Issue link: https://www.e-digitaleditions.com/i/1462155
Pharmaceutical Technology QUALITY AND REGULATORY SOURCEBOOK EBOOK MARCH 2022 21 able and reproducible multivariate mechanisms can be used for good practice (GxP) purposes; AI must be managed in this way to open the per- ceived black box. There are some industr y initiatives to help overcome this hurdle. For example, Xavier Uni- versity has several AI working teams that have been working with industry leaders in coordina- tion with FDA to develop Good Machine Learn- ing Practices (GMLP), an AI Maturity Model, show real-use cases, and drive other initiatives related to AI implementation. The Parenteral Drug Association has also published an AI Al- gorithm Qualification study (4). The goal of this research initiative was to provide a methodol- ogy based on a design of experiment (DoE) ap- proach to agnostically qualify the Isolation Forest algorithm. This strategy can be applied to real datasets containing different sources of variabil- ity or using synthetic datasets containing sev- eral variables with different controlled features that are the object of study, such as correlated variables, various standard deviations, orders of magnitude, etc. An experimental engine was specifically built as part of the project. The strat- egy was focused on detecting defects in a motor using an AI algorithm for outlier detection. The experiments carried out have been defined using a DoE to cover all operational conditions. The resulting guidelines can be applied to other AI algorithms for regulated drug and medical device manufacturing environments. Following this blueprint, pharmaceutical and biotech companies can use AI/ML to optimize their highly regulated manufacturing processes. The use of batch data to streamline production, reduce costs, improve yield, and keep up with to- day's increasingly complex regulatory landscape will enable Pharma 4.0 and the business transfor- mation benefits that come with it. Qualification method AI/ML allows pharmaceutical manufacturers and life sciences companies to interrogate, analyze, and manage huge data sets generated by complex systems. Using qualified AI production chain algorithms provides new ways to scale operations and production. A qualified algorithm presumes a ready-to-use state, making the model validation process easier. The concept of qualification is associated with systems, facilities, or equipment and is a process to ensure that these things are achieving the expected acceptance criteria defined in quality attributes and parameters. In a regulatory context, "quali- fication" is the status of quality assurance for sys- tems, facilities, or equipment supported by docu- mented evidence that proves the right use of them. Qualification is different from validation. "Valida- tion" is the documented evidence to ensure that a specific system, process, or facility is generating an output with predefined quality attributes and specifications consistently (3). In summary, valida- tion acts over processes and procedures while qual- ification affects assets. Generalization is possible for qualification, but not for validation. Usually, the validation of a system or equipment implies its previous qualification. Following this reasoning, the validity of AI applications should be studied by A qualified algorithm presumes a ready-to-use state, making the model validation process easier.