BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Network modelling and Graph Neural Networks for emergency healthca
 re management - Annamaria Defilippo\, PhD Student - Magna Graecia Universi
 ty of Catanzaro
DTSTART:20250224T180000Z
DTEND:20250224T183000Z
UID:TALK228856@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:Patients needing emergency department (ED) services are sorted
  into urgency categories using triage\, often through severity indexes lik
 e the Emergency Severity Index (ESI)\, traditionally done manually by nurs
 es. This manual triage process\, while effective\, can be time-consuming a
 nd prone to human error due to the subjective nature of the assessment.\nI
 n this talk\, it will be introduced a network-based patient modelling appr
 oach using Graph Neural Networks (GNNs) to automate triage by leveraging i
 nter-patient similarities and inter-feature relationships. This approach a
 ims to streamline the triage process\, enhancing both accuracy and efficie
 ncy.\nThe proposal presented in this session considers two models: one tha
 t views patients as nodes in a similarity graph (Patient-Level Modelling)\
 , and another that forms a graph for each patient with nodes as features t
 hat connect based on mutual information (Feature-Level Modelling).\nThe pr
 eliminary findings from applying these models confirm the effectiveness of
  these methods. The automated triage system shows promise in accurately ca
 tegorizing patients according to urgency levels\, thereby potentially redu
 cing the workload on medical staff and minimizing the chances of human err
 or.\nMoreover\, there are exciting future possibilities for enhancing tran
 sparency and clinical applicability through explainability techniques and 
 the integration of the two models.\n\n\nGoogle Meet joining info: https://
 meet.google.com/tmt-xpeg-xhm
LOCATION:Lecture Theatre 2\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR