BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Palaner//palaner Single Event//EN
METHOD:PUBLISH
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:meetup-9f9e648dca@palaner.app
DTSTAMP:20260802T142256Z
DTSTART:20260812T050000Z
DTEND:20260812T063000Z
SUMMARY:Agentic AI in Healthcare [in person]
LOCATION:Microsoft
DESCRIPTION:Saved in Palaner as: saved\n\nPalaner match: 41% · why: lots of
  people going\n\nThis month we are going to have an in person meeting abou
 t AI in Healthcare. BI and Fabric\, and also a review of the most importan
 t updates recently on the Power BI and Microsoft Fabric\, and the AI-relat
 ed topics about it. Join us for this session of learning and networking.\n
 Agenda:\n5:00-5:30 PM Welcome and Networking/Refreshments\n5:30-6:30 PM AI
  in Healthcare\n6:30-7:00 PM Wrap Up\nTopic 1: Agentic AI in Healthcare: T
 he RAG-Powered Diagnostic Assistant\n\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\nP
 arametric LLM memory cannot support clinical deployment: no provenance\, n
 o patient-specific grounding at inference time\, and unacceptable hallucin
 ation rates. Retrieval-Augmented Generation solves this by grounding an ag
 entic system's reasoning loop in dynamically retrieved\, evidence-backed c
 ontext instead of static model weights.\n\nThe architecture: Chunking and 
 embedding heterogeneous clinical data (imaging reports\, discharge summari
 es\, structured EHR fields)\, dense vector indexing with approximate neare
 st-neighbour retrieval\, and context injection into the perceive-reason-ac
 t-learn cycle.\n\nKey trade-offs: Retrieval latency versus context window 
 budget\, chunk size versus semantic coherence\, embedding model selection 
 for domain terminology\, and multi-agent coordination for ranking and rout
 ing evidence across specialist sub-agents.\n\nResults: 94.1% diagnostic ac
 curacy in liver cancer detection\; door-to-balloon time reduced from 211 t
 o 68 minutes in a real cardiac workflow via concurrent retrieval across im
 aging\, history\, and evidence.\n\nOpen problems: Explainability at the re
 trieval-to-reasoning boundary\, bias propagation through retrieved context
 \, and the validation gap between proof-of-concept and clinical-grade reli
 ability.\n\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\n\nAbout the sp
 eaker: Dr. Saranya Selvarangan\nA seasoned academic professional consultan
 t\, and data science professional with over 10 years of experience in teac
 hing\, research\, and consultancy across Information Systems\, Business An
 alytics\, Artificial Intelligence\, and Data Science. Proficient in develo
 ping AI solutions\, conducting advanced data analysis\, and guiding resear
 ch initiatives. I hold a PhD in Information and Communication Engineering 
 and am currently pursuing a Master's in Business Data Science\, applying m
 y knowledge extensively within educational institutions and universities. 
 Skilled in curriculum development\, student mentoring\, and leveraging AI-
 driven insights for problem-solving\, I am committed to enhancing both aca
 demic and industry outcomes through research and education\n\nMeetup Spons
 or:\n\n\\RADACAD \\is a leading Power BI training and consulting firm and 
 a Microsoft Gold Partner in Data Analytics. We provide training to upskill
 \, consultation to enhance\, and expert development support for Data analy
 tics projects using Microsoft Fabric and Power BI.\n\nThis is an in-person
 -only event. Please register to attend here\n\nSpaces are limited\, so sec
 ure your spot today and be part of the conversation shaping the future of 
 technology!\nRSVP Now & Secure Your Spot!\n\nWe look forward to welcoming 
 you to an evening of insightful discussions and networking with the Micros
 oft community!
URL:https://www.meetup.com/meetup-group-lyenjapw/events/315808242/
CATEGORIES:palaner,saved
STATUS:TENTATIVE
END:VEVENT
END:VCALENDAR