{"title":"Research on Detection Model of Penicillin Potency Content based on Near-Infrared Spectroscopy Technology.","authors":"Jianxia Wang, Nan Shen, Xiaojun Wang, Yan Wang","doi":"10.2174/0115734099366520250226084836","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>The potency content of penicillin serves as a crucial indicator for measuring its pharmacological effects, playing a vital role in quality control and clinical applications. In recent years, with the continuous improvement of production efficiency and quality requirements in the pharmaceutical industry, the need for high-frequency monitoring of drug potency has become increasingly urgent. Infrared spectroscopy, as an emerging research tool, has demonstrated immense potential in the field of drug potency testing.</p><p><strong>Objective: </strong>The objective of this study is to develop a real-time monitoring model for penicillin potency content utilizing near-infrared (NIR) spectroscopy data. This model aims to enable rapid and accurate detection of potency content during the penicillin production process, ultimately enhancing production efficiency and reducing costs.</p><p><strong>Method: </strong>During the penicillin production process, NIR spectroscopy data from penicillin samples were scanned and collected to form a comprehensive dataset. Five distinct spectral preprocessing methods were combined with three regression models to construct detection models. By comparing the performance of different combinations, the optimal model configuration was identified.</p><p><strong>Results: </strong>The optimal model configuration identified in this study integrates the Savitzky-Golay filtering method with ridge regression. Under this optimal model, the coefficient of determination for the test set reached 0.990669, indicating an extremely high degree of agreement between the model's predicted values and the actual measured values. This real-time monitoring model for penicillin potency content can be applied as a rapid and non-destructive monitoring method in factory settings.</p><p><strong>Conclusion: </strong>This study successfully developed a real-time monitoring model for penicillin potency based on NIR spectroscopy technology. The research findings not only provide strong support for potency monitoring during the penicillin production process but also offer new insights and methodologies for non-destructive testing of other pharmaceuticals and chemicals.</p>","PeriodicalId":93961,"journal":{"name":"Current computer-aided drug design","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current computer-aided drug design","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2174/0115734099366520250226084836","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Background: The potency content of penicillin serves as a crucial indicator for measuring its pharmacological effects, playing a vital role in quality control and clinical applications. In recent years, with the continuous improvement of production efficiency and quality requirements in the pharmaceutical industry, the need for high-frequency monitoring of drug potency has become increasingly urgent. Infrared spectroscopy, as an emerging research tool, has demonstrated immense potential in the field of drug potency testing.
Objective: The objective of this study is to develop a real-time monitoring model for penicillin potency content utilizing near-infrared (NIR) spectroscopy data. This model aims to enable rapid and accurate detection of potency content during the penicillin production process, ultimately enhancing production efficiency and reducing costs.
Method: During the penicillin production process, NIR spectroscopy data from penicillin samples were scanned and collected to form a comprehensive dataset. Five distinct spectral preprocessing methods were combined with three regression models to construct detection models. By comparing the performance of different combinations, the optimal model configuration was identified.
Results: The optimal model configuration identified in this study integrates the Savitzky-Golay filtering method with ridge regression. Under this optimal model, the coefficient of determination for the test set reached 0.990669, indicating an extremely high degree of agreement between the model's predicted values and the actual measured values. This real-time monitoring model for penicillin potency content can be applied as a rapid and non-destructive monitoring method in factory settings.
Conclusion: This study successfully developed a real-time monitoring model for penicillin potency based on NIR spectroscopy technology. The research findings not only provide strong support for potency monitoring during the penicillin production process but also offer new insights and methodologies for non-destructive testing of other pharmaceuticals and chemicals.