1 river flow model of diseases riichiro mizoguchi, kouji kozaki, hiroko kou, yuki yamagata isir,...

1

River Flow Model of Diseases

Riichiro Mizoguchi, Kouji Kozaki, Hiroko Kou, Yuki Yamagata

ISIR, Osaka University

Takeshi Imai, Kayo Waki, Kazuhiko OheDepartment of Medical Informatics, Graduate School of Medicine

The University of Tokyo

2

Agenda of my talk

Our goal Ontological discussion on a new definition of diseases

River flow model of diseases Ontology of causal chains Cumulative continuous processes

A new computational model of diseases Core causal chain Clinical imbalance

3

OGMS: Ontology for General Medical Science

supervised by Barry Smith It includes fundamental clinical terms such as

Disorder, Pathological process, Disease, Etiological process, Disease course, etc.

Pros: Sound and very carefully defined Cons:

(a) Disease is defined as a disposition which is a potentiality for the realization of chains of physical/physiological changes in the human body.(b) Although they mention “elevated level of glucose in the blood” to talk about diabetes, there is no type it instantiates in OGMS.

4

Our goal

To augment OGMS by concretizing the definition of diseases in it

To do that, we propose:1. an clinician-friendly definition of diseases as causal

chains of clinical disorders 2. a new type for talking about so-called

“major pathological condition”3. an idea of imbalance as a concretization of disposition4. a computational model of diseases to make our

proposals operational

5

Some definitions1. A enacts B =def A is a continuant and B is an external process of A who/which partic

ipates in it as a whole in which it is maximal among participants who/which play the same role in the process. Examples: when you walk, you (not your legs) enact your walking, the motion of your legs is internal process of your walking which you cannot enact.

2. Event =def a non-dissective unitary entity in the temporal space. Examples include a conference, an arrival, etc. It must be dealt with as a whole in any case.

3. Process =def a dissective non-unitary entity in the temporal space like walking, singing, etc.

An event is constituted of processes, except an instant event.

4. External process of A =def a process enacted by a continuant A.5. Internal process of A =def a process enacted by a part of A. Examples: In a walkin

g process of A, leg motion is an internal process of A whose external process is the walking.

6. Object =def anything which enacts a process [2][2] Antony Galton and Riichiro Mizoguchi: The water falls but the waterfall does not fall: New perspectives on

objects, processes and events, Applied Ontology 4(2), pp. 71-107 (2009)

6

Definition 1: Disease

A disease is a dependent continuant constituted of one or more causal chains of clinical disorders appearing in a human body and initiated by at least one disorder.

In OGMS: disposition (i) to undergo pathological processes that (ii) exist in an organism because of one or more disorders in that organism (excerpt from OGMS paper)

The main issue here is How is a causal chain of clinical disorders a

continuant?

7

Informal account of our river flow model of diseases

A river is similar to a disease as a causal chain

After it has been born as a river (as a disease), then it extends further (causes some disorders) to another lake or to the sea. While extending, it branches (the branching perhaps causes the appearance of another disorder or symptom).

Finally, it may dry up because of climate change (cure). Thus, the life of a river corresponds well to the life of a disease. Thus

– in concordance with OGMS – both a river and a disease are continuants

Note here that a river is not a disposition

8

What is a causal chain

Causal chain =def a chain of entities linked by causality. There can be a causal chain of disorders, causal chain of processes, causal chain of events, etc.[1]

[1] What constitutes causality is outside the scope of our research.

Two kinds of processes:

1. Cumulative continuous process: a process that proceeds without completing the current process/event at every instant in time

⇒ causal chain of processes2. Non-cumulative process: a process that proceeds by completing the

current process at every instant in time ⇒ causal chain of events

9

Cumulative continuous process

Examples: a fall of water and a river flow We know waterfall and river are a continuant associated with these two

processes. Following the definition of object in , what we need is to identify processes which each of these two enacts.

“The water falls, but the waterfall doesn’t fall”. That is, what a waterfall is doing is not the water falling but migration upstream as it carves its way into the rock

What a river enacts is not the water flow but change of the shape of its course, and the water flow of the river is its internal process

Similarly, a causal chain as a flow of causality (propagation of causality) is an internal process of the causal chain which is a continuant that enacts branching, extension, and diminishing processes as its external processes

Any disease has dynamic flows of the propagation of causality as its internal processes, and is the enactor of processes such as branching and extending its causal chain of disorders as its external processes

10

From Ontology

to a computational model

11

Computational model of diseases

Two kinds of diseases (Type1) those whose etiological and pathological processes are well-understood and (Type2) other diseases, and

we mainly discuss type 1 diseases, but it also applies to type 2 diseases.

Definition 2: Is-a relation between diseases Disease A is a super class of disease B if all of the causal chains at the class

level of disease A are included in those of disease B. The inclusion of nodes (disorders) is judged by taking an is-a relation between the nodes into account, as well as sameness of the nodes.

Definition 3: Core causal chain of a disease The causal chain of a disease included in the chains of all its subclass diseases

is defined as the core causal chain of the disease. An example of the core causal chain in the case of (non-latent) diabetes is:

deficiency of insulin elevated level of glucose in the blood

12

Clinical imbalance modelas a concretization of disposition

From now on, I discuss only Type 1 diseases In OGMS, a disposition to diabetes inheres in <deficiency of insulin> The idea of disturbance of homeostasis to capture <deficiency of insulin> For each parameter participating in homeostasis, there must be the notion of bala

nce and regulation functions

Definition 4: Clinical imbalance Clinical imbalance is a local phenomenon of homeostasis in the human body and is

defined as a state where the difference between the amounts of supply and demand is out of the range specified for the parameter under consideration.

In the case of diabetes, the supply is the performed amount of the insulin operation and the demand is the required amount of the insulin operation.

In a normal case, the difference between the two amounts is within a certain range, that is to say, “balanced”. In an abnormal case, on the other hand, an imbalance (deficiency of insulin) occurs, which can be a disposition to the initiation of the pathological process of diabetes.

13

Use cases of the imbalance model

Four qualitative values We applied the imbalance model to

Diabetes including latent one Infectious diseases (including Hepatitis) Osteoporosis

fracture caused by osteoporosis is modeled using medium value for demand to resist the normal pressure given as an external cause and small value for supply to resist external pressure

Hyperlipidemia Ischemic heart diseases Etc.

14

Concluding remarks

A new definition of diseases has been proposed within the framework of OGMS

A disease is understood as a causal chain of clinical disorders

OGMS’s disposition has been concretized as an Imbalance model based on the OGMS’s notion of homeostasis

We obtained a computational model of diseases We are now building an ontology of about 6000 diseases of

12 different divisions

15

Thank you!

(Especially to Barry who kindly polished up my poor English)

16

Definition of diseases in OGMS

a disposition (i) to undergo pathological processes that (ii) exist in an organism because of one or more disorders in that organism (excerpt from OGMS paper)

This is a beautiful definition from a philosophical point of view. At the same time, however, it is not very friendly to clinicians because it lacks practicality

17

The life of a river

overflow

minimal flow reaches the sea

branches

would reachanother lake

might dry up

extends and changesits route, etc.

18

An example of causal chain of events

…

Causality (Causal relationship)

A person is injured by a car accident

A witness calls an ambulance

An ambulance arrives

…

t

Duration of events

t1 t2 t3 t4 t5 t6

Each event is caused by a completed previous event.

• The whole event corresponds to non-cumulative process

t0 t7

it arrives ata hospital

< <<

19

An example of causal chainof a disease

…

A blood clot is growing ．

The cross section of blood vessel is reducing.

Amount of oxygen supplied is reducing.

Cells are dying.

…

tt1 t2

Each process is not completed (keeps going) at every instant in time while the whole process is proceeding.

• It is a cumulative continuous process• So, the issue is how to derive a continuant from

a cumulative continuous process

Disorder


Duration of process

A continuous　 process

A continuous　 process

A blood clot grows

The cross section of blood vessel reduces

Amount of oxygen supplied reduces

Cells die

21

Homeostasis in OGMS

a disposition of the whole organism (or of some causally relatively isolated part of the organism, such as a single cell) to regulate its bodily processes in such a way as (1) to maintain bodily qualities within a certain range or profile and (2) to respond successfully to departures from this range caused by internal influences or environmental influences such as poisoning

22

Types of diabetes constituted of causal chains

Disorder (nodes)

Causal Relationship

Causal chain of a disease(each color represents a disease)

Legend

loss of sight

Elevated level of glucose in the blood

Type I diabetesDiabetes-related Blindness

Steroid diabetes

Diabetes…

…

……

…

…

…

… … …

…

possible causes and effects

Destruction of pancreatic beta cells

Lack of insulin I in the blood

Long-term steroid treatment

Deficiency of insulin

Core causal chain of non-latent diabetes

Coloredregions

23

A representation of the clinical imbalance model

…

……

…

… …

Clinical imbalance

imbalance

Performing amount of insulin operation(supply)

(possible)Maximal amount of insulin operation

Balance between supply and demand

Required amount of insulin operation (demand)

loss of sight


Type I diabetes

Diabetes-related Blindness

Steroid diabetes

DiabetesDestruction of pancreatic beta cells



Imbalanceoutcome

24

Four qualitative values

Legends of values

balance

Performing amount of the operation (supply)

(a) Normal situation

balance

imbalance

(possible)Maximal amount of the operation


Required amount of the operation (demand)

Demand side Supply side

Very Large

Large

Middle

Small









(b) Hard or very active situation

(c) When demand increases up to very large

25

defective insulin secretion

…

Clinical imbalance

insulin resistance

Medium

Medium

Medium

25

Latent diabetes in daily life


(Possible) maximal amount of insulin operation

Balance

Performing amount of insulin operation

Daily mean load


26

Clinical Imbalanceovereating

Large

Medium

Medium

26

Manifestation (Latent diabetes)

increased demand of insulin supply




Imbalance ImbalanceOutcome

Exceed the daily mean load level


insulin resistance Elevated level o

f glucose in the bood


27

break the biological defense

break the biological defense

proliferationproliferationInvasionInvasion

Invasion (Infection) stageInvasion (Infection) stage Onset stageOnset stage

colonizationcolonization

onsetonset

Infectious diseaseTwo-stage imbalance model for infectious diseases:① Invasion stage 　②Onset stage

27

28

Common coldCommon cold

No virus

virus invasion

Medium

　　

Medium

No onset stageNo onset stage

OnsetOnset

virus invasion

virusExternalcause

t1 　　　　　　　　　　

timet2 　　　　　　　　　　

t3 　　　　　　　　　　

virus infection Onset

28

Required amount of defense against invasion (demand)

Performing amount of defense againstinvasion

(Possible) maximal amount of immune defense against invasion

no deficiency ofimmune defense

runny nose, throat pain 　

(Possible) maximal amount of defense against outcome

Performing amount of defense againstoutcome

deficiency of defense

balance

Invasion stageInvasion stage Onset stageOnset stage

Large

Medium

29

imbalance of pathogen invasion defense imbalance of pathogen invasion defense

deficiency of immune defenseagainst invasion

Performing amount of defense againstinvasion(supply)

(Possible) maximal amount of immune defense against invasion

　　Pathogeninvasion


pathogenInfection



　　

imbalance of onset imbalance of onset

Other attack

Onsetpathogen infection

OnsetOnset

pathogenInvasion

pathogenExternal cause

t1 　　　　　　　　　　

timet2 　　　　　　　　　　

t3 　　　　　　　　　　

29

ImbalanceOutcome


Required amount of defense against outcome (demand)

ImbalanceOutcome

Imbalance Imbalance

deficiency of defense againstonset

Pathogen

hepatic fibrosis


onset

Large

Large

Very large

Large

Very large

Large

30

Hepatitis （ asymptomatic children ）Hepatitis （ asymptomatic children ）

Along with the development of immune system, symptom caused by excess immune defense appears.

VeryLarge

excess immune defense (supply)excess immune defense (supply)

destruction of hepatic cells

destruction of hepatic cells→ hepatic fibrosis→ cirrhosis

Adverse effect of immune defense

no deficiency of immune defense

Performing amount of immune defense againstvirus invasion (supply)

(Possible) maximal amount of immune defense against virus invasion

　　Hepatitis virusinvasion

Hepatitis virusInfection


Small

Small

VeryLarge

no deficiency of immune defense

balanceLarge

onset

In the childhood, due to immune system is immature, the maximal amount of immune defense was very small

demand

Performing amount of immune defense (supply)

ImbalanceVerysmall

31

OsteoporosisOsteoporosis

deficiency of bone formation

Performing amount of bone formation (supply)

(Possible) maximal amount of bone formation

　　

bone density loss

amount of bone resorption (demand)

(Possible) maximal amount of resist external pressure

　　

fracturefracture

falling

31

ImbalanceOutcome

Performing amount of resist external pressure (supply)

amount of external pressure(demand)

Imbalance

deficiency of resist external pressure

OsteoporosisOsteoporosis

fractureSmall

Small

Medium

steroidadministration

menopause

internal cause

external cause

Small

Small

osteoclastactivation

ImbalanceImbalanceOutcomeSmall

Large

32

environmental factors(high-fat diet…)

Intolerable load on the lipid influx

Small

32

Clinical imbalance

Lipid influx

(Possible) maximal amount of lipid efflux

Performing amount of lipid efflux

ImbalanceImbalanceOutcome

Elevated level of lipid in the blood (hyperlipidemia)

Balance between efflux and influx of lipids

genetic factors

HyperlipidemiaHyperlipidemia

Small

Large

34

How can we correctly capture latent diabetes?

34

II

Pancreaticβ cell

G

Insulin secretion

GG

G

blood

LiverMuscle …

IIIIII

IIII

Glucose uptakeregulation

G GG G

G

Overload!

Type 1 Diabetes ： chronic severe elevated blood glucose by destruction of insulin-producing beta-cells

Latent diabetes ： There is no symptom in the daily life, but will have problems with elevated blood glucose by some loading on insulin action.

35

MalariaMalaria

deficiency of immune defenseaction

Performing amount of defense action (supply)

(Possible) maximal amount of immune defense action

　　

Externalcause

plasmodiuminvasionvia a mosquito bite

Infection (liver cell)

Required amount of defense action (demand)

Large

Very Large

ImbalanceOutcomeLarge

Imbalance brain damage（ Cerebral Malaria ）

destruction of red blood cells

plasmodium infection

anemia

OnsetOnset

plasmodiuminvasion

Externalcause

t1 　　　　　　　　　　

timet2 　　　　　　　　　　

t3 　　　　　　　　　　

Onset of disease


Plasmodium

Plasmodium

36

Imbalance model•Ischemic heart disease 　 myocardial infarction

Imbalance: O2 demand 　ｖｓ．　 O2 supplyImbalance outcome ：　 ischemic necrosis

•OsteoporosisImbalance: bone resorption 　ｖｓ． bone formationImbalance outcome ：　 bone density loss

•Fracture　　　 Imbalance: external pressure ｖｓ．　 Resistance to external pressure–Fracture caused by osteoporosis

Imbalance: external pressure ｖｓ．　 Resistance to external pressure

–Fracture caused by traffic accidentImbalance: external pressure ｖｓ．　 Resistance to external pressure

•HyperlipidemiaImbalance: lipid influx 　ｖｓ．　 lipid efflux in the blood Imbalance outcome ：　 hyperlipidemia (elevated levels of lipids in the blood)

36

smallsmallmediummedium

Imbalance OutcomeFracture

very largevery large mediummediumImbalance Outcome

Fracture

37

Supplement the cases of resistant peripheral receptors which are also covered by

our model. The model except the imbalance model is applicable to either type of

diseases thanks to the flexibility of the granularity and the degree of “well-understood”-ness by putting “unknown” node of cause

38

An example of causal chain of processes

…

The engine isgeneratingtorque

The shaft is transmitting the torque to wheels.

Wheels are transforming the torque toforward-going force

The car is movingforward

…

t

Duration of process

t1 t2

Each process is not completed (keeps going) at every instant in time while the whole process is proceeding.

• A causal chain as a flow of causality (propagation of causality) is internal process (=cumulative continuous process) of the causal chain. - The flow corresponds to a flow of the water in a river.


Continuous　processes

Continuous　processes

39

Causal chain

1. Causal chain =def a chain of entities linked by causality. There can be a causal chain of disorders, causal chain of processes, causal chain of events, etc.[1]

[1] What constitutes causality is outside the scope of our research.

40


Disorder (nodes)

Causal Relationship

Core causal chain of a disease(each color represents a disease)

Legends

loss of sight



Steroid diabetes

Diabetes…

…

……

…

…

…

… … …

…






42

No virus

virus invasion

Small

Small


Performing amount of defense againstinvasion

(Possible) maximal amount of imumne defense against invasion


balance

Healthy personHealthy person

no deficiency ofimmune defense

OnsetOnset

virus invasion

virusExternalcause

t1 　　　　　　　　　　

timet2 　　　　　　　　　　

t3 　　　　　　　　　　

virus infection Onset


　　

No onset stageNo onset stage

42



deficiency of defense

Small

43

Computational disease model

43

…

……

…

causal chaincausal chain

clinicaldisorder

clinicaldisorder

CausalRelationship

44


Disorder (nodes)

Causal Relationship

Causal chain of a disease(each color represents a disease)

Legend

loss of sight



Steroid diabetes

Diabetes…

…

……

…

…

…

… … …

…






Core causal chain of diabetes

45

DiabetesClinical imbalance model

…

……

…

… …

Clinical imbalance

imbalance

Performing amount of insulin operation(supply)

(possible)Maximal amount of insulin operation



loss of sightType I diabetes

Diabetes related Blindness

Steroid diabetes

DiabetesDestruction of pancreatic beta cell

Lack of insulin in the blood

Long term steroid treatment

Elevated level of glucose in the bood

46 46

Required amount of the operation(demand)

Cause

internal cause

external cause

ImbalanceOutcomeImbalance

Performing amount of the operation(supply)

Clinical Imbalance

(Possible)Maximal amount of the operation


Very Large

Large

Medium

Small

SupplySide

DemandSide

47


Intolerable load on the demand side

Daily mean load

Type I (non-latent) diabetes

Small

Small

Medium

47

Clinical imbalance




ImbalanceImbalanceOutcome



48

Informal account of diseases

After it begins to exist, it undergoes extending, branching, and disappearing processes before the patient is cured, not always though

Such an entity (a disease) can change according to its phase while keeping its identity, and hence it seems to be a continuant

It could be something related directly to a manifestation process of the disease rather than a disposition itself

At the same time, a disease should not be a process (occurrent) This is why defining a disease is difficult

Although the introduction of the notion of disposition is one way to solve this problem, disposition is a bit too far from what its manifestation process implies/suggests

1 river flow model of diseases riichiro mizoguchi, kouji kozaki, hiroko kou, yuki yamagata isir,...

Documents