A multidisciplinary task force of 14 European AI experts has come up with a set of broad guidelines for anyone using the technology to analyze big data in healthcare.
While the group’s “points to consider” specifically focus on rheumatic and musculoskeletal diseases, the recommendations readily translate to other medical specialties as well as to other parts of the world.
The team’s paper was published online June 22 in Annals of the Rheumatic Diseases.
Laure Gossec, MD, PhD, of Sorbonne University and 13 fellow members of the European League Against Rheumatism came up with their list after reviewing the literature and analyzing the current state of big-data usage in their field and in other areas of medicine.
After allocating levels of evidence and strengths of recommendations, and then calculating mean levels of agreement, the taskforce members arrived at three overarching principles and 10 points to consider.
The principles are ethical aspects, big data’s potential to change healthcare delivery and the ultimate goal of working with big data in the first place.
On the latter, the authors comment that their field needs to constantly aim “to be of benefit to people with rheumatic and musculoskeletal disorders.”
Here are the opening statements for three of their 10 more granular points.
1. As the amount of big data increases, the need for data harmonization becomes more apparent, with the possibility for using different data sources through application of global standards.
“It is essential to ensure that existing and future datasets can be used and, in particular, pooled for big data approaches,” the authors write. “To this end, they must be harmonized/aligned to facilitate interoperability of data.”
2. While interdisciplinary collaboration is beneficial and required for all research projects, it is even more important in big data projects where expertise is dispersed among different stakeholders.
“The task force insisted on the importance of collaboration between appropriate stakeholders at the analysis stage, for example, where AI methods require appropriate expertise, and at all phases of a big data project,” the authors write. “Interdisciplinary collaborations should intervene at different times across a project, to enable the most appropriate design to be chosen, while ensuring that data collection and the type of analysis are fit for purpose.”
3. Interdisciplinary training for clinical, biological or imaging researchers, healthcare professionals and computational biologists/data scientists in the field of big data is important and links closely with the need for collaborations in the field of big data.
Machine-learning methods “are becoming ubiquitous and have major implications for scientific discovery; however, healthcare professionals are not perfectly aware of the correct use of these methods, whereas data scientists may lack the clinical knowledge to design studies and interpret the findings,” the authors note. “Given the current relative lack of expertise related to big data in the field of rheumatic and musculoskeletal disorders, and given the rapid changes in this field, certain organizations should set up or facilitate training sessions.”
In their discussion, Gossec et al. point out that their guidelines comprise an informal toolset to help set the parameters of future discussions around AI and big data in healthcare—and could well be applied by other medical disciplines.
“[W]e expect that these points to consider [will] inspire governmental and research organizations, healthcare providers, researchers and patients to increase relevant training of the stakeholders,” they write, adding that the work may also help develop benchmarks and guidelines for reproducible research.
The paper is available in full for free.