The Ohio State University College of Medicine Office of Curriculum and Scholarship recently held an educational forum that explored the growing movement of open science; open science is a set of principles and practices aimed at making scientific research and its outputs accessible, transparent and reusable for the benefit of both scientists and society. Presentations during this forum by a group of experts gathered from across multiple disciplines at Ohio State shared how open science principles can strengthen research quality, impact and innovation.
Matt Southward, PhD, assistant professor of clinical psychology and director of the Mood, Anxiety & Transdiagnostic Treatment Laboratory at the university, began the forum with a presentation on key factors of open science. These included:
- Transparency on how data is collected and analyzed. This can include explanations on deviations in data collection methods and justifications for your approach.
- Reproducibility. This concept includes open code and materials that other researchers can access through repositories to use and examine and even offer evaluations and checks and balances on the work.
- Statistical power that shows what the research is powering for, including sample and effect size and dimensional measures.
Xia Ning, PhD, MBA, professor and division chief of Artificial Intelligence in Digital Health in the Department of Biomedical Informatics at the College of Medicine, expanded the term “open science” to “open-minded science” – which included:
- The need to consider alternative viewpoints, challenge existing beliefs and adapt one’s understanding based on evidence.
- She uses her computer science and engineering experience to extract knowledge from data using artificial intelligence (AI) and large language models (LLM) and shows how different AI software and different LLMs can change research results.
- The different levels of data use restrictions and security within research, which includes public access, controlled access and governed access.
- The acronym FAIR, which refers to data that is findable, accessible, interoperable and reusable.
Lang Li, PhD, professor and chair of the Department of Biomedical Informatics in the College of Medicine, presented on work using electronic medical record data sources and medical informatics approaches to investigate risk factors (e.g., genomics and drug interactions) that predict either drug efficacy or adverse drug events. He stressed that:
- AI and humans co-learn together.
- Humans can’t do the work alone, but can train machines and humans to synthesize knowledge, together, to produce outcomes they couldn’t produce on their own.
Then, Kelsey Badger, MS, assistant professor and research data librarian at The Ohio State University Libraries, shared information on open data and open access. She shared the university’s open science resources, stressing that data-sharing requirements have changed, and also that there are now updated federal access policies that include information being held in data repositories.
