Publications by authors named "Jeff Leek"
Article Synopsis
- Generative AI can have issues like bias, data privacy concerns, misinterpretation of prompts, and generating inaccurate information, which are known as hallucinations.
- A scoping review analyzing 120 articles on generative AI in medicine identifies key challenges and suggests solutions for digital health practitioners to navigate these pitfalls.
- The review highlights frequent issues such as bias, privacy, and compliance, while also noting less discussed topics like overreliance on AI and adversarial attacks.
Article Synopsis
- - Health equity means everyone has equal chances to achieve their best health, but human genomics research has not reflected the diversity of the population, leading to health disparities.
- - The National Human Genome Research Institute (NHGRI) acknowledges these inequities and has gathered experts to provide recommendations and review the current state of health equity in genomics.
- - This report outlines the gaps and opportunities in bridging human genomics with health equity, emphasizing the need for more diverse participation in genomics research to improve health outcomes for all.
Previously, we have described our feeder-free, xeno-free approach to generate megakaryocytes (MKs) in culture from human induced pluripotent stem cells (iPSCs). Here, we focus specifically on the integrity of these MKs using: (1) genotype discordance between parent cell DNA to iPSC cell DNA and onward to the differentiated MK DNA; (2) genomic structural integrity using copy number variation (CNV); and (3) transcriptomic signatures of the derived MK lines compared to the iPSC lines. We detected a very low rate of genotype discordance; estimates were 0.
View Article and Find Full Text PDFScatterplots are the most common way for statisticians, scientists, and the public to visually detect relationships between measured variables. At the same time, and despite widely publicized controversy, P-values remain the most commonly used measure to statistically justify relationships identified between variables. Here we measure the ability to detect statistically significant relationships from scatterplots in a randomized trial of 2,039 students in a statistics massive open online course (MOOC).
View Article and Find Full Text PDF