Giuseppe Leone, Oncology Resident at Istituto Nazionale Tumori, Milan and Researcher at AI-ON-Lab, shared a post on LinkedIn:
“Really happy to have taken part in the 1st Annual Meeting ESAC – European Interdisciplinary Society of AI for Cancer Research in Milan, where together with Vittoria Peppoloni we received the Best Poster award in our category.
The clinical rationale behind this work is clear: clinical trial matching in oncology is still difficult, due to limited awareness of active trials, the time required for manual screening, and the complexity of eligibility criteria. For this reason, several AI-based trial matching pipelines have been developed to improve efficiency and support clinicians.
We believe the real challenge is how to make these tools truly reliable in the real world. That is why we use a neurosymbolic approach: to get the best of both worlds, with LLMs for data extraction and symbolic reasoning for deterministic matching. In a context where data quality is still not always optimal, this allows us to make the process more robust, transparent, and clinically useful.
The next step will be to keep working on data quality and infrastructure, because – as also highlighted by ESMO – trial matching alone is not enough if discovery is limited, recruitment information is not updated, and referral pathways remain fragmented.
Big thanks to Vittoria Peppoloni and Federica Corso. A special thanks to Arsela Prelaj and all AI-ON- Lab team for the trust and support throughout this journey.
Proud of this result and excited for what comes next.”
Other articles featuring Giuseppe Leone on OncoDaily.
