Prediction of Milk Transfer of Drugs Using Machine Learning Methods / 医薬品情報学

Objective: Currently, limited information is available on the milk transfer properties of drugs when consumed by lactating women. Therefore, we aim to construct a prediction model of milk transfer of drugs using machine learning methods.Methods: We obtained data from Hale’s Medications & Mothers’ Milk (MMM) and SciFinder®, and then constructed the datasets. The physicochemical and pharmacokinetic data were used as feature variables with M/P ratio ≥ 1 and M/P ratio < 1 as the objective variables, classified into two groups as the classification of milk transferability. In this study, analyses were conducted using machine learning methods: logistic regression, linear support vector machine (linear SVM), kernel method support vector machine (kernel SVM), random forest, and k-nearest neighbor classification. The results were compared to those obtained with the linear regression equation of Yamauchi et al. from a previous study. The analysis was performed using scikit-learn (version 0.24.2) with python (version 3.8.10).Results: Model construction and validation were performed on the training data comprising 159 drugs. The results revealed that the random forest had the highest accuracy, area under the receiver operating characteristic curve (AUC), and F value. Additionally, the results with test data A and B (n = 36, 31), which were not used for training, showed that both F value and accuracy for the random forest and the kernel method SVM exceeded those with the linear regression equation of Yamauchi et al. Conclusion: We were able to construct a predictive model of milk transferability with relatively high performance using a machine learning method capable of nonlinear separation. The predictive model in this study can be applied to drugs with unknown M/P ratios for providing a new source of information on milk transfer.

Development of an Application for Using the Pharmaceuticals and Medical Devices Agency Side Effect Dataset More Effectively / 医薬品情報学

<b>Objective: </b>The Pharmaceuticals and Medical Devices Agency (PMDA) discloses reports with accumulated side effect information in comma-separated value (CSV) format.  It is difficult to use the information in this type of text file because the amount of data is large and composed of multiple fields.  Therefore, we developed an application that presents the data in a way that is easier to read and understand.<br><b>Methods: </b>The application can display the whole dataset, or the search results of certain medicines and side effects within the database in Microsoft Access 2013.  It exports data from search results into an Excel spreadsheet organized by medicine and side effect.<br><b>Results: </b>This application makes it possible to understand statistics contained in the side effect dataset, such as the number of cases, the medicines, and the side effects themselves.  Moreover, the application allows the totaled search results for the medicines and the side effects to be graphed.  It also makes it possible to understand the sex and age distribution of patients, as well as the days elapsed before developing a side effect.<br><b>Conclusions: </b>Recently, the importance of information concerning the safety of medicine has increased.  This system could facilitate the effective use of side effect information and the creation of medicine risk management plans in medical institutions.

Construction Case with an Electronic Hospital Formulary to Be Able to Do Rapid Retrieval Using a Portable Terminal iPhone® / 医薬品情報学

<b>Objective: </b>The hospital is changing its formulary reference from paper-based to intranet.  There was concern that both paper-based and intranet versions of the formulary would be necessary.  Revising the paper-based hospital formulary each time package inserts are revised is difficult.  For your review we report on the creation of the iPhone® electronic formulary which enables rapid off-line formulary retrieval and easy updates while at the same time providing low cost service in a light device.<br><b>Methods: </b>The CSV (Comma Separated Value) of the hospital formulary dictionary was made using a standard personal computer.  The CSV data file was converted using JAMES2DIC into a HTML file format.  Next, the converted HTML file is transformed into the EPWING (Electronic Publishing WING) format using EBStudio.  Finally, we forward the EPWING dictionary file from the personal computer to the iPhone®.  The retrieval becomes possible by using EBPocket for iOS of EPWING/electronic book viewer software for the iPhone®.  The number of items was assumed to be 29 items thought for a lot of inquiries to exist.<br><b>Results: </b>We compared the paper-based formulary with the iPhone® electronic formulary.  As a result, the iPhone4® electronic formulary shortened the retrieval time, was smaller, lighter, and excellent at a lower price.<br><b>Conclusion: </b>The iPhone4® electronic formulary enables the user to perform complex full-text searches and retrieve information at a much higher speed than is possible with paper based formularies.  It has the additional advantage of seamless integration and deployment of formulary additions or reference material revisions.  We believe we have successfully created a practical electronic formulary.