R - imeko

Transcription

R - imeko
1
XVIII IMEKO WORLD CONGRESS
Rio de Janeiro, RJ, BRAZIL
May, September
7 -12, 2006
16-22
2006
Universality of Measurement
in Sciences
in Medical
Sciences
Prof. Attila NASZLADY M.D.,D.Sc.
D.Sc
General Director of Policlinic of Hospitaller Bros
Past President of the European Federation for Medical Informatics
Academician in Pontificia Accademia Tiberina, Vatican
Knight in Order of St.Sepulcre of Jerusalem
1
2
UNIVERSAL RELATIONSHIPS
„Mens
i
force
agitat
*
voltage
affinity
molem”
∆e
route =
*
charge =
*
mass =
This presentation contains three main points:
t
1.What are the measurable variables in medicine.
W
O
t
=
R
P
O
2.Which are the possibilities for transporting the measured
W
E
R
data from one provider to the others.
K
pressure
Σ
ικ
the quantitative as well as qualitative relationships among
* volume =
∗
δεκ = ∆e
3.How can we process the data to construct information from
them.
Naszlady 1998
Fig.1
The different information carriers are texts as the patient’ s
I.
In fo rm a tio n C a rrie rs in m e d ic in e
History. These textual informations can be processed by
matematical-statistical methodologies and exress the state of
- T e x ts
(a lfa -n u m . d a ta m a t.-s ta t. a n a ly s is , p ro b a b ility ..)
- C u rv e s
the patient at the time of investigation. The curves are usually
amplitude/time functions and after having processed express
( s p e c tra l a n a ly s is , F F T , a u to c o rre la tio n … )
the adaptivity of the patient. The pictures the structural
- P ic tu re s
conditions, and allow us to forcast the outcome of the disease.
( d ig ita liz a tio n , 3 -D re c o n s tru c tio n s … )
Fig 2.
2
3
DATA AQUISITION
IN MEDICINE
•
direct:
•
inductíve: A
Pharmacology
S
A
L
•
convers:
S
•
invers:
L
S
System identification A
have been identified and measured, then the fourth
unidentified can also be determined.
S
A
other sciences. If three of the four other components
R
L
Physiology
Clinics
The data aquisition in medicine is the same as in any
L
R
The unknown variable is indicated by red colour on
R
the Fig. No.3.
R
A = Action; S = System; L= Law; R = Reaction
Naszlady 1998
Fig.3.
S te r e o - v e c to r c a r d io g r a m
ey
The traditional ECG takes only 12 voltage samples
n ose
from the body surface. From the Frank – spacial
e
X
Y
ey
orthogonal leads we succeeded to reconstruct the
e
Z
whole electrical field around the heart. The small
bars represent time in msec, their high density
means conduction disturbances. Red colour shows
the left and the blue one the right side of the heart.
Normally the borderline circle is convex to outside.
Fig 4.
3
4
D o sa g e o p tim iz a tio n o f D ru g s B a se d o n P h a rm a c o k in e tic s C o m p u ta tio n
C onc.
to x ic
The dosage calculation is frequently very complicated,
if we want to administer drugs individually. That is the
reason for ordering them as e.g.: three times a day
irrespectively the individal parameters of the patient. By
e ffe c tiv e
1.
T 1 /2 =
Fig.5.
mmH g
100
4.
3 ,0
5.
4 ,5
ml / s
10
8.
9.
t (d a y)
6 ,0
1 0 T 1 /2
computer modelling of the pharmacodynamic process
of the drug within the body so we can determine
the individually optimum dosage of the medicaments.
Coronary blood flow is measured regularly in ml/min.
P -LV
As one can observe on the Figure 6. at the same blood
C SV
4 ml
2
333
100
1 ,5
3.
T h e tim e o rg a n iza tio n o f th e co ro na ry circu la tio n
50
mmH g
2.
2 d a ys
6.
7.
ml / s
10
C o ro na ry b loo d f low
m se c
flow per minute, differs from blood flow per beat (CSV).
666
60/m in *
4
ml / b ea t = 2 4 0 m l / m i n
On the computer model there have not been changed
80/m in *
3
ml / b ea t = 2 4 0 m l / m i n
neither pressure nor vascular conductance (cross
sectional diameter). The alteration of blood supply in the
C SV
3 ml
50
2
300
450
C o ro na ry b loo d f low
heart is the consequence of time organization only.
P -LV
m se c
Fig. 6.
4
5
On the routin chest X-ray picture the main airways are
Not well visible. By postprocessing the same picture
as it can be seen on the right side image, the main
respiratory tract ( the Trachea) can be seen more
clearly. The 8 bits/ pixel image representation proved
to be enough for this medically important distinction.
Inversally it is possible to set the gray scale so that
mainly the bones ( ribs ) appear more intensively.
Fig.7.
The 8 bits/pixel resolution image has a 256 levels
gray – scale, the distinction of that by human eye is
almost impossible. Transforming, however, to
pseudocolour representation of the different darkness
of gray levels helps in disdinguishing them. So the air
filled bubbles, (blebs) not visible on the original
picture in the lungs became identificable even for
medical students as well.
P s e u d o -c o lo r re p re s e n ta tio n o f c h e s t X -ra y p ic tu re
Fig 8.
5
6
By means of ICT application the two-dimensional (2-D)
3 -D re c o n s tr u c tio n o f th e H e a rt
ultrasound images (e.g. echocardiograms) can easily be
reconstructed to 3-D images, as one can see on the Fig.9.
The red colour indicates left ventricle, the blue colour the
right ventricle of a patien’s heart. At the apex of the left
ventricle myocardial postinfarcted sac (aneurysma) can be
quantitatively measured in ml. So surgical intervention
fr o m u ltra s o u n d p ic tu r e s
became possible and has been performed.
Fig. 9.
The measured data transport is unavoidable when the
II.
D a ta tr a n s p o r te r s in m e d ic in e
specialisation has become extremely graded and therefore
the patients have to have shared care. ICT offers for that
purpose three main modes: Electronic Patient Record,
- - E le c tr o n ic P a tie n t R e c o r d
- - In tr a n e t tr a n s fe r o f d a ta a n d in fo r m a tio n s
Intranet communication within the health institution, and
Internet data and information transfer among remote
- - In te r n e t c o m m u n ic a tio n o f p ts ’s in fo - s
health providers. Privacy protection, however, is a „must”.
Fig 10.
6
7
The extended mobility of population requires so called
shared care. It means that all the medical information
and insurance eligibility are accessible for the health
provider. A chip-laser hybrid smart card is the best
solution for this purpose. Safe, secure, confidential.
We have developed such an Electronic Health Care
Record (EHCR) in five EU languages with automatic
translation from any to any other one.
N a s z la d y A . n a s z la d y J
Fig.11.
C o m ple te Ele ctron ic H ealth C are R eco rd
o n c hip c a rd
E.Sz .E.M .:E ü .S zem ély es E lek tro n ik u s M em or ia
Em erge ncy
P h a rm a c y
P ic tu re s
This Chip-doki system contains the following sections
with different authorized accessibility: Identification,
Emergency, Patient’s history, and Status, Laboratory,
P o rtra it
Pharmacy, Pictures, Jobs, and Expenditure sections.
Being this smart card EHCR always at the patient,
his/her reliable data and informations are available
for authenticated providers at any place and any time.
( A palm-top W/R device is enough to use it)
N as z la dy 1 98 8
Fig. 12.
7
8
The storing capacity of this hybrid card is big enough
to archive the patient’s all textual, graphical (curves),
and picture type informations. The card is quatitatively
unlimited database because if needed a second ehcr
card may be created for the same patient. On Fig 13.
Chest X-ray, ECG, SPECT, 3-D echo, Stereo- VCG,
MRI, and coronarographic images are shown
N a sz la d y 1 9 9 8
allowing to compare these finding to each others.
Fig. 13.
H e a lth p r o v id e r
discharge for the patient by the Insurance Company;
co
.D
de
at
a
a
I.D
rs
pe
P
s
er
on
a
de
co
ta
P a t ie n t
D e p e r s o n a liz e d
via web to a central electronic archive for retrieval
ta
at
a
aa
.d
at
ed
d
e
+
z
i
l
od
da
I.D
.d
ed
ed
li z
c
liz
na
I .D
doctor or doctors. Depersonalized data could be sent
so
na
the next step is filling up with data by the patient’s
er
so
The work-flow is shown on Fig.14. starting with card
ep
er
at
D
P
co
+d
de
De
I.D
In su ran ce C o
A r c h iv e
d a ta
either for care or scientific purposes. This system can
R e s e a r c h C tr .
Fig. 14.
8
satisfy medical, legal, privacy protective requirements.
9
After having been data enough in number and quality
III.P ro c e s s in g o f d a ta
health authorities may process them with statisticalmathematical methods, e.g. population mass screening;
another important application of ICT in medical
---fro m p o p u la tio n m a s s s c re e n in g
sciences is the computerized stuctural and/or functional
model creation. In such a model there are no physical
---fo r m o d e llin g o f c lin ic a l p h y s io lo g y
or ethical limitation to perform experiments which
cannot or may not be carried out on human beings.
Fig. 15.
A part of results in a population health condition study
which aimed at assessing complaints, signs and
symptoms and was carried out by smart card in1995 are
shown on Fig.16. The different coloured bars represent
the frequency distribution of Cardiovascular(CV),
Respiratory (R), Gastro-Intestinal (G-I), SkeletoMuscular (Sk-M), and Neuro-Psychological ( N-Psy)
N as
N zaslazla
dy dy
19 19
96 98
Fig. 16.
9
Complaints and number of participants is shown by GIS.
10
At the bifurcation (branching) of the carotid artery a
F u n c tio n a l A n a lys is o f V a s c u la r-r e c ep to rs
biological stretch receptor takes place in the neck.,
Fc
F0
FR
C
F0
FR
Fc
R
simulating the biological structure could be described
by loop-equation sec. Kirchoff. Rearranging the
Τsy
P
A r t.p u lm .
dP
C
dt
1
C
Fc dt
for regulating blood pressure. The electrical model
(K irc h o ff- e q u .)
integral equation and completed by measured
parameters it becomes clear for an expert in biology
FC
N as zlad y 1 9 9 8
that it is not only pressure but flow receptor too.
Fig.17.
A complete cario-pulmonary computer model were
constructed with paper-and-pencil method, discribed
C a r d io C A R D IO COM PUTER
by Kirchoff equations. The heart was a flow-source
PULM O NARY
M ODEL
(current-generator) , the other simulations were
realized by parallel capacitive (vascular compliance),
by ohmlike resistive (vascular resistance) elements.
Quatitative reliability had been proved by comparing
N a s z la dNy a1s9z 9la8d y 1 9 9 8
Fig. 18.
10
the virtual results to the real ones.
11
A n a lo g u e C o m p u te r P ro g ra m fo r B lo o d P C O 2 C a lc u la tio n
The cardio-pulmonary analogue computer model ’s
Rk
-1
Program structure can be seen on Fig. 19. The symboles
-1
11
2 τ
nT
1
1
1 -R k
1
1
of the operational amplifiers are internationally accepted
1
2π
nT
X
1
nR 1 C 1
P aC O 2
and used. These are integrators, summators, multipliers
1
nR m i n C 1
Y
potentiometers, completely set with adequate, human,
11
1
1
FM
nC 1 P M
(-P 3 ) o
real parameters measured previously in clinical routin
Pv CO2
PM
R 2F M
1
1
1
1
1
-f o
investigation.
+P5
FM
nC 2 P M
(P 5 ) o
PM
FMR2
Fig. 19.
The periodical upper curve represents alveolar wave-
K a r d io - p u lm o n a lis C o m pu t e r M o d e l
like ventillation of the lung ( V’A), the lower two pCO2
V ’A ( c m 3 s -1 )
105
time-function curves show the partial pressure of the
70
p C O 2(H g m m )
0 ,3 5 L
5 s ec
0
V C O 2 =0 ,2 4 L /m in
80
60
40
20
v
a
Q’ is the cardiac output. When it is reduced from 6 to 3
Liter/ minute, the checked arterial pCO2 decreased
FR C = 2 ,4 L
10
Q ’ = 6 ,0 L /m in
carbon dioxide in the arterial (a) and venous (v) blood.
35
V A = 4 ,2 L / m in
20
Q ’ =3 , 0L /m i n
30
indicating „improvement”, but the venous pCO2 level is
( se c )
elevating and this means „secret” danger for the patient.
N as zl a dy 199 8
Fig. 20.
11
12
Input impedance( Zin) of the great arteries are
In p u t im p e d a n c e o f th e g re a t a rte rie s
frequency-dependent and the values of Zin are
S y ste m ic c ir c u la tio n .
P u lm o n a r y c ir c u la tio n .
Z in
500
100
500
1000
Z in
100
r
F R = 1 /m in
200
r
explanation for tachycardia when the heart is in
F R = 1 /m in
trouble. From the formula it can be observe, that
300
Z in
100
F R = 5 0 /m in
200
F R = 1 0 /m in
ω (omega) representing heart rate is inversally related
-j
-j
Z in = R
to Zin of both great arteries (aorta and pulmonary).
1
1+ jω R C
N a sz la d y 1 9 9 6
Fig.21.
R egu larity in H eart Ra te of M am m a ls
f1
G2
f2
G1
resulted in discovering a rule of REGULARITY IN
*
G
1
-1/3
k
k 1 )
L
f1
f1 L
E. g. :27 kg : H R= 10 0/m i n;
Dog or child
The higher the heart rate, the less the Zin .
The computer modelling of cardiovascular system
3
f 2 G 2 1 /3
f1
decreasing with increasing heart rate. That is a good
(G
1
-1 /3
1
HEART RATE OF MAMMALS. That expresses an
)
invers relationship between heart rate (f) and body size.
L
The heart rate of two mammals are inversally related to
k = 3 00
the cubic root of their body mass (some length value L)
k
6 4k g : HR= 7 5/m in;
ma n
125 kg : H R= 60 /m i n
pig or m an
N a szl a d y 19 69
Mathematical desciption revealed that frequency times
length of aorta (up to main branching)is a constant (k).
(cf. Phenomenon of resonancy in physics)
Fig.22.
12
13
T H E P R O C E S S O F P A T IE N T C A R E
re quir e d
ord ers
b os s
t est
su bo rdin at es
The work flow of patient care may be descibed as a
„ is”
d ist urb
th e sta te of iss ue
feed –back operation as one can see on Fig 23.
Usually physision could not be able to achieve the best
R es u lts o f me as u rem e nts
Ne w d a ta
treatment at the first attempt, so each patient’s care
Fe e dba ck
M ea s ura b le d is turb
may be regarded as an investigation for the optimum.
On a parallel computer model the doctor may be able
L iv in g s ys te m
to perform „experiments” to find the quatitatively
int erv e ntio ns
mo de l
c on tro lle d
res u lt s
optimal treatment to be applied to the patient.
F ee dfor w a rd
E. R. Car son 1 98 5
Fig.23.
Summary: Medical informatics started in Hungary at the early 60-ies of the XX-th century. Data collection was not allowed in our socialist
regime that time, so we begun the application of ICT with analogue computer modelling. My first result was the recognition of analogy
between the stroke of lightening in an electrical long distance transmission line and the heart beat in the arterial system, in 1964. The same
phenomena could be observed in both structure: pressure-voltage wave propagation, wave reflections, lateral leakage of flow, etc. Mathematical
computer models have been constructed with branching lumped quadrupoles and their loops were described by Kirchoff-equations, in 1966. We
discovered the Regularity in heart rate of mammals based on resonance theory, in 1969. Coronary blood circulation, cardio-pulmonary
pressure-volume-flow interactions have been successfully studied this way as a virtual reality and produced a lot of new recognitions, in the
1970-ies. Pharmacokinetic computer models helped us to calculate individual dosage of various medicaments, in 1990. This should be
introduced in patient’s care for optimisation of treatment in the future. Another very useful application of ICT would be in data processing of
population mass screening as well as in shared care. We have performed a GIS supported pilot study in this field, in 1995. We also had some
results in 3-D reconstruction and volume calculation of irregular-shaped heart ventricles. This would be usable in cardiac surgery especially
combined with telekinetic manipulations. Worldwide initiations are observable in the field of making applicable the chip-card technology for
Electronic Health Care Record containing complete patient’s documentation:texts, curves, pictures. We also worked out one smart card, in 5
languages for cross-border application too.
13