Co-authorship network analysis in cardiovascular research utilizing machine learning (2009-2019).

Area of Science:

• Cardiovascular Medicine
• Computational Science
• Network Science

Background:

• Machine learning (ML) is increasingly used in medical research, necessitating interdisciplinary collaboration between clinicians and data scientists.
• No prior analysis has mapped research collaboration networks in cardiovascular medicine utilizing machine intelligence.
• Understanding these networks is crucial for fostering effective scientific partnerships.

Purpose of the Study:

• To analyze the co-authorship network structure in machine learning applications within cardiovascular medicine.
• To identify leading researchers and understand collaboration patterns.
• To provide insights for future scientific collaboration planning in this interdisciplinary field.

Main Methods:

• Co-authorship network analysis of 2857 articles (2009-2019) from Web of Science.
• Network metrics calculated: density, average degree, clustering coefficient, number of communities.
• Leading authors identified using centrality metrics; network structure validated with Barabasi-Albert model simulation.

Main Results:

• The network comprised 13,979 nodes and 68,668 edges, showing linear growth over time.
• Network exhibits scale-free properties, with D. Berman and S. Neubauer identified as highly central authors.
• Of the top-ranked authors, 63.6% held medical degrees (MD), while 36.4% had PhDs in computational science.

Conclusions:

• Co-authorship network analysis reveals the structure of collaboration in machine learning for cardiovascular disease.
• The findings highlight the interdisciplinary nature of the field, with both medical and computational experts playing key roles.
• This analysis offers valuable insights for strategic planning of future scientific collaborations in cardiovascular research leveraging machine intelligence.