Development of A Chromatographic Database for Applying Machine Learning for Drug Analysis
Ashwini M. Mawal*1, Ishit S. Rindhe2, Shaikh Hanif Ashfaque3, Suresh K. Gaikwad4, Mahesh Bhadane5, Parag Chavan6
1Ashwini M. Mawal, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
2Ishit S. Rindhe, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
3Shaikh Hanif Ashfaque, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
4Suresh K Gaikwad, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
5Mahesh Bhadane, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
6Dr. Parag Chavan, Assistant Professor, Department of Chemistry, School of Science, Sandip University, Nashik, Maharashtra, India.
Manuscript received on 01 June 2026 | First Revised Manuscript received on 10 June 2026 | Second Revised Manuscript received on 02 July 2026 | Manuscript Accepted on 15 July 2026 | Manuscript published on 30 July 2026 | PP: 10-14 | Volume-13 Issue-7, July 2026 | Retrieval Number: 100.1/ijies.L113712121225 | DOI: 10.35940/ijies.L1137.12121225
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This study introduces a machine learning-based framework to optimise chromatographic data analysis in pharmaceutical quality assurance. We developed a chromatographic peak database and trained an XGBoost model— widely adopted in industry for its scalability and robustness across diverse data ranges—to predict the acceptability of chromatograms. The dataset included 100 CSV files representing real-world chromatographic waveforms, from which we extracted features such as retention time, peak height, and peak width for training. Our model achieved 65% accuracy, 3% precision, 38% recall, and an average RMSE, highlighting ML’s potential to augment current manual quality assessment workflows. No benchmark model existed for this application; therefore, our comparative analysis with the Random Forest model provided the only available baseline. Unlike traditional, time-consuming manual reviews—where quality analysts may evaluate up to 100 chromatograms daily, each taking approximately five minutes—our predictive system streamlines the process by flagging only non-compliant cases, effectively reducing human workload by over 90%. This innovation lays foundational work for AI-driven chromatography, especially in areas where literature integrating ML with chromatographic analysis remains sparse.
Keywords: Gas Chromatography, CSV, Data Processing, Waveform Plotting, Peak Identification, Training Data, AI Algorithm, Machine Learning, Model Building, Model Testing, Predicted Model, Manual Input.
Scope of the Article: Computer Science and Engineering