Biotechnology, Life Sciences and Control Engineering – Some Synthesis Issues

1

Biotechnology, Life Sciences and Control Engineering – Some

Synthesis Issues

Aidan O’Dwyer

School of Control Systems and Electrical Engineering, Dublin Institute of Technology, Ireland

[email protected]

Colloquium on Control in Systems Biology, Sheffield, March 2007

2

Structure of Presentation

Part A: An outline survey of the synergistic work between biotechnology, life sciences and control engineering.

Part B: Examples of control engineering concepts applied to biological applications, in two sections.

Section 1 outlines how a model of the pupil reflex of the human eye to light may be estimated from an appropriate number of data points.

Section 2 reports on an innovative undergraduate experiment, developed by the author, which estimates a model of a persons’ eye-brain-hand motor response.

3

Part A: An outline survey of the synergistic work between biotechnology, life sciences and control engineering.

• There is an increasing interest in life science related areas in the electrical and control engineering community.

• One manifestation of this interest is the development of interdisciplinary programmes, such as the B.Sc. in Medical Physics and Bioengineering, offered by the Dublin Institute of Technology from September 2005.

• A significant amount of research work in control engineering has focused on aspects of human physiological system, for example blood glucose regulation, cardiovascular modelling and control, and limb control.

4

• Engineering approaches in the study of biological sciences are thus well explored.

• However, “we are in the midst of revolutionary developments in biological sciences”, with “new systems-level knowledge being urgently required” (Sontag, E.D. (2004). Syst. Biol., 1(1), 9-18). In one example, an influential U.S. “future trends” report has signalled, for example, the importance of the role of control theory in understanding the working of cell networks (in molecular biology).

• In recognition of this, the Institute of Electrical Engineers (IEE) has starting publishing, from June 2004, a new journal entitled IEE Proceedings on Systems Biology, which aims to study intra- and inter-cellular dynamics, using systems and signals oriented approaches.

5

• Interestingly, examples have been reported over thirty years ago e.g. control of enzyme activity (As reported in Fleming, D.G. and Feinberg, B.N. (1976). “CRC Handbook of Engineering in Medicine and Biology”, CRC Press, pages 27-30).

• Recent conferences on the topic have included the IFAC Conference on the Foundations of Systems Biology in Engineering (www.fosbe.org), California, August 2005; to be held again in Germany, September 2007.

• More widely, the control of biotechnological processes was one of the major themes of the 2005 IFAC World Congress in Automatic Control.

Systems biology: other strands of activity

6

National Institute of Bioprocessing Research and Training (Ireland)

• http://www.nibrt.ie• NIBRT is a recently established institute providing training

and research for the Bioprocessing Industry in Ireland. • NIBRT supports fundamental, applied and industrial research

in Cell Biology & Product Expression Systems. Protein Stimulation & Biopharmaceutical Design. Bioreaction Engineering. Downstream Processing. Process & Product Analytics - Automation.

7

Part B: Control engineering concepts

applied to biological applications. Section 1: Model of the pupil reflex of the

human eye to light.

[From: http://www.physics.brocku.ca/www/faculty/sternin/120/images/f26030.jpg]

8

 Images: http://www.physics.brocku.ca/www/faculty/sternin/120/images/f26030.jpg,http://www.opt.pacificu.edu/journal/Articles/citek_dre/CITEKDRE.HTM,http://www.meddean.luc.edu/lumen/MedEd/medicine/pulmonar/phydx/s10a.jpg

 Pupil size in darkness Dazzled

Full vision: pupil contracted

9

The pupil reflex to light may be described as a closed loop control system

Pupil neuromuscular

apparatus

How can a transfer function model be obtained for the pupil neuromuscular apparatus?1. Step testing in open loop:

….. difficult to perform experimentally 2. Frequency domain testing in open loop:Reference: Stark, L. and Sherman, P. (1957). “A servoanalytic study of consensual pupil reflex to light”, Journal of Neurophysiology, 20, 17-26. 

10

• The disturbance inputs were sine wave variations in light intensity onto the retina, at a variety of frequencies.

• The pupil area (also a sine wave variation) was measured continuously by reflecting an infra-red light from the iris to the photocell.

• The magnitude and phase of pupil area relative to light intensity was determined.

• Open loop frequency response data determined:

Pupil neuromuscular

apparatus

f 0.14 0.7 0.9 1.3 1.34 2.1 2.3 2.4 2.7 3.0 3.4

M 0.16 0.15 0.12 0.13 0.11 0.06 0.07 0.05 0.05 0.02 0.02

P -60 -90 -140 -190 -190 -320 -380 -420 -470 -510 -530

11

Results

• Four models can be developed (using graphical methods:

s2.01

e16.0)s(G

s42.0

m

2

s38.0

m)s133.01(

e18.0)s(G

s18.01)s1.01(

e16.0)s(G

s38.0

m

3

s18.0

m)s1.01(

e16.0)s(G

Stark and Sherman, 1957

“Analytical” model developed:

“Gradient” model developed:

s17.01

e16.0G

s39.0

m

s13.01

e13.0G

s38.0

m

12

Model validationBode plot of process and models

Process Graphical method Gradient model ( ) Analytical model (*) Stark and Sherman

13

Other interesting data sets

• Human motor response of patients with Parkinson’s syndrome

14

• Hand control system

Reference: Stark, L. (1968). “Neurological control systems: studies in bioengineering”, Plenum Press, New York.

15

Section 2: Eye-brain-hand motor response

This section reports on the development of an innovative undergraduate experiment which estimates a model of a persons’ eye-brain-hand motor response. In the experiment the person is successively asked to track, with a mouse, ten sine wave signals at different frequencies on a computer screen. Based on the average of the data recorded, the persons eye-brain-hand motor response in the frequency domain is recorded (and may be summarized on a Bode plot).Subsequently, the parameters of a single-input, single-output (SISO) process model may be determined.

16

A typical example of one sine wave input signal, and a person’s tracking attempt, is shown.

17

ProcedureA computer with a data acquisition card and suitable software is used to record the input sine wave (generated in SIMULINK) and the persons’ tracking attempt. Implementation:

18

The peak-to-peak amplitude and phase shift of the output was recorded for each frequency.

Time

acay

ysc

tp

º3602

t

a

aAR p

c

y

Amplitude ratio, Phase,

19

Based on an average of the amplitudes and phase differences recorded, a Bode plot is drawn from the data. The figure shows the Bode plot of all the data gathered.

-4

-2

0

2

4

Magnitu

de (

dB

)

10-1

100

101

-225

-180

-135

-90

-45

0

Phase (

deg) blue = right hand data, AOD, May 99

green = left hand data, AOD, May 99black = right hand data, AOD, Jan 06red = left hand data, AOD, Jan 06cyan = data, students, May 99 onwardsmagenta = data, older group, May 99black line = average of all the data

Bode plot of all non-leading data gathered from eye-brain-hand experiment

Frequency (rad/sec)

20

Model determination• A model may be determined directly from the Bode

plot, or using an analytical technique.• It is clear from the figure that subjects tend to be able

to follow the amplitude of the sine wave accurately. Therefore, on average, the gain, , of the model is approximately 1.0.

• The phase varies with frequency in an approximately linear manner. A time delay (reaction time), , may be fitted to this data. The average eye-brain-hand motor response reaction time is 0.25 seconds.

mK

m

21

Some experimental resultsBode plot of authors’ response (age 36): right hand

22

Model identified: ,

Validation of model and original data:

08.1Km ssec24.0m

Model

Experiment

23

Youth versus Middle Age …

-4

-2

0

2

4

Magnitu

de (

dB

)

10-1

100

101

-225

-180

-135

-90

-45

0

Phase (

deg)

blue = dominant hand, older (and avg)green = dominant hand, younger (and avg.)magenta = non-dominant hand, older (and avg.)red = non-dominant hand, younger (and avg.)

Bode plot - dominant and non-dominant hand data (exc. leading) - older, younger

Frequency (rad/sec)

24

Condition Delay determinedAverage of all data gathered 0.25 s

Average, dominant hand data, students, May 99 - Jan 06 0.14 s

Average, dominant hand data, 22 year old male, May 99 0.16 s

Average, dominant hand data, 36 year old male, May 99 0.20 s

Average, dominant hand data, 42 year old male, Jan 06 0.32 s

Average, non-dominant hand data, students, May 99 - Jan 06 0.20 s

Average, non-dominant hand data, 22 year old male, May 99 0.19 s

Average, non-dominant hand data, 42 year old male, Jan 06 0.41 s

Summary of results obtained

25

Pedagogical Issues - 1

• Since its development in 1999, this experiment has been carried out by students taking a control engineering option in the programmes in electrical/electronic engineering at DIT.

• The experiment does not require strong mathematical foundations.

• The author has found that students are enthusiastic about the experiment and frequently spend over the allocated time on aspects of it.

26

Pedagogical Issues - 2Formal Student Feedback 2005-6Please answer the following questions. To answer each question, please write a number between 1 and 5, with 5 - strongly agree, 4 – agree, 3 – unsure, 2 – disagree 1 – strongly disagree

1. The work was a beneficial learning experience (compared to other exercises)

– Average score: 4.3

2. The work is user-friendly – Average score: 4.2

3. The work complements and enhances my understanding of lecture material

– Average score: 4.5

4. The work is fun and sustained my interest – Average score: 4.5

5. I became more interested in the material because of this work

– Average score: 4.0

6.  There is enough time to perform the work – Average score: 3.0

7.  I would recommend this work to others – Average score: 4.3

27

Pedagogical Issues - 3Student feedback is very positive; the reasons for this, in my opinion, are:

• The experiment provides direct feedback to the user on the PC screen.

• The experiment is not excessively time-consuming; a typical experiment time to gather one set of data, at 10 frequencies, is 10 minutes.

• A competitive edge among (typically, male) students is frequently observed, with a desire to have the shortest reaction time.

• A motivational aspect for some students is the application of the idea in biomedical engineering, possibly in the diagnosis of some motor response disorders.

28

Conclusions• The contribution briefly outlines the synergistic work that is

occurring between biotechnology, life sciences and control engineering.

• Control engineering concepts may be applied readily to biological applications. Two examples of transfer function modeling of such applications are described.

• The second application described has been used by the author in the undergraduate control engineering laboratory for a number of years. The experiment does not require strong mathematical foundations. Student feedback is positive.

• In the wider perspective (in Ireland), the recent establishment of the (Irish) National Institute of Bioprocessing Research and Training (http://www.nibrt.ie/), whose mission includes training and research for the biotechnology industry, is facilitating further synergies.