BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Modular Clinical Decision Support Networks—Predicting What You N
 eed with What You Have
DTSTART:20230217T100000
DTEND:20230217T110000
DTSTAMP:20260502T104727Z
UID:f5b29fa8312e2c1a34a87155c1f516cc216b962541fba6ed35eac7e3
CATEGORIES:Conferences - Seminars
DESCRIPTION:Mary-Anne Hartley \nThe EDIC program is happy to invite you to
  a public talk by Mary-Anne Hartley\, a post-doctoral scientist in Machine
  Learning and Optimization Laboratory at EPFL.\nThe aim of the talk is to 
 present her achievements to a broad audience to prepare for hiring intervi
 ews coming up soon. You are warmly welcome to listen to the talk and parti
 cipate in the Q&A session at the end of the presentation.\n\nAbstract\nCli
 nical Decision Support Systems (CDSS) are mobile applications that have th
 e potential to improve and standardize care with probabilistic guidance. H
 owever\, many CDSS deploy static\, generic rule-based logic and require an
  inflexible combination of inputs that results in inconsistent performance
  in volatile or evolving clinical environments.\nParadoxically\, it is the
 se environments that have the most to gain from probabilistic guidance (as
  well as the most to lose when it goes wrong).\nIn this talk\, I will desc
 ribe the data-driven solution we developed to address this issue: MoDN (Mo
 dular Clinical Decision Support Networks).\nMoDN is a novel\, interpretabl
 e-by-design\, modular decision tree network\, comprising a flexible number
  and composition of question-specific encoder modules\, which can be assem
 bled in real-time to build tailored decision networks at the point-of-care
 . In short\, it allows the clinician to combine the information that they 
 have and inspect its ability to predict any information that they need.\nI
  will first explain the motivation for MoDN and how it works\, and then sh
 ow the various advantages of its modular design like: 1) providing interpr
 etable\, continuous predictive feedback to the clinician\, 2) being robust
  to the bias of systematic missingness\, 3) enabling multi-modal modeling 
 (i.e. combining text\, images\, sound\, etc.) and 4) allowing privacy-pres
 erving collaborative learning between imperfectly interoperable data sets.
 \n\nBio\nAnnie is an interdisciplinary biomedical data scientist and clini
 cian with a focus on creating and implementing digital health solutions de
 signed for low-resource and humanitarian settings.\nShe leads the Intellig
 ent Global Health (iGH) research group in the computer science department 
 of the Swiss Institute of Technology (EPFL): www.go.epfl.ch/igh iGH collab
 orates with international NGOs (Doctors Without Borders\, the Internationa
 l Committee of the Red Cross\, etc.) and several research/clinical institu
 tions in Switzerland and Africa to collect large-scale\, locally owned dat
 a/biobanks. The group has a special interest in ethical data management\, 
 privacy-preserving distributed collaborative learning\, and the developmen
 t of clinical decision-support tools that aim to improve care by filling r
 esource gaps with probabilistic guidance.\nAs an MD\, she regularly volunt
 eers her medical skills in her home country of South Africa\, where she al
 so embeds her research and several capacity-building initiatives for clini
 cal support (www.help-tintswalo.org) and bilateral academic exchange.\nAnn
 ie obtained an undergraduate in biomedical sciences at the University of C
 ape Town\, South Africa\, followed by a PhD and MD at the University of La
 usanne in Switzerland\, and several further postgraduate specializations i
 n public health and data science\, including an MSc at the London School o
 f Hygiene and Tropical Medicine (LSHTM).\n\n\n 
LOCATION:BC 410 https://plan.epfl.ch/?room==BC%20410
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR