Background: Research in children is essential for them to benefit from the outcomes of research but involvement must be weighed against potential harms. In many countries and circumstances, medical research legally requires parental consent until the age of 18 years, with poorly defined recommendations for assent prior to this. However, there is little research exploring how these decisions are made by families and the ethical implications of this.
View Article and Find Full Text PDFOne-dose prophylactic HPV vaccination of pre-adolescents may reduce cervical cancer deaths dramatically in lower-resource settings, but the benefits of achieving immediate high coverage among pre-adolescents would not be realized for 20 to 40 years. Prophylactic vaccine efficacy is reduced after sexual debut, and current therapeutic intervention candidates designed to treat existing HPV infections or precancerous lesions have yielded insufficient evidence to warrant widespread use. However, we are developing a feasible, scalable, high-quality cervical screening approach that could prevent hundreds of thousands of deaths, while we work to achieve high coverage of one-dose vaccination for adolescent cohorts.
View Article and Find Full Text PDFA number of challenges hinder artificial intelligence (AI) models from effective clinical translation. Foremost among these challenges is the lack of generalizability, which is defined as the ability of a model to perform well on datasets that have different characteristics from the training data. We recently investigated the development of an AI pipeline on digital images of the cervix, utilizing a multi-heterogeneous dataset of 9,462 women (17,013 images) and a multi-stage model selection and optimization approach, to generate a diagnostic classifier able to classify images of the cervix into "normal", "indeterminate" and "precancer/cancer" (denoted as "precancer+") categories.
View Article and Find Full Text PDF