cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Triwiyanto
Contact Email
triwiyanto123@gmail.com
Phone
+628155126883
Journal Mail Official
editorial.jeeemi@gmail.com
Editorial Address
Department of Electromedical Engineering, Poltekkes Kemenkes Surabaya Jl. Pucang Jajar Timur No. 10, Surabaya, Indonesia
Location
Kota surabaya,
Jawa timur
INDONESIA
Journal of Electronics, Electromedical Engineering, and Medical Informatics
Published by Polilteknik Kesehatan Kemenkes Surabaya
ISSN : -     EISSN : 26568632     DOI : https://doi.org/10.35882/jeeemi
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering
The Journal of Electronics, Electromedical Engineering, and Medical Informatics (JEEEMI) is a peer-reviewed open-access journal. The journal invites scientists and engineers throughout the world to exchange and disseminate theoretical and practice-oriented topics which covers three (3) majors areas of research that includes 1) Electronics, 2) Biomedical Engineering, and 3)Medical Informatics (emphasize on hardware and software design). Submitted papers must be written in English for an initial review stage by editors and further review process by a minimum of two reviewers.
Arjuna Subject : Teknik (semua kategori) - Teknik Biomedis
Articles 368 Documents
 33   34   35   36   37 
TPE-FFT: A Novel Two-Phase Execution Fast Fourier Transform for Ultra-Low-Latency Bioinformatics Architecture Saravanakumar C; Usha Bhanu N; Subhashini N; Preethi Sambandam Raju
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.1933

Abstract

The Fast Fourier Transform (FFT) is widely employed for frequency-domain analysis of genomic and biomedical signals; however, conventional block-oriented FFT processing generally requires the complete input sequence before the transform can be finalized. When genomic data are acquired sequentially, computations that depend on the final sample may remain unresolved until that sample arrives, resulting in additional post-arrival processing latency. This study presents a Two-Phase Execution Fast Fourier Transform (TPE-FFT), a dependency-aware execution strategy for radix-2 Decimation-in-Time (DIT) FFT processing. The proposed method exploits the linearity of the Discrete Fourier Transform (DFT) to separate the contribution of the final acquired sample from the contributions of the preceding samples. During Phase I, computations independent of the final sample are performed using a temporary zero representation, while the corresponding final-sample contribution is evaluated during Phase II and incorporated through an exact Fourier-domain correction. Variable-length genomic sequences are mapped to a common radix-2 transform length using zero-padding, and the proposed method is evaluated against a conventional radix-2 DIT-FFT implementation under the same computational conditions. A total of 1,075 genomic sequences are evaluated. TPE-FFT achieves positive measured execution-time reduction for 1,053 sequences (97.95%). The cumulative execution time decreases from 225.56469 s for the conventional implementation to 7.80065 s for TPE-FFT, corresponding to a 96.54% aggregate reduction and a 28.92× runtime ratio. For a representative 33,681-sample genomic sequence, execution time decreases from 15.15 s to 0.51 s, corresponding to a 96.63% reduction. The results indicate that dependency-aware scheduling can substantially reduce the computational workload remaining after final-sample arrival while retaining the underlying O(Nlog₂N) FFT complexity. The proposed strategy therefore provides a computational basis for further investigation of low-latency FFT processing in sequential genomic signal-processing applications.
Optimized FPGA Based Modified MFCC Architecture Using Time Division Multiplexed FFT Processing Muthumari M; R. S. Valarmathi
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.1938

Abstract

Advances in real-time speech recognition have led to an increasing demand for compact as well as energy-efficient MFCC hardware capable of generating accurate and consistent speech features with low computational latency. Conventional MFCC solutions frequently necessitate specialized processing resources and several math operations, resulting in increased size, power consumption, and hardware complexity. To address these restrictions, this paper proposes a modified MFCC architecture that incorporates dual window operation, a Radix-2 Time Division Multiplexed FFT (TDM FFT), and an Approximate Radix-4 Distributed Arithmetic (DA) multiplier. The proposed framework utilizes shared computational abilities for speech framing, dual window operations, FFT calculation, magnitude prediction, Mel filter bank analysis, logarithmic conversion, as well as MFCC coefficient generation. A Xilinx KV260 Zynq UltraScale+ FPGA is used to implement the standard architecture, and 45 nm CMOS ASIC synthesis is used for further assessment. The MFCC coefficients produced by MATLAB are compared with the appropriate Verilog outputs to perform functional validation. In comparison to the standard architecture, experimental results show a 50% reduction in LUT/ALM utilization, a 60% reduction in DSP resource usage, and a 45% reduction in power consumption. ASIC synthesis achieves a roughly 50% reduction in core silicon area while further reducing processing latency from 1024 to 320 clock cycles, or a 69% reduction. Time-division multiplexing, approximate distributed arithmetic processing, and efficient computing resource sharing all contribute to these gains. The findings show that the suggested architecture strikes a good balance among MFCC feature characteristics, processing speed, power consumption, as well as hardware efficiency. Thus, for low-power, low-latency, along with resource constrained real-time voice recognition applications, the proposed architecture offers a feasible and scalable approach.
ECA-ECGNet: An Efficient Channel Attention-Enhanced 1D CNN for ECG Arrhythmia Detection Maher Al-Asaady; Mohanad Rashid; Mahmood Mahmood
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.1971

Abstract

Cardiovascular diseases remain a major cause of mortality worldwide, with cardiac arrhythmias representing a potentially serious condition. Although deep learning has improved automated electrocardiogram (ECG) analysis, many recent models rely on recurrent, residual, transformer-based, or multi-input architectures, thereby increasing computational complexity. Moreover, reported classification performance can vary substantially according to the adopted data-partitioning strategy. This study proposes an Efficient Channel Attention ECG Network, termed ECA-ECGNet, for automated five-class ECG beat classification using the MIT-BIH Arrhythmia Database. ECG beats were segmented, normalized, and grouped into five AAMI categories: Normal (N), Supraventricular (S), Ventricular (V), Fusion (F), and Unknown (Q). The ECA-ECGNet integrates three Conv1D-based feature-extraction blocks with lightweight ECA modules to emphasize informative channel features. To mitigate the severe class imbalance inherent in clinical ECG data, categorical focal loss was implemented during model training to dynamically scale the loss based on prediction confidence. ECA-ECGNet was assessed under complementary intra-patient and inter-patient evaluation settings. The model achieved an accuracy of 99.20%, a macro F1-score of 0.953, and a weighted F1-score of 0.992. In the inter-patient setting, complete subject groups were separated between model development and testing, resulting in an accuracy of 97.96%, a macro F1-score of 0.9165, and a weighted F1-score of 0.9798. The implementation of focal loss substantially narrowed the performance gap between majority and minority classes, demonstrating enhanced generalization for Supraventricular and Ventricular arrhythmias. Ultimately, the principal novelty of ECA-ECGNet is the synergistic application of structural channel attention and dynamic focal loss. This framework shows that diagnostic accuracy on unseen patients is achievable without high computational overhead, supporting real-time arrhythmia screening on low-resource clinical devices
Segmentation of Acute Ischemic Stroke Infarct using Multi-scale U-Net Architecture on Non-contrast CT Scan Images Osmalina Rahma; Aisyah Widayani; Nimas Maharani Putri; Rifai Chai
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.1980

Abstract

Acute ischemic stroke must be treated within a few hours, and the first scan is usually a non-contrast CT (NCCT), on which an early infarction is often hard to see. Automatic segmentation assist image interpretation, and multi-scale convolution is a commonly proposed approach. This study measured how much such a block improves U-Net performance for this task, and what it costs. Every convolutional block of U-Net, in the encoder and the decoder, was replaced by a block with three parallel branches using 1×1, 3×3 and 5×5 kernels. Both networks were trained on the public Acute Ischemic Stroke Dataset, on 375 studies with supratentorial infarcts. The data were divided by study; the scans of the dataset’s own test split were kept for testing, and augmentation was applied to the training subset alone. Four configurations were compared: each architecture with and without augmentation. On the 51 test patients, Multi-scale U-Net did not improve overlap. Its Dice score was 0.6372 against 0.6632 for U-Net with augmentation, and the paired difference per patient was 0.0233 in favour of U-Net (95 % CI −0.060 to 0.013, p = 0.19). The multi-scale block instead altered the balance between the two types error: sensitivity increased by 0.060 (p = 0.002), and precision decreased by 0.100 (p < 0.001). Augmentation improved both networks (+0.106 and +0.048 Dice per patient, p < 0.001), mainly by cutting false-positive lesions from 15.5 to 5.6 per patient. The multi-scale block costs 2.31 times as many parameters and 3.37 times as much inference time. An ablation over the target definition favoured the dataset authors' labels: adding the label for infarct invisible on NCCT cost 0.040 Dice and tripled false-positive lesions. Both networks segment small infarcts poorly, with Dice below 0.16 under 100 pixels, which remains the main obstacle.
Neuro-Spectral Causal Fusion Network with Lesion Dependency Reasoning for Explainable Diabetic Retinopathy Detection K. Venu; B. Thanga Parvathi; Kunchanapalli Rama Krishna; Yuvaraj S.
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 9 No 1 (2027): January
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v9i1.1998

Abstract

Background: Diabetic Retinopathy (DR) is a progressive microvascular complication of diabetes and a leading cause of preventable vision impairment worldwide. Early and accurate identification of DR severity is essential for timely intervention and effective disease management. However, conventional retinal screening relies heavily on manual examination of fundus images, which is labor-intensive, subjective, and dependent on specialist expertise. Although recent deep-learning models have demonstrated promising diagnostic performance, many existing approaches inadequately capture lesion-level relationships, underutilize clinical information, and provide limited interpretability for clinical decision-making. Aims: this study proposes a novel Neuro-Spectral Causal Fusion Network (NSCF-Net) for explainable diabetic retinopathy detection and severity grading. Methods: The proposed framework integrates an adaptive Neuro-Spectral Enhancement Module to improve lesion visibility, a Lesion Dependency Reasoning Graph to model pathological interactions among retinal abnormalities, and a Dual-Path Hybrid Transformer to jointly learn local lesion characteristics and global retinal context. Furthermore, a Causal Multimodal Fusion Unit combines retinal image representations with clinically relevant parameters, including HbA1c level, blood pressure, age, and duration of diabetes. To improve transparency and clinical trustworthiness, an Explainable Lesion Attribution Module generates counterfactual attention maps and hierarchical saliency visualizations highlighting diagnostically significant retinal regions. The proposed framework was evaluated on the APTOS 2019 diabetic retinopathy dataset, consisting of 3,662 retinal fundus images distributed across five DR severity classes. Results: Experimental results demonstrate the effectiveness of NSCF-Net, achieving 99.1% accuracy, 98.9% precision, 99.0% recall, and 98.8% F1-score. Comparative analyses and ablation studies further confirm the contribution of neuro-spectral enhancement, lesion dependency reasoning, multimodal fusion, and explainability components to overall performance improvement. Conclusion: These findings indicate that NSCF-Net provides an accurate, interpretable, and clinically reliable framework for automated diabetic retinopathy screening, severity assessment, and computer-aided ophthalmic decision support.
Diagnostic Accuracy of Four Large Language Models for the Cipto Triage Method: A Comparative Evaluation for Emergency Department Triage in Resource-Limited Settings Ahmad Ridwan Fauzi; Adelia Chitra Sazkia; Muhammad Ardiansyah
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.2022

Abstract

Triage accuracy remains suboptimal globally, particularly in developing countries employing simplified systems such as the Cipto Triage Method (CTM), and LLM performance on three-level triage systems in non-English contexts has not been evaluated. This study evaluated the diagnostic accuracy of four LLMs (DeepSeek V4 Pro, DeepSeek V4 Flash, Tencent Hy3, MiMo V2.5) for CTM classification as decision support tools for emergency department (ED) triage in resource-limited settings. The models were tested on a stratified random sample of 1,200 clinical records from the MIETIC corpus (v1.0.0; 5,528 unique cases after deduplication), relabeled from the Emergency Severity Index to CTM by four board-certified emergency physicians (400 cases per category), with physician consensus as the reference standard. Primary outcomes were overall accuracy and Cohen kappa; secondary outcomes included per-class sensitivity, specificity, positive and negative predictive value with 95% Wilson-score confidence intervals, undertriage detection rate, and API cost. Accuracy ranged from 66.5% to 68.1% (kappa 0.50–0.52). DeepSeek V4 Flash achieved the highest accuracy (68.1%, 95% CI 65.4–70.7%) at $0.207 total cost. No pairwise McNemar comparison was statistically significant (p > 0.05). All models exceeded benchmarks previously reported for ChatGPT on the Canadian Triage and Acuity Scale, while kappa values were comparable to GPT-4 on the Korean Triage and Acuity Scale. Undertriage detection varied markedly: MiMo V2.5 identified 56.6% of undertriaged cases versus 11.9% for DeepSeek V4 Pro. Lower-cost models performed equivalently to more expensive alternatives, though undertriage detection remained suboptimal across all models. These findings support the potential role of LLMs as adjunctive decision aids in resource-limited EDs, pending prospective validation in authentic clinical settings.
AI-StuntDetect: Development of Early Stunting Detection System for Young Children Based on CNN and MediaPipe Pose Asmaul Husna RS; Purnamawati Purnamawati; Hendra Jaya; Anas Arfandi
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.2041

Abstract

Stunting affects a substantial proportion of Indonesian children under five and is monitored through routine anthropometric measurement at posyandu, community health posts staffed largely by volunteer cadres. This measurement depends on calibrated instruments and consistent technique, which are not uniformly available across sites, so a low-cost screening aid could extend coverage where conventional practice is weakest. This study aims to develop AI-StuntDetect, a web-based system that estimates children's height and weight from a single mobile-phone photograph, classifies nutritional status, and evaluates whether its image-derived estimates add value beyond simple demographic predictors. A Convolutional Neural Network estimates weight directly from the image, while MediaPipe Pose landmarks are converted into body-proportion features and regressed onto height with a Random Forest. The models were developed on 117 children (1,593 images) using child-level partitioning and validated externally on 48 children from three independent posyandu. Performance was compared against a mean-value baseline and against age-only and age-plus-sex regressions fitted to the training data, and predicted heights were classified against the WHO and Kemenkes reference standards. Height estimation achieved an MAE of 4.092 cm and an R² of 0.863, reducing error by 31.9% relative to the strongest training-derived baseline. Weight estimation achieved an MAE of 1.771 kg and an R² of 0.590 but did not improve on the age-only regression (1.675 kg). Nutritional status was classified with 70.8% accuracy and a macro-averaged F1 of 0.546, with weak sensitivity for the intermediate stunted category (0.167). A single photograph therefore supports useful height estimation, whereas image-based weight estimation adds nothing beyond what age predicts. AI-StuntDetect is best characterized as an experimental screening prototype suited to triage under manual confirmation.
A Graph-Assisted Post-OCR Framework with Damerau-Levenshtein Spelling Correction for Drug Name Recognition from Pharmaceutical Packaging Fito Satrio; Muhammad Raafi'u Firmansyah; Diah Septiani
Journal of Electronics, Electromedical Engineering, and Medical Informatics Vol 8 No 4 (2026): October
Publisher : Department of Electromedical Engineering, POLTEKKES KEMENKES SURABAYA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35882/jeeemi.v8i4.2062

Abstract

Medication errors remain a persistent threat to patient safety, with a global prevalence of approximately 3%, rising to 11% in elderly care and 7% in intensive care units. In Indonesia, the large number of BPOM-registered pharmaceutical products increases the risk of drug misidentification from look-alike, sound-alike (LASA) names. At the same time, conventional OCR pipelines remain limited by heterogeneous layouts, font variation, and brand–generic name ambiguity. This study develops an integrated, lightweight post-OCR framework resolving both character-level OCR errors and brand–generic naming ambiguity without costly large language models. The main contributions are a Damerau-Levenshtein Spelling Correction (DLSC) module correcting character-level errors against a validated drug dictionary, and a Graph Traversal Correction (GTC) mechanism modeling brand–generic relationships as a directed graph for disambiguation. The framework was evaluated on a self-collected, web-scraped dataset of 9,500 Indonesian drug packaging images spanning 3,678 brand labels, split 80:20 at the brand-label level; three pretrained OCR engines (Tesseract, EasyOCR, PaddleOCR) were benchmarked without domain-specific fine-tuning under both CPU and GPU configurations, with the training subset serving only as a consistency check rather than fitting data, since DLSC and GTC are rule-based and undergo no parameter training. PaddleOCR + DLSC + GTC achieved the best performance, raising accuracy from 50.94% (Raw OCR) to 96.13% on training and 96.69% on testing data, with WER reduced to 3.82%/3.44% and CER to 3.64%/3.76%. Paired McNemar statistical testing confirmed GTC’s improvement over DLSC is significant across all three engines, and baseline comparison confirmed this gain stems specifically from explicit transposition modeling rather than generic dictionary correction. GTC added only 0.05–0.06 ms beyond DLSC regardless of CPU or GPU configuration. Findings indicate that combining lightweight edit-distance correction with graph-based disambiguation substantially improves drug-name recognition accuracy under resource-constrained, pretrained-model deployment conditions

Page 37 of 37 | Total Record : 368

 33   34   35   36   37 

Filter by Year

2019 2026


Reset
Filter By Issues
All Issue Vol 9 No 1 (2027): January Vol 8 No 4 (2026): October Vol 8 No 3 (2026): July Vol 8 No 2 (2026): April Vol 8 No 1 (2026): January Vol 7 No 4 (2025): October Vol 7 No 3 (2025): July Vol 7 No 2 (2025): April Vol 7 No 1 (2025): January Vol 6 No 4 (2024): October Vol 6 No 3 (2024): July Vol 6 No 2 (2024): April Vol 6 No 1 (2024): January Vol 5 No 4 (2023): October Vol 5 No 3 (2023): July Vol 5 No 2 (2023): April Vol 5 No 1 (2023): January Vol 4 No 4 (2022): October Vol 4 No 3 (2022): July Vol 4 No 2 (2022): April Vol 4 No 1 (2022): January Vol 3 No 3 (2021): October Vol 3 No 2 (2021): July Vol 3 No 1 (2021): January Vol 2 No 3 (2020): October Vol 2 No 2 (2020): July Vol 2 No 1 (2020): January Vol 1 No 2 (2019): October Vol 1 No 1 (2019): July More Issue