The University of Auckland

Project #79: Heart Cell Model Acceleration using FPGAs

Back

Description:

To improve the understanding of heart in health and disease, computational models are used in synergy with experimental techniques. Mathematical and computational models of biological cells and tissues play a significant role in drug discovery, testing of cardiac devices, teaching, and reduce the need for experiments to be performed on animals. With the emergence of increasingly advanced experimental techniques has led to an increase in the complexity of mathematical models (e.g. 140 complex ordinary differential equations for one cardiac cell). Therefore, the major issue is that tissue simulations (few hundreds to thousands of cells) may involve millions of grid points. To overcome this problem and improve computational tractability, simplified cardiac cell model (which maintains the balance between the amount of anatomical detail against the time and data capacity required to perform simulations) has been developed. The focus of this project is the implementation of existing developed simplified cardiac cell model (involving only sinusoids and co-sinusoids) on FPGA.

As mentioned above, this project is aimed at implementing cardiac cell model on FPGA and involves only sines and cosines. There are three different ways to implement sine and cosines on FPGA: 1. Using look up table approach, 2. CORDIC, 3. Using adders. The objectives of the project are therefore:

 

1.     Implementation of cardiac cell model on FPGA using all the above-mentioned techniques.

2.     Identifying the implementation technique that will have the best computational speed v/s resource usage.

 

Outcome:

Prerequisites

None

Specialisations

Categories

Supervisor

Co-supervisor

Team

Allocated (Not available for preferences)

Lab

Lab allocations have not been finalised