Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Next-Generation Subcutaneous Injector Device

Author(s):

M. Naumann
Conference/Journal:

Semester Thesis, HS15 (10462)
Abstract:

Altran AG, in collaboration with a large pharmaceutical company, is developing a nextgeneration injector device, which is electronic and software controlled. The purpose is to provide automatic subcutaneous injection of liquids, for example liquid medication for patients with multiple sclerosis. When the injection button is pressed on the device, a small motor inserts the needle, while a second motor pushes a piston into a cartridge and injects a certain predefined amount. In order to control the movement of these two drives (needle and liquid), a position/velocity cascade controller was implemented and tested in previous projects.

My task was to analyze the behavior of the existing controller through various experiments, improve it and determine its usefulness with respect to a different controller.

In this thesis, the focus was on the liquid drive control. I ran multiple experiments and tuned the PID parameters for the existing cascade controller with a Ziegler-Nichols[1] approach. Using a linear coil magnet (LAL) I created three different force profiles, which can be applied as counter-pressure to simulate either different tissues or external disturbances and synchronized (the application of) these force profiles with the start- and endpoint of the movement of the liquid drive. Furthermore, I implemented a simulation using SIMULINK and created an advanced controller using load observation and load compensation. This new controller proved to be useful for this type of application as it performed well in simulation, with higher loads and added sensor noise, as well as on the real system. Finally my task of creating a dashboard for visualizing data was successful. I developed an interface using R Shiny on which it is possible to upload experimental data that initially gets put into a table and is visualized by two graphs on a second page. It can serve as a good basis for a client-server dashboard to monitor patient adherence and compliance by healthcare professionals in the future.

Supervisors: Saso Jezernik (Altran Zürich), Manfred Morari

Year:

2016
Type of Publication:

(13)Semester/Bachelor Thesis
Supervisor:

M. Morari

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2016:IFA_5374
}
Permanent link