Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Agrikultura Media Farmasi : Jurnal Ilmu Farmasi (Journal Of Pharmaceutical Science)
Harmiansyah, Harmiansyah
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Chemistry Immunology & microbiology Materials Science & Nanotechnology Medicine & Pharmacology

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

Found 2 Documents
Search

 1 
Deteksi Cerdas Penyakit Tanaman Kopi Robusta Berbasis Deep Learning Menggunakan Variasi YOLO Harmiansyah, Harmiansyah; Oviana, Ella Trilia; Fitrawan, Mhd Kadar; Putra, Pramana; Diptaningsari, Danar; Meidaliyantisyah, Meidaliyantisyah
Agrikultura Vol 36, No 3 (2025): Desember, 2025
Publisher : Fakultas Pertanian Universitas Padjadjaran

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24198/agrikultura.v36i3.64004

Abstract

Tanaman kopi menjadi populer karena produk minuman kopi yang memiliki aroma dan rasa unik. Ditengah populernya tanaman kopi terdapat permasalahan dalam pengendalian penyakit seperti bercak daun dan karat daun yang berdampak pada produksi tanaman kopi menurun. Sehingga dibutuhkan sistem deteksi cerdas berakurasi tinggi untuk mengidentifikasi jenis penyakit pada tanaman kopi sebagai langkah penanganan dini.Tujuan pada penelitian ini adalah implementasi menggunakan variasi model pralatih YOLO (You Only Look Once) untuk mendeteksi penyakit tanaman kopi robusta berdasarkan citra daun. Penelitian ini menggunakan 3 jenis model pralatih dari YOLO yaitu YOLOv5, YOLOv7 dan YOLOv8 dengan parameter hyperparameter yaitu 150 epoch, batch size 16 dan learning rate 0,001 sedangkan untuk optimizer yang digunakan adalah SGD (Stochastic Gradient Descent). Dataset penelitian adalah citra daun tanaman kopi yang didapatkan dari pengambilan manual menggunakan tools handphone dengan spesifiaksi kamera 20 MP berlokasi di Kebun Percobaan Tegineneng Natar, Balai Penerapan Modernisasi Pertanian (BRMP) Lampung. Dataset diberi augmentasi berupa shear, blur dan rotation. Berdasarkan hasil kinerja model deteksi objek berbasis YOLO, model terbaik yang didapatkan adalah YOLOv8 dengan nilai mAP@50 sebesar 99,5% dan mAP@50-95 adalah 94,6% dalam waktu training selama 1,748 jam. Testing yang dilakukan menggunakan YOLOv8 menghasilkan nilai evaluasi metrik yaitu nilai akurasi 100%, presisi 100%, recall 100% dan F1 score 100% untuk kelas daun sehat dan daun karat. Sedangkan kelas daun bercak mendapatkan nilai akurasi 94%, presisi 100%, recall 94% dan F1 score 97%.
Unveiling the Multitarget Anti-Aging Mechanisms of Andaliman (Zanthoxylum acanthopodium) Essential Oil through Integrated in Silico Network Pharmacology and Docking Analysis Yasir, Angga; Setiawan, Tirta; Harmiansyah, Harmiansyah
Media Farmasi: Jurnal Ilmu Farmasi Vol. 23 No. 1 (2026): March 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/mf.v23i1.31437

Abstract

Research on natural anti-aging agents often encounters challenges due to the complex and multifactorial nature of skin-aging mechanisms, as well as the limited exploration of essential oils with multitarget actions. Addressing these gaps, this study aimed to elucidate the multitarget molecular mechanisms of andaliman (Zanthoxylum acanthopodium) essential oil against skin aging using an integrated in silico network pharmacology and molecular docking approach. The study identified three dominant bioactive compounds, geranyl acetate, citronellal, and citronellol acetate, through literature-based GC–MS data. Their potential protein targets were predicted via Swiss Target Prediction and compared with aging-related genes, followed by protein–protein interaction (PPI) analysis, hub gene identification, and pathway enrichment using STRING, Cytoscape, and Shiny GO platforms. Molecular docking was then conducted to validate the interactions of key compounds with the core proteins. The analysis revealed six central target genes (CDK1, CCND1, EGFR, SRC, GSK3B, and HDAC1) that regulate cell proliferation, adhesion, and epigenetic modification. Pathway enrichment indicated significant involvement in cell cycle, focal adhesion, and thyroid hormone signaling pathways. Docking simulations demonstrated stable ligand–protein interactions, with binding affinities ranging from −5.67 to −7.37 kcal/mol, particularly between geranyl acetate and CDK1 as well as citronellol acetate and CDK1. These findings provide comprehensive computational evidence that andaliman essential oil exerts anti-aging activity through simultaneous modulation of proliferative, adhesive, and epigenetic pathways. The results reinforce its potential application as a natural multitarget active ingredient for anti-aging cosmetic formulations, offering a molecular foundation for further experimental validation in biological models.
Co-Authors Diptaningsari, Danar Fitrawan, Mhd Kadar Meidaliyantisyah, Meidaliyantisyah Oviana, Ella Trilia Putra, Pramana Tirta Setiawan, Tirta Yasir, Angga