Mark Birse

Machine learning (ML), artificial intelligence (AI) and other modern statistical methods are providing new opportunities to operationalize previously untapped and rapidly growing sources of data for patient benefit. Whilst there is a lot of promising research currently being undertaken, the literature as a whole lacks: transparency; clear reporting to facilitate replicability; exploration for potential ethical concerns; and, clear demonstrations of effectiveness. There are many reasons for why… Show more

Machine learning (ML), artificial intelligence (AI) and other modern statistical methods are providing new opportunities to operationalize previously untapped and rapidly growing sources of data for patient benefit. Whilst there is a lot of promising research currently being undertaken, the literature as a whole lacks: transparency; clear reporting to facilitate replicability; exploration for potential ethical concerns; and, clear demonstrations of effectiveness. There are many reasons for why these issues exist, but one of the most important that we provide a preliminary solution for here is the current lack of ML/AI- specific best practice guidance. Although there is no consensus on what best practice looks in this field, we believe that interdisciplinary groups pursuing research and impact projects in the ML/AI for health domain would benefit from answering a series of questions based on the important issues that exist when undertaking work of this nature. Here we present 20 questions that span the entire project life cycle, from inception, data analysis, and model evaluation, to implementation, as a means to facilitate project planning and post-hoc (structured) independent evaluation. By beginning to answer these questions in different settings, we can start to understand what constitutes a good answer, and we expect that the resulting discussion will be central to developing an international consensus framework for transparent, replicable, ethical and effective research in artificial intelligence (AI-TREE) for health.
Comments: Show less

Over the last five years, the MHRA Inspectorate has been developing its ability to educate as much as it regulates,
with the aim of driving a culture of compliance with our stakeholders. When serious non-compliance is identified, it’s a painful experience for all; be that for patients with potential disruption to their treatment, for the regulator through the increased workload to ensure patient safety and continued availability of critical medicines, and for industry as a result of the… Show more

Over the last five years, the MHRA Inspectorate has been developing its ability to educate as much as it regulates,
with the aim of driving a culture of compliance with our stakeholders. When serious non-compliance is identified, it’s a painful experience for all; be that for patients with potential disruption to their treatment, for the regulator through the increased workload to ensure patient safety and continued availability of critical medicines, and for industry as a result of the reputational and financial impacts.
One key aspect of this work has been the provision of greater clarity around the agency’s expectations regarding data integrity. Readers of the MHRA Inspectorate blog will be well aware of our work in this field and this article outlines our approach and provides examples of anonymised findings that have been identified across all parts of the Inspectorate.
The MHRA Inspectorate is made up of four groups which cover our operational activities, strategy and innovation, as well as overseeing how we operate as a risk-based regulator using science to underpin the decisions we make. The Inspectorate employs around 75 inspectors working across five GXP areas: good clinical practice (GCP); good distribution practice (GDP); good laboratory practice (GLP); good manufacturing practice (GMP); and good pharmacovigilance practice (GPvP).
MHRA inspectors are regularly requested by other regulatory agencies to provide specialist data integrity training for their inspectors. Show less

The sterility test is a key microbiological test that is required to be performed to support the release of sterile products. A sterility test failure is a time consuming, stressful event often involving a great deal of extra work for a number of people under severe time pressures. It is essential that companies plan for these events prior to them happening so individuals are aware of their roles and responsibilities.