ABSTRACT
Modeling is an important tool to utilize at the beginning of an infectious disease outbreak, as it allows estimation of parameters - such as the basic reproduction number, R0-that can be used to postulate how the outbreak may continue to spread. However, there exist many challenges that need to be accounted for, such as an unknown first case date, retrospective reporting of 'probable' cases, changing dynamics between case count and death count trends, and the implementation of multiple control efforts and their delayed or diminished effects. Using the near-daily data provided from the recent outbreak of Sudan ebolavirus in Uganda as a case study, we create a model and present a framework aimed at overcoming these aforementioned challenges. The impact of each challenge is examined by comparing model estimates and fits throughout our framework. Indeed, we found that allowing for multiple fatality rates over the course of an outbreak generally resulted in better fitting models. On the other hand, not knowing the start date of an outbreak appeared to have large and non-uniform effects on parameter estimates, particularly at the beginning stages of an outbreak. While models that did not account for the decaying effect of interventions on transmission underestimated R0, all decay models run on the full dataset yielded precise R0 estimates, demonstrating the robustness of R0 as a measure of disease spread when examining data from the entire outbreak.
ABSTRACT
Given the growing use of machine learning (ML) technologies in health care, regulatory bodies face unique challenges in governing their clinical use. Under the regulatory framework of the Food and Drug Administration, approved ML algorithms are practically locked, preventing their adaptation in the ever-changing clinical environment, defeating the unique adaptive trait of ML technology in learning from real-world feedback. At the same time, regulations must enforce a strict level of patient safety to mitigate risk at a systemic level. Given that ML algorithms often support, or at times replace, the role of medical professionals, we have proposed a novel regulatory pathway analogous to the regulation of medical professionals, encompassing the life cycle of an algorithm from inception, development to clinical implementation, and continual clinical adaptation. We then discuss in-depth technical and nontechnical challenges to its implementation and offer potential solutions to unleash the full potential of ML technology in health care while ensuring quality, equity, and safety. References for this article were identified through searches of PubMed with the search terms "Artificial intelligence," "Machine learning," and "regulation" from June 25, 2017, until June 25, 2022. Articles were also identified through searches of the reference list of the articles. Only papers published in English were reviewed. The final reference list was generated based on originality and relevance to the broad scope of this paper.
