LightProbe - CNN-Based-Image-Reconstruction

Our Current Wireless Probe

Short Description

Wireless ultrasound probes face to main problems:

• On one hand, the high data-rates that has to be transferred from the probe to the connecting device (i.e. to the Smartphone).
• On the other hand, the thermal dissipation of the electronics within the probe.

In this project, we want to evaluate an idea to use CNN-based imaging reconstruction to solve both problems at the same time.

(The idea is more concrete, but we do not want to post it publicly here :-) )

Tasks

The main goal of the project is to evaluate our idea in detail: Your tasks are

• Simulation of the ultrasound imaging system to evaluate the idea (feasibility & performance)
• Quantify energy and data-rate savings with the idea
• Implement the idea on our system (only in case of a Master Thesis)

Prerequisites

• Signal processing basics
• VLSI 1 / Microcontroller experience (only in case of a Master Theis)

Status: Available

Looking for Interested Students

Supervision: Pascal Hager

Character

40% Ultrasound Simulations (including training dataset generation)

40% Simulations of the Imaging Process

20% Evaluations

+200% Implementation (only for Master Thesis)

Professor

Luca Benini

↑ top