Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Sinkron : Jurnal dan Penelitian Teknik Informatika International Journal of Artificial Intelligence in Medical Issues
Ahmad Kholish Fauzan Shobiry
Universitas Negeri Malang
Computer Science & IT Dentistry Health Professions Medicine & Pharmacology Public Health

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

Found 2 Documents
Search

 1 
Hybrid Feature Benchmark for Blood Cell Classification Using ResNet50 and EfficientNetV2 Features with SVM and ANN Classifiers via Unsupervised Segmentation Ahmad Kholish Fauzan Shobiry; Rahma Puspitasari
International Journal of Artificial Intelligence in Medical Issues Vol. 3 No. 2 (2025): International Journal of Artificial Intelligence in Medical Issues
Publisher : Yocto Brain

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56705/ijaimi.v3i2.364

Abstract

Automated blood cell classification supports hematological diagnosis by providing objective and efficient analysis, but end-to-end deep learning models often require substantial computational resources that limit deployment on low-resource clinical devices. This study evaluates whether frozen deep features extracted from EfficientNetV2B0 or ResNet50 provide better separability for the eight BloodMNIST classes, and examines which classical classifier offers the most practical balance of accuracy, model size, and training time. The BloodMNIST dataset, consisting of 11,959 training images, 1,712 validation images, and 3,421 test images, is processed using data augmentation and Otsu-based unsupervised segmentation before the resulting masks are replicated into three channels and passed into pretrained ImageNet CNNs used strictly as frozen feature extractors. The extracted features are classified using Support Vector Machine with grid search, K-Nearest Neighbor, Artificial Neural Network, and Random Forest, with performance assessed through accuracy, precision, recall, and F1-score. EfficientNetV2 with Support Vector Machine achieves the highest performance, reaching 76.8% test accuracy, 75.3% precision, 72.6% recall, and a 73.6% F1-score, while EfficientNetV2 with Artificial Neural Network provides a comparable 76.2% accuracy and a 73.0% F1-score with a compact 2 MB model size. These findings highlight a clear trade-off between accuracy, model size, and computational cost, demonstrating that hybrid deep-feature pipelines offer lightweight and effective solutions for blood cell classification in resource-constrained clinical settings
Incremental Effects of Augmentation Strategies on Pre-Augmented Biomedical Waste Detection Using YOLOv11n Jevri Tri Ardiansah; Ahmad Kholish Fauzan Shobiry; Anik Nur Handayani; Mohammad Muzayyin Amrulloh; Taiga Haruta
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16531

Abstract

Biomedical waste carries infectious and hazardous risk that makes accurate automated sorting valuable, and object detection offers a path toward automation, yet published studies rarely measure how much online data augmentation contributes once training images carry offline augmentation and mean Average Precision alone can conceal how augmentation shifts the balance between missed detections and false alarms.This study measures the incremental effect of Mosaic, MixUp, and Copy-Paste augmentation on YOLOv11n trained for biomedical waste detection, and verifies whether each augmentation executes as configured. We designed a 2×2×2 factorial ablation across eight configurations, trained YOLOv11n three times per configuration with different random seeds on a 14-class biomedical waste dataset and evaluated each run's best checkpoint on a held-out test set using mean Average Precision, precision, recall, and a corrected confusion matrix retaining the background class. We verified framework behavior through prediction-level identity, loss-level comparison, and instrumented tracing. Configurations combining Mosaic and MixUp raised mean Average Precision by 0.0053 over baseline, MixUp alone raised recall from 0.8950 to 0.9071, and Mosaic with MixUp raised precision to 0.9650. False negatives outnumbered false positives three to one across every configuration, identifying class-versus-background rather than inter-class confusion as the dominant failure mode. Copy-Paste showed zero measurable effect once verified, consistent with its restriction to segmentation tasks in the underlying framework. Online augmentation produces modest but distinguishable, metric-specific gains on pre-augmented biomedical waste data, and verifying framework behavior before attributing results to an augmentation strategy is necessary for reliable reporting.
Co-Authors Anik Nur Handayani Jevri Tri Ardiansah Mohammad Muzayyin Amrulloh Rahma Puspitasari Taiga Haruta