Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal International Journal of Public Health Excellence (IJPHE) Jurnal Multidisiplin West Science
Nurul Qomariah, Dinial Utami
Unknown Affiliation
Economics, Econometrics & Finance Health Professions Medicine & Pharmacology Nursing Public Health

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

Found 2 Documents
Search

 1 
Automatic Pill Detection Using Faster R-CNN with an AlexNet Backbone Elvira, Ade Irma; Kurniasar, Arvita Agus; Maulana, Bima Wahyu; Nurul Qomariah, Dinial Utami
Jurnal Multidisiplin West Science Vol 4 No 12 (2025): Jurnal Multidisiplin West Science
Publisher : Westscience Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58812/jmws.v4i12.3068

Abstract

Object detection is a crucial component in the development of automated systems in the healthcare domain, particularly in pharmaceutical applications such as pill identification and management. One of the main challenges in image-based pill detection systems is achieving high accuracy and robust generalization under variations in pill shape, color, and illumination conditions. This study applies the Faster R-CNN framework with an AlexNet backbone to detect and classify pill objects in digital images. The model is trained using multiple epoch configurations to analyze the effect of training duration on detection performance. Experimental results show that the proposed approach achieves an accuracy of up to 98%, demonstrating strong detection capability. Increasing the number of training epochs improves the stability and consistency of pill recognition. These results indicate that AlexNet-based Faster R-CNN is effective for pharmaceutical applications, particularly in drug distribution, packaging, and pill counting systems that require high precision and reliability.
Automatic Pill Counting Using YOLOv8 to Improve Medication Distribution Accuracy Nurul Qomariah, Dinial Utami; Elvira, Ade Irma; Agus Kurniasari, Arvita; Wahyu Maulana, Bima
International Journal of Public Health Excellence (IJPHE) Vol. 5 No. 2 (2026): January-May
Publisher : PT Inovasi Pratama Internasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55299/ijphe.v5i2.1724

Abstract

Object detection is a critical component in various modern applications, including healthcare systems, smart agriculture, and industrial automation. The main challenge in developing detection systems lies in achieving high accuracy and strong generalization capabilities under diverse image conditions. This study aims to implement and evaluate the YOLOv8 model, a detection method known for its speed and efficiency. The model is trained using two scenarios—10 epochs and 50 epochs—to examine the impact of training duration on system performance. Evaluation results show that training for 10 epochs produces very good performance, with a precision of 0.98, recall of 0.94, and mAP of 0.98. Increasing the training to 50 epochs yields even more optimal results, achieving a precision of 0.99, recall of 1.00, and mAP of 0.99. Based on these findings, YOLOv8 demonstrates excellent adaptability to the dataset and is suitable for real-time detection applications that require high accuracy
Co-Authors Agus Kurniasari, Arvita Elvira, Ade Irma Kurniasar, Arvita Agus Maulana, Bima Wahyu Wahyu Maulana, Bima