We are pleased to announce the publication of a new ChatMED research article, “Multi-Agent Readiness Scoring Methodology in Bioinformatics Domain,” in the MDPI journal Future Internet. The paper was authored by Blagojche Gjorgjioski, Djansel Bukovec, Ivana Vichentijevikj, Ivan Kitanovski, Kostadin Mishev, and Monika Simjanoska Misheva, and was published on 2 August 2026.
The study introduces the Multi-Agent Readiness Score (MARS), a framework for assessing whether bioinformatics Large Language Models are not only scientifically capable, but also technically and operationally prepared to participate in interoperable multi-agent and clinical workflows. MARS evaluates models across governance and accessibility, biological competence, technical maturity, and agentic orchestration, while incorporating requirements related to the EU AI Act, HL7 FHIR, HL7 CDA, and MyHealth@EU.
Applied to 43 genomic LLMs, the framework revealed an important gap between biological performance and real-world deployment readiness: most models remain research prototypes or unsuitable for direct multi-agent integration, highlighting interoperability, structured interfaces, provenance, and engineering maturity as major barriers to scalable biomedical AI.
Read the full open-access article in Future Internet: https://www.mdpi.com/1999-5903/18/8/409?utm_source=chatgpt.com.
