Tinjauan Sistematis dan Penilaian Kelayakan Meta-Analisis Algoritma Klasifikasi untuk Deteksi Keaslian Minyak Atsiri sebagai Bahan Parfum

Yoga Aji Sukma; Desi Permata Sari

doi:10.37729/intek.v9i1.7675

Yoga Aji Sukma Institut Teknologi Sumatera
Desi Permata Sari Institut Teknologi Sumatera

DOI: https://doi.org/10.37729/intek.v9i1.7675

Abstract

Keaslian minyak atsiri sebagai bahan parfum memerlukan quality control yang mampu mengubah data sensorik, kromatografik, dan spektroskopik menjadi keputusan yang dapat diaudit. Penelitian ini menyintesis evidence mengenai algoritma klasifikasi untuk deteksi keaslian, diskriminasi, mutu, asal, dan indikasi adulterasi minyak atsiri dari sudut pandang informatika. Metode yang digunakan adalah systematic review berbasis PRISMA/PRISMA-DTA melalui literature search OpenAlex dan PubMed/PMC, screening berbasis eligibility criteria, verifikasi full text, serta data extraction atas data modality/instrumen, algoritma, validation design, dan performance metric. Dari 160 rekaman awal, 36 studi masuk calon qualitative synthesis dan 5 studi memiliki quantitative classifier-performance data. Lima studi tersebut melibatkan electronic nose, HS-GC-MS, GC-FID, dan FTIR dengan algoritma LDA/QDA, SVM, ANN/MLP, Fuzzy ARTMAP, O2PLS-DA, dan HCA. Akurasi/CCR berada pada kisaran 82% hingga 100%, tetapi tidak digabungkan sebagai pooled accuracy estimate karena tujuan klasifikasi, data modality/instrumen, validation design, validation sample, dan risk of bias tidak sebanding. Sintesis menunjukkan bahwa algoritma klasifikasi berpotensi mendukung autentikasi minyak atsiri apabila penelitian berikutnya melaporkan external validation, confusion matrix, jumlah independent validation sample, dan AI-based quality-control workflow secara transparan.

References

Type and magnitude of non-compliance and adulteration in neroli, mandarin and bergamot essential oils purchased online: potential consumer vulnerability, Scientific Reports, 2021. doi: 10.1038/s41598-021-90307-2.

Classification and Identification of Essential Oils from Herbs and Fruits Based on a MOS Electronic-Nose Technology, Chemosensors, 2021. doi: 10.3390/chemosensors9060142.

Classification of Bitter Orange Essential Oils According to Fruit Ripening Stage by Untargeted Chemical Profiling and Machine Learning, Sensors, 2018. doi: 10.3390/s18061922.

Proposal of a Gas Sensor-Based Device for Detecting Adulteration in Essential Oil of Cistus ladanifer, Sustainability, 2023. doi: 10.3390/su15043357.

Integration of FTIR Spectroscopy, Volatile Compound Profiling, and Chemometric Techniques for Advanced Geographical and Varietal Analysis of Moroccan Eucalyptus Essential Oils, Sensors, 2024. doi: 10.3390/s24227337.

Performance Comparison of Fuzzy ARTMAP and LDA in Qualitative Classification of Iranian Rosa damascena Essential Oils by an Electronic Nose, Sensors, 2016. doi: 10.3390/s16050636.

Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies: The PRISMA-DTA Statement, JAMA, 2018. doi: 10.1001/jama.2017.19163.

PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods, BMJ, 2025. doi: 10.1136/bmj-2024-082505.

PROBAST: Prediction model study Risk Of Bias Assessment Tool, PROBAST, 2019. URL: https://www.probast.org/wp-content/uploads/2020/02/PROBAST_20190515.pdf.

QUADAS-2 Tool for the Quality Assessment of Diagnostic Accuracy Studies, NCBI Bookshelf, 2015. URL: https://www.ncbi.nlm.nih.gov/books/NBK333349/.

Cebi, Nur, Quantification of the Geranium Essential Oil, Palmarosa Essential Oil and Phenylethyl Alcohol in Rosa damascena Essential Oil Using ATR-FTIR Spectroscopy Combined with Chemometrics, Foods, vol. 10, no. 8, pp. 1848, 2021. doi: 10.3390/foods10081848.

Agatonovic-Kustrin, Snezana et al., Essential Oil Quality and Purity Evaluation via FT-IR Spectroscopy and Pattern Recognition Techniques, Applied Sciences, vol. 10, no. 20, pp. 7294, 2020. doi: 10.3390/app10207294.

Rasekh, Mansour et al., Performance Analysis of MAU-9 Electronic-Nose MOS Sensor Array Components and ANN Classification Methods for Discrimination of Herb and Fruit Essential Oils, Chemosensors, vol. 9, no. 9, pp. 243, 2021. doi: 10.3390/chemosensors9090243.

Nurani, Laela Hayu et al., Authentication of Patchouli Oil from Various Growing Regions Using GC-MS Method with Chemometric Combination on the Products in the Market, Jurnal Farmasi Sains dan Praktis, pp. 90--96, 2024. doi: 10.31603/pharmacy.v10i1.10523.

Allenspach, Martina et al., Authenticity Control of Pine Sylvestris Essential Oil by Chiral Gas Chromatographic Analysis of Alpha-Pinene, Scientific Reports, vol. 11, no. 1, 2021. doi: 10.1038/s41598-021-96356-x.