Mohasin Ali Miah
Research Subjects
Mohasin Ali Miah has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
Deepam Gangopadhyay
2 shared publicationsSumit Kundu
2 shared publicationsRunu Banerjee Roy
1 shared publication- Runu Banerjee Roy
Department of Instrumentation and Electronics Engineering, Jadavpur University, Kolkata, India
1 shared publication Tanmoy Ghosh
1 shared publication
Latest Publications
- Open Access
- 7 May 2026
- Published by Elsevier in SSRN Electronic Journal
p-Cymene Profiling in Cumin Essential Oil by Developing a Molecular Imprinted Quartz Crystal Microbalance Sensor Development and Machine Learning
A novel quartz crystal microbalance (QCM)-based sensor, PCy-MIP-QCM, was developed for the selective and sensitive detection of p-cymene, a key flavor compound in cumin essential oil (CEO) widely used in food and pharmaceutical applications. The sensor was fabricated by coating the QCM electrode with a molecularly imprinted polymer (MIP) synthesized using p-cymene as the template and methacrylic acid (MAA) and styrene as functional monomers. The MIP layer enabled selective adsorption of p-cymene, producing measurable frequency shifts corresponding to mass changes. The sensor exhibited high sensitivity (0.198 Hz/ppm), excellent selectivity, repeatability, reproducibility, and long-term stability, along with a wide linear detection range (10–400 ppm) and a low limit of detection (7.020 ppm). A strong correlation between sensor responses and gas chromatography–mass spectrometry (GC–MS) measurements was observed for real CEO samples. To further enhance predictive capability, four regression models like Principal Component Regression (PCR), Partial Least Squares Regression (PLSR), K-Nearest Neighbor Regression (KNNR), and Ensemble Multiple Linear Regression (EMLR) were applied. Among these, EMLR has showed the best performance with the lowest RMSE (0.034), highest R² (0.998), and an average predictive accuracy of 99.095%. A paired t-test comparing predicted and GC–MS values yielded p > 0.050 (95% confidence) for all the models. This indicated no significant differences between GC-MS and the models, thereby confirming the reliability of the prediction framework. These results highlight the sensor’s potential as a rapid, reliable, and cost-effective tool for aroma profiling and quality assessment of spice-derived essential oils.
- 7 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open Accessp-Cymene Profiling in Cumin Essential Oil by Developing a Molecular Imprinted Quartz Crystal Microbalance Sensor Development and Machine Learning
A novel quartz crystal microbalance (QCM)-based sensor, PCy-MIP-QCM, was developed for the selective and sensitive detection of p-cymene, a key flavor compound in cumin essential oil (CEO) widely used in food and pharmaceutical applications. The sensor was fabricated by coating the QCM electrode with a molecularly imprinted polymer (MIP) synthesized using p-cymene as the template and methacrylic acid (MAA) and styrene as functional monomers. The MIP layer enabled selective adsorption of p-cymene, producing measurable frequency shifts corresponding to mass changes. The sensor exhibited high sensitivity (0.198 Hz/ppm), excellent selectivity, repeatability, reproducibility, and long-term stability, along with a wide linear detection range (10–400 ppm) and a low limit of detection (7.020 ppm). A strong correlation between sensor responses and gas chromatography–mass spectrometry (GC–MS) measurements was observed for real CEO samples. To further enhance predictive capability, four regression models like Principal Component Regression (PCR), Partial Least Squares Regression (PLSR), K-Nearest Neighbor Regression (KNNR), and Ensemble Multiple Linear Regression (EMLR) were applied. Among these, EMLR has showed the best performance with the lowest RMSE (0.034), highest R² (0.998), and an average predictive accuracy of 99.095%. A paired t-test comparing predicted and GC–MS values yielded p > 0.050 (95% confidence) for all the models. This indicated no significant differences between GC-MS and the models, thereby confirming the reliability of the prediction framework. These results highlight the sensor’s potential as a rapid, reliable, and cost-effective tool for aroma profiling and quality assessment of spice-derived essential oils.