GATT); the development of a Rule-Based Reasoning (RBR) expert system to determine prediabetes status using Fasting Plasma Glucose (FPG) and Postprandial Plasma Glucose (PP2) measurements; and the application of Rapid Application Development (RAD) methods in the development of Flutter-based mobile applications and Laravel Inertia Vue–based web applications. The results of this study demonstrate that the telemedicine system operates in both offline and online modes and incorporates AI functionality on mobile devices and servers without requiring extensive computational resources. All system functionalities successfully passed testing, and the expert system achieved 100% accuracy in determining prediabetes status. In conclusion, the integration of telemedicine and AI-based expert systems provides an effective, economical, and flexible solution that can be widely implemented in Indonesia to reduce the increasing incidence of prediabetes through continuous digital health monitoring.
Comparative Study of Filter, Wrapper, and Hybrid Feature Selection Using Tree-Based Classifiers for Software Defect Prediction Rahmayanti, Rahmayanti; Herteno, Rudy; Saputro, Setyo Wahyu; Saragih, Triando Hamonangan; Abadi, Friska
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 8 No. 1 (2026): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/ijeeemi.v8i1.294

Abstract

Software defect prediction (SDP) is essential for improving software reliability by enabling the early identification of modules that may contain defects before the release stage. SDP commonly exhibits redundant or non-contributory metrics, underscoring the need for feature selection to derive a more informative subset. To address this problem, the present study investigates and compares the effectiveness of three feature-selection strategies: SelectKBest (SKB), Recursive Feature Elimination (RFE), and the hybrid SKB+RFE, in enhancing the performance of tree-based classifiers on the NASA Metrics Data Program (MDP) data collections. The study utilizes three classification algorithms, namely Random Forest (RF), Extra Trees (ET), and Bagging (Decision Tree), with Area Under the Curve (AUC) serving as the primary metric for assessing model performance. Experimental results reveal that the RFE and Extra Trees combination yields the top performance, producing an average AUC of 0.7855. This is subsequently followed by the SKB+RFE+ET configuration, which achieves an AUC of 0.7809, and SKB+ET at 0.7776. These findings demonstrate that iterative wrapper-based approaches such as RFE can identify more relevant and effective feature subsets than filter or hybrid strategies, with the RFE+Extra Trees configuration yielding the strongest overall predictive performance and wrapper-based methods exhibiting higher stability across heterogeneous datasets. Even without hyperparameter tuning and relying solely on class-weighting rather than explicit resampling techniques, the findings offer empirical insight into the isolated influence of feature selection on predictive performance. Overall, the study confirms that RFE combined with Extra Trees offers the strongest predictive performance on NASA MDP data collections and forms a foundation for developing more adaptive and robust models.
Real-Time, Multi-Command Drone Navigation Using a Consumer-Grade EEG-Based SSVEP BCI Wijaya, Anderias Eko; Nurizati, Nurizati; Hermawan, Rian; Suhendra, Muhammad Agung
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 8 No. 1 (2026): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/ijeeemi.v8i1.295

Abstract

Steady-state visual evoked potential (SSVEP) brain-computer interfaces (BCIs) provide a non-invasive method for hands-free device control. However, their practical applications are limited by reliance on costly laboratory-grade electroencephalography (EEG) systems. This study addresses this gap by designing and evaluating a real-time, six-command SSVEP-BCI for drone navigation using a consumer-grade EEG headset. An adaptive processing pipeline was developed to extract spectral and spatial features, which were classified using Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Network (ANN) models. Analysis of data from 30 participants revealed that the RF classifier achieved an optimal balance between performance and speed, with a high classification accuracy of 87.24% and a low computational latency of 0.09 seconds, resulting in a high information transfer rate (ITR) of 35.0 bits/min. In contrast, the ANN was insufficiently accurate, and SVM performance was marginal. These findings demonstrate the viability of low-cost, multi-command SSVEP-BCIs for applications in assistive technology, teleoperation, and human-computer interaction.
MEWT-Enhanced EEGNet for ASD EEG Classification: Performance Evaluation with k-Fold Cross-Validation Fathur Rahman, Imam; Melinda, Melinda; Yunidar, Yunidar; Basir, Nurlida; Rafiki, Aufa
Indonesian Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol. 8 No. 1 (2026): February
Publisher : Jurusan Teknik Elektromedik, Politeknik Kesehatan Kemenkes Surabaya, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/ijeeemi.v8i1.313

Abstract

Accurate and reliable classification of autism spectrum disorder (ASD) from electroencephalography (EEG) signals remains challenging due to the inherently nonstationary, nonlinear, and multichannel nature of EEG data. These properties complicate the extraction of discriminative features that are both stable and computationally efficient. To address this challenge, this study proposes a compact deep-learning pipeline that integrates the Multivariate Empirical Wavelet Transform (MEWT) with EEGNet for ASD–EEG classification. MEWT decomposes multichannel EEG signals into spectrally aligned subbands while preserving inter-channel relationships. The resulting MEWT-based features are then processed by EEGNet, a lightweight convolutional neural network specifically designed for EEG-based learning tasks. Performance was evaluated using 5-fold cross-validation. The proposed MEWT with the the EEGNet model achieved a mean test accuracy of 98.35%, with consistently high precision (98.23%), recall (98.45%), F1-score (98.34%), and specificity (98.24%) across all folds. Confusion-matrix results indicated very few and well-balanced false positives and false negatives, supporting stable discrimination between ASD and control EEG segments. A one-sample one-tailed t-test against a 50% chance level confirmed that all evaluated metrics were significantly above chance (p < 0.0001). When benchmarked against previously reported results on the same dataset, the proposed approach slightly improved upon EMD with EEGNet (97.99%) and clearly outperformed EWT with EEGNet (95.08%), suggesting that MEWT-derived multichannel features are well matched to compact convolutional architectures for ASD–EEG analysis. Despite these strong results, the study is limited by a small, single-site cohort and the use of a single deep-learning model. Future work will focus on standardized retraining across multiple feature families and validation on larger and more diverse populations to further assess robustness and generalizability