Article Per Year (5 Year)
p-Index From 2021 - 2026
0.882
P-Index
This Author published in this journals
All Journal Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI) Viva Medika: Jurnal Kesehatan, Kebidanan dan Keperawatan Journal of Advanced Health Informatics Research Control Systems and Optimization Letters
Annastasya Nabila Elsa Wulandari
Universitas Harapan Bangsa
Aerospace Engineering Automotive Engineering Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering Health Professions Medicine & Pharmacology Nursing Public Health

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

Found 5 Documents
Search

 1 
Lightweight Hybrid Linformer-Mamba U-Net for Efficient Retinal Microaneurysm Segmentation Arif Setia Sandi Ariyanto; Deny Nugroho Triwibowo; Agriby Diandra Chaniago; Indah Trivilia; Annastasya Nabila Elsa Wulandari
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 11 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v11i4.31598

Abstract

Diabetic retinopathy is a major microvascular complication of diabetes and a leading cause of vision loss among the working-age population. Microaneurysms (MAs), as the earliest clinical indicators of disease progression, remain challenging to segment due to their small size, low contrast, and extreme class imbalance. This study proposes a lightweight hybrid U-Net architecture for microaneurysm segmentation in retinal fundus images, designed to balance detection sensitivity and computational efficiency for deployment in resource-constrained environments. The proposed architecture integrates depthwise separable convolutions for efficient local feature extraction, a Transformer-Lite bottleneck based on Linformer self-attention for global contextual modeling, and a Mamba State Space Model (SSM)–based decoder to enhance feature propagation and spatial continuity.  The research contribution of this work is threefold: the introduction of an efficient hybrid U-Net combining Linformer and Mamba SSM for microaneurysm segmentation; a deployment-oriented evaluation protocol that explicitly distinguishes patch-level learning behavior from full-image reconstruction performance; and a transparent analysis of false positive behavior under extreme background dominance.  Experiments were conducted on the IDRiD dataset, consisting of 81 retinal images, using patient-level data splitting prior to patch extraction to prevent data leakage.  The results indicate that while patch-level evaluation demonstrates effective lesion-centric learning, deployment-realistic full-image evaluation reveals a notable performance degradation caused by false positive accumulation in extensive background regions. Nevertheless, the model maintains high recall, indicating preserved lesion sensitivity. These findings suggest that lightweight architectural design can deliver meaningful performance and is well suited for screening-oriented decision-support systems that prioritize efficiency and sensitivity.
Reproducible Biomedical NER and Proxy Relation Extraction for Drug–Adverse Event Analysis in Breast Cancer Deny Nugroho Triwibowo; Hadi Jayusman; Rachman Hidayat; Anisya; Annastasya Nabila Elsa Wulandari
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 12 No. 1 (2026): March
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v12i1.31594

Abstract

Pharmacovigilance requires automated systems to extract biomedical entities and their relationships from text, as manual processes are inefficient and prone to error. This study develops a reproducible pipeline for Named Entity Recognition (NER) and pattern-based proxy relation formation, focusing on drug side effects related to breast cancer. The research contribution is twofold: a domain-specific annotated dataset for pharmacovigilance NER, and a reproducible pipeline for proxy-based relation analysis. The experimental setup combines MobileBERT, DistilBERT, TinyBERT, and ALBERT. Evaluation is conducted using accuracy, precision, recall, F1-score, ROC AUC, and computational efficiency metrics. The results show that ALBERT achieves the highest NER performance (F1-score = 0.9261), while DistilBERT attains the best ROC AUC (0.9037). TinyBERT is the most efficient model, with 4.57 million parameters, 4.68 G FLOPs, and an average training time of 45.8 seconds per scenario. The proposed pipeline demonstrates a trade-off between accuracy and computational efficiency under the evaluated setting. The generated relations act as sentence-level proxy indicators of potential drug–adverse event associations and serve as a preliminary triage layer requiring expert validation rather than a high-precision system. However, the approach does not account for negation, uncertainty, or cross-sentence context, which may introduce false positive associations. Despite these limitations, the pipeline provides a reproducible baseline for exploratory pharmacovigilance analysis.
Understanding Large Language Models: A Review Annastasya Nabila Elsa Wulandari; Purwono Purwono; Alfian Ma’arif; Noorulden Basil; Hamzah M. Marhoon
Control Systems and Optimization Letters Vol 4, No 2 (2026)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/csol.v4i2.292

Abstract

Large Language Models (LLMs) have experienced rapid development and have been established as the dominant paradigm in modern Natural Language Processing (NLP), with high performance demonstrated across various language understanding and generation tasks. Increasing architectural complexity has led to the need for a structured conceptual framework to explain how architectural design, training paradigms, and inference mechanisms are collectively associated with model behavior. A conceptual and analytical review of LLMs is presented in this article through an examination of the relationship between Transformer-based architectures, multi-stage training processes, and the resulting capabilities and limitations. Encoder-only, decoder-only, and encoder–decoder architectural variants are examined in relation to structural characteristics and functional implications. The roles of pretraining, supervised fine-tuning, and instruction tuning are analyzed to clarify how output characteristics are shaped during model development. This study emphasizes how architectural and training strategies causally influence generative capabilities and inherent limitations. Fundamental issues, including hallucination, bias, data dependency, computational cost, and evaluation challenges, are critically examined as consequences of the probabilistic modeling paradigm adopted in LLMs. This review contributes a structured analytical perspective for evaluating LLMs design choices and their operational consequences, supporting more informed development and deployment practices.
AI Agents for Nursing Task Augmentation: A Focused Literature Overview of Clinical, Operational, Educational, and Governance Implications Sony Kartika Wibisono; Purwono Purwono; Muhammad Ahmad Baballe; Imam Ahmad Ashari; Annastasya Nabila Elsa Wulandari
Viva Medika Vol 19 No 2 (2026)
Publisher : LPPM Universitas Harapan Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35960/vm.v19i2.2250

Abstract

The rapid development of artificial intelligence has accelerated the emergence of AI agents, defined as autonomous or semi-autonomous systems that integrate perception, contextual reasoning, decision-making, interaction, and action within defined workflows. Although AI in nursing has been widely reviewed, existing syntheses often combine predictive models, generative tools, decision-support systems, and agent-based architectures, leaving the specific contributions and implementation maturity of AI agents insufficiently differentiated. This focused literature overview examined peer-reviewed publications published between 2021 and 2025 using targeted database searches and a structured narrative synthesis. The review classified the evidence according to agent architecture, automated nursing tasks, implementation maturity, and reported clinical and operational implications. Three overlapping architectural categories were identified: LLM-driven agents, cognitive agents, and multi-agent systems. Applications were concentrated in clinical documentation and handover, predictive monitoring, medication safety, triage, and clinical decision support. The evidence suggests potential improvements in timeliness, documentation consistency, risk detection, workflow integration, and the reduction of repetitive administrative work. The maturity of the evidence varied considerably. Monitoring and sensor-enabled safety systems showed closer links to rsmitheal-world practice, whereas LLM-driven documentation and multi-agent triage systems were more frequently supported by conceptual, prototype, or simulation-based evidence. The synthesis therefore supports supervised task augmentation rather than the replacement of professional nursing judgment. Major implementation requirements include system reliability, bias mitigation, transparent accountability, human oversight, workforce competency, organizational readiness, and clinical governance. Nursing education should prepare practitioners to evaluate AI-generated outputs, recognize uncertainty and inappropriate recommendations, and apply appropriate escalation procedures. Future research should prioritize real-world, multisite, and longitudinal evaluations that measure patient safety, workload redistribution, verification burden, and clinical outcomes. This review provides a focused conceptual distinction between conventional AI decision-support tools and agent-based systems while integrating their clinical, operational, educational, and governance implications.
Explainable AI for Mental Health and Biomedical Decision Systems: A Comprehensive Review Pramesti Dewi; Purwono Purwono; Annastasya Nabila Elsa Wulandari; Indah Trivilia
Journal of Advanced Health Informatics Research Vol. 4 No. 2 (2026)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jahir.v4i2.376

Abstract

The application of Artificial Intelligence (AI) in mental health is experiencing rapid development, while algorithmic transparency and clinical translation readiness still face fundamental obstacles. This review synthesizes empirical findings related to the use of Explainable Artificial Intelligence (XAI) in mental health and biomedical decision systems, focusing on three evaluative aspects, namely explainability architecture, validation strength, and clinical integration. The literature search followed the PRISMA 2020 guidelines across five major databases for publications from 2020 to 2026 and yielded nine studies that met the inclusion criteria. The synthesis results show the dominance of post-hoc approaches, particularly SHAP, which are commonly applied to ensemble and boosting models, while intrinsic models and counterfactual approaches are still rarely used. The majority of studies rely on internal validation, while independent external validation and prospective application in real clinical workflows are relatively limited. User-based evaluation of explainability has also been understudied, with algorithmic transparency more often understood as technical feature attribution rather than as a verified mechanism in clinical decision-making. These findings indicate a persistent gap between methodological advances and the level of clinical translation maturity. Explainability has not been systematically integrated with robust validation designs or user-oriented evaluations. This review proposes a translation evaluation framework that combines technical and clinical dimensions to assess the readiness for XAI implementation more comprehensively. The development of XAI in mental health requires evaluation standardization, strengthened external validation, and prospective testing focused on clinical impact and user trust.
Co-Authors Agriby Diandra Chaniago Alfian Ma’arif Anisya Arif Setia Sandi A. Deny Nugroho Triwibowo Hadi Jayusman Hamzah M. Marhoon Imam Ahmad Ashari, Imam Ahmad Indah Trivilia Muhammad Ahmad Baballe Noorulden Basil Pramesti Dewi Purwono Purwono Purwono, Purwono Rachman Hidayat Sony Kartika Wibisono