BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:From Clinic to Computation: Understanding Emergent Dynamics in AF 
 Using Cardiac Models\, AI and UQ - Dhani Dharmaprani (University of Adelai
 de)
DTSTART:20240603T135000Z
DTEND:20240603T141000Z
UID:TALK214513@talks.cam.ac.uk
DESCRIPTION:Introduction: Cardiac digital twins present a multiscale physi
 cs and physiology constrained framework for studying atrial fibrillation (
 AF). RENEWAL-AF identified novel indices of AF tissue level dynamics that 
 correlate with AF recurrence\, termination\, and persistence. However\, it
  is unclear how tissue properties may impact these measures\, which is typ
 ically also computationally intensive to explore.\nMethods: This pilot stu
 dy addresses this using Gaussian Process Emulators (GPEs) to perform rapid
  parameter analysis. Simulations of AF (4x4cm 2D grids) using the Courtema
 nche and Luo-Rudy models were analysed (n=200 each)\, focusing on ionic co
 nductance (Courtemance: GNa\, GK1\, GCaL\; Luo-Rudy: GNa\, GK1\, GK1bar\, 
 GSi) and tissue conductivities (longitudinal\, transverse). Parameters wer
 e sampled using a Latin hypercube design. Using virtual catheters simulati
 ng the HD-grid catheter\, we measured: i) correlation length (&xi\;)\; ii)
  rate of spiral wave formation (&lambda\;f)\; iii) rate of spiral destruct
 ion (&lambda\;d) - indicators of atrial electrical desynchrony. Simulated 
 &xi\;\, &lambda\;f and &lambda\;d values were used to train GPEs implement
 ed through GPErks. GPE performance was assessed using R2 and MSE\,and used
  for sensitivity analysis to quantify the importance of each model paramet
 er.\nResults: No significant difference was found between mean simulated v
 alues calculated in-silico using the virtual catheter (&xi\;=27.15 (95%CI:
 22.69\,31.61)\, &lambda\;f=6.28 (95%CI:6.10\,6.42))\,&nbsp\; &lambda\;f=3.
 81 (95%CI:3.38\,4.24)\,&nbsp\; and mean clinical HD-grid measurements (&xi
 \;=34.63 (95%CI:27.78\,41.49)\, &lambda\;f =6.54 (95%CI:5.41\,8.61))\, &la
 mbda\;d=3.67 (95%CI:2.61\,4.72)\, (all P>0.05). For the Courtemanche model
 \, GPEs returned an R2=0.62 for &xi\;\, compared to 0.66 for &lambda\;f an
 d 0.64 for &lambda\;d. For the Luo-Rudy model\, &xi\; R2=0.78\, &lambda\;f
  R2 = 0.69\, and &lambda\;d R2 =0.67. Sobol variance identified longitudin
 al tissue conductivity and GSi as most influential on &xi\;\, which was al
 so observed in addition to GK1 for &lambda\;f and &lambda\;d.\nConclusion:
  Despite scope to optimise and improve GPE performance further\, this pilo
 t demonstrates GPEs' potential to efficiently map bespoke tissue scale AF 
 metrics to model parameters\, potentially further supporting cardiac digit
 al twins for clinical/research use.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR