Education Pathways for Biomedical Data Science (R25)

项目概要/摘要 该项目汇集了德雷克塞尔大学和费城儿童医院（CHOP）， 自适应生物医学数据科学教育计划，使研究人员能够学习和使用新兴数据 科学方法我们提出了一个在线计划，以准备研究人员直接在他们的数据驱动的工作。 目前的领域，同时也确定跨学科合作的途径。我们将开发新的课程 途径，利用现有的教育资源，创建桥梁材料，并提供实习“实验室” 为参与者提供与特定领域项目相匹配的活动，以获得实践经验。 生物医学数据科学的教育途径将与CHOP学术培训办公室合作 和推广计划，使不同的学习者观众参与新颖的教学研究。学员将 从许多现有CHOP培训计划的积极参与者中招募，包括一种新的数据科学 教育计划，研究生培训，博士后导师，医生奖学金研究，和临床 研究人员培训。注册后，教育项目经理将参与者聚集到协作 实践社区，他们将在那里接受指导，并为发展途径作出贡献。 我们承认，我们仍在学习生物医学数据科学的最有效干预措施 教育，无论是在我们的特定社区和广泛的医学教育（例如费德勒等人， 2015; Rowhani-Faarid，艾伦，and巴内特，2017）.我们将找到新的证据，证明在知识、技能、 和学习者的态度。 我们将开发和实施生物医学数据科学素养工具 基于新兴的奖学金。我们还将通过混合方法收集学习者反馈，以适应和发展 我们的模块化资源，确保强大的学习成果，并与NIH战略计划保持一致， 数据科学。 对于传统计算机和信息科学之外的研究人员来说，数据科学教学是一个具体的 迈向数据驱动的全民科学素养。我们将确保参与者从本计划中脱颖而出， 通过实践经验巩固计算和算法知识。对于非计算机研究人员， 我们还将为个人提供必要的基础数据流畅性，以便为以下方面做出有意义的贡献 机器学习研究，这将使数据驱动的系统，洞察到决策的转换，决策， 数据驱动的决策管理。最后，对于所有参与者，我们将努力帮助个人 从广泛的背景和身份，确定新的方向，其中发展自己的职业生涯。

Project Summary / Abstract This project brings together Drexel University and the Children’s Hospital of Philadelphia (CHOP) to create an adaptive Biomedical Data Science education program to empower researchers to learn and use emerging data science methods. We propose an in-line program to prepare researchers for data-driven work directly in their current field, while also identifying avenues for interdisciplinary collaboration. We will develop novel curricula pathways, leverage existing educational resources, create bridge materials, and provide practicum “lab” activities matching domain-specific projects to participants for hands-on experience. Educational Pathways in Biomedical Data Science will collaborate with the CHOP Office of Academic Training and Outreach Programs to engage a diverse learner audience in novel pedagogical research. Learners will be recruited from active participants in many existing CHOP training initiatives, including a novel data science education program, graduate student training, postdoc mentorship, physician fellowship research, and clinical research staff training. After enrollment, education program managers will cluster participants into collaborative communities of practice where they will receive mentorship and contribute to development of pathways. We acknowledge that we are still learning the most effective interventions for biomedical data science education, both within our specific communities and broadly within medical education (e.g. Federer et al., 2015; Rowhani-Faarid, Allen, and Barnett, 2017). We will develop new evidence of gaps in knowledge, skills, and attitudes among learners. We will develop and implement biomedical data science literacy instruments based on emerging scholarship. We will also gather learner feedback via mixed methods to adapt and evolve our modular resources, ensure robust learner outcomes, and align deliverables with the NIH Strategic Plan for Data Science. Data Science instruction for researchers outside of traditional computer and information sciences is a concrete step toward data-driven scientific literacy for all. We will ensure that participants emerge from this program with computational and algorithmic literacy solidified through hands-on experience. For non-computing researchers, we also will provide the foundational data fluencies necessary for individuals to contribute meaningfully to machine learning research, which will enable data-driven systems, insight-to-decision transformation, decision- making, and data-driven decision management. Finally, for all participants, we will strive to help individuals from a broad spectrum of backgrounds and identities identify new directions in which to develop their careers.