BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:From personalised medicine to decision-making for emergency care: 
 Role of smart devices and analytics - Dr Siddharth Arora (University of Ox
 ford)
DTSTART:20191017T103000Z
DTEND:20191017T113000Z
UID:TALK132628@talks.cam.ac.uk
CONTACT:Shuya Zhong
DESCRIPTION:The aim of this talk would be to present the following two pro
 jects:\n\n(i)  Using smartphones for detecting and monitoring the symptoms
  of Parkinson’s disease (PD) – we demonstrate that using inbuilt smart
 phone sensors\, we can assess key motor symptoms associated with PD such a
 s: voice\, balance\, gait\, finger tapping\, reaction time\, rest tremor\,
  and postural tremor. Using a range of features extracted from these smart
 phone recordings\, we were able to: (a) distinguish PD participants from c
 ontrols with a high accuracy\, (b) monitor symptom severity remotely\, and
  (c) identify participants who are at risk of developing Parkinson’s. Th
 ese findings are a step towards developing a remote diagnostic support too
 l for PD.\n\n(ii) Predicting patient waiting times in the emergency depart
 ment (ED) - Accurate predictions of individual patient waiting times in th
 e ED can help improve overall patient satisfaction and assist healthcare o
 rganisations to streamline patient-flow based on informed staff and resour
 ce allocation. Waiting time is inherently uncertain\, and so a point forec
 ast is unhelpful and potentially misleading\, and can thus result in great
 er dissatisfaction among patients. Using anonymized patient-level ED data\
 , in this ongoing study\, we adopt a machine learning approach to: (a) pre
 dict patient waiting times for both major and minor triage categories\, an
 d (b) identify the key variables that have the highest impact on modelling
  accuracy. These predictions can assist patients select a hospital in thei
 r vicinity that has the shortest waiting time.
LOCATION:Institute for Manufacturing\, University of Cambridge
END:VEVENT
END:VCALENDAR