Moving Medical Semeiotics to the Digital Realm

SEMEOTICONS Approach to Face Signs of Cardiometabolic Risk

Giuseppe Coppini

, Riccardo Favilla

Amalia Gastaldelli

, Sara Colantonio

and Paolo Marraccini

CNR Institute of Clinical Physiology, Pisa, Italy

CNR Institute of Information Science and Technology, Pisa, Italy

Keywords: Face Semeiotics, Cardiovascular Diseases, Cardio-metabolic Risk Factors, Cardio-Metabolic Risk Signs,

Personalized Prevention, Self-monitoring, Virtual Individual Model, Well-being Index.

Abstract: In modern medicine signs derivable from face observation remain an important part of the physical

examination that, together with the anamnesis, constitutes the basis for a rational decision-making.

Therefore, face semiotics may be considered as a potential source of information for obtaining markers of

obesity, metabolomics, cardiovascular homeostasis and psychophysical status. Once properly mapped to

computational descriptors, their systematic exploitation is expected allowing the building of effective self-

monitoring systems. In this perspective, in the frame of the FP7 project SEMEOTICONS, the most relevant

face signs of cardio-metabolic risk are reviewed and analysed so as to drive their detection, quantification

and integration into a virtual individual model useful for cardio-metabolic risk prevention.

1 INTRODUCTION

Human face has been always considered to be a

mirror of emotions, mood and health status. Face

signs have been studied since the time of Aristotle.

Hippocrates already described aspects of

pathological conditions related to face, becoming

common heritage to associate face traits with

character, psychological dispositions and health

status.

In present days, medical semeiotics deems the

face as an important revelaer of precious information

about the healthy or unhealthy status of individuals,

produced by the combination of biophysical signs

and expressive features. Experienced medical

doctors acquire a personal and typical ability in

reading and interpreting the complex and composite

signs of patients’ face. These signs usually suggest

how to steer the medical examination and may

contribute to suggest which diagnostic investigations

are to be prescribed. So far, although its striking

importance, semeiotic evaluation has not raised a

systematic scientific interest. Despite the face

represent a naturally pre-eminent mean for

communication among human beings and

modulating inter-personal interaction, the valuable

pieces of information conveyed by human face have

not been comprehensively investigated with the aim

to assist medical professional and individuals with

computational tools.

Nowadays, results from clinical and

epidemiological studies strongly stimulate the

development of personalized health care systems.

This is also in view of new and more efficient

strategies for disease prevention, which is a strategic

objective for modern health systems. On the other

hand, personalized systems working at individual

home offer a significant way to contain health care

costs with improved effectiveness and efficiency.

Proficient self-monitoring coupled to individual

education and coaching is key point to build systems

capable to help people staying healthy. The FP7

project SEMEOTICONS (SEMEiotic Oriented

Technology for Individual’s CardiOmetabolic risk

self-assessmeNt and Self-monitoring) has as a major

technological objective the building of a

multisensory system having the appearance of a

conventional mirror, called the Wize Mirror, to be

hosted in everyday life environments (including

individuals’ home, pharmacies, fitness centres and

schools). The Wize Mirror will exploit face

semeiotics to instantiate a Virtual Individual Model

(Honka et al., 2011) from which a Well-Being Index

(WBI) will be obtained. Starting form an initially

defined state, tracking WBI temporal evolution will

enable individual to self-monitor life-style related

606

Coppini G., Favilla R., gastaldelli A., Colantonio S. and Marraccini P..

Moving Medical Semeiotics to the Digital Realm - SEMEOTICONS Approach to Face Signs of Cardiometabolic Risk.

DOI: 10.5220/0004939106060613

In Proceedings of the International Conference on Health Informatics (SUPERHEAL-2014), pages 606-613

ISBN: 978-989-758-010-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

risks for cardio-metabolic diseases and implement

personalized prevention actions.

In the following, we will review and analyse

some basic aspects of face semeiotics so as to define

the traits of a related computational model.

2 MEDICAL SEMEIOTICS

The face is a fine descriptor of a person’s well-being

state and, people, not only doctors, commonly derive

from the observation of the face significant clues

about psychophysical condition. Evidence on the

state of nutrition, fitness, and mental state can be

obtained. In addition, conditions affecting the colour

or the appearance of the skin can be also revealed.

The appearance and features of the face allow the

distinction among ethnicity, gender, age and

emotions (such as happiness, sadness, fear, anxiety,

and pain). Face changes can be due to alterations of

skeletal and/or muscular structure, subcutaneous

tissue, colour of the skin and eyeballs appearance.

For examples, chronic endocrinological diseases

(achondroplasia, acromegaly) and congenital anemia

(thalassemia) may produce characteristic alterations

of bone structures. Diseases of the nervous system

(Parkinson, myasthenia, tetanus) may cause typical

variations of the muscular structures. Other local and

systemic illness may induce modifications of the

superficial tissues due to changes of water content,

growth of adipose tissue, and deposition of

mucoproteins such as in the case of myxoedema

(hypothyroidism). Haemoglobin concentration,

oxygen saturation, vasodilation or vasoconstriction

affects the colour of facial skin (pallor, redness, and

cyanosis); moreover the deposit of other substances

may be responsible of pathologic appearance of the

skin, as bilirubin in jaundice. Local accumulation of

cholesterol may become evident with the appearance

of xanthelasmas in the eyelid and arcus cornealis, a

white ring in front of the periphery of the iris.

Moreover some clusters of characteristic features of

the face are considered pathognomonic of specific

medical conditions such as mitral face (mitral

stenosis), Hippocratic face (sepsis), lunaris face

(Cushing's syndrome, obesity) and other well-known

semeiotic facies.

From this brief summary, it is evident that

building a comprehensive model of face semeiotics

able to capture all the available pieces of

information is an extremely complex task.

Therefore, focusing on a specific application helps

to make the problem tractable. Moreover, working

with a “real world” setting is expected offer a

significantly general framework for further

utilization.

That led us to focus on cardiovascular diseases

(CVD) and cardio-metabolic risk for which the need

of personalized prevention strategies has gained a

universal acceptance.

3 CARDIOMETABOLIC RISK

Atherosclerotic cardiovascular diseases (CVDs),

including heart disease and stroke, are the leading

causes of mortality worldwide (World Health

Organization, 2008). The atherosclerotic illness

develops insidiously, and clinical manifestations

often become evident in its advanced stages.

Altogether, frequently, the major events, such as

serious health complications, disability and death

occur between 40 and 60 years of age. Moreover, the

majority of patients who survive a myocardial

infarction do not fully recover the ventricular

function, and many stroke survivors have physical

limitation in the daily activities. This explains why

CVDs represent one of the major challenges to the

health systems and considerable efforts are profuse

to treat clinical manifestations of CVDs. These

efforts have granted significant advances with actual

improvements in patients’ outcome, quod ad vitam

and valitudinem (Ford et al., 2007).

Despite the success of the pharmacological,

interventional, and surgical treatment of the CVDs,

it is obvious that all these therapies cannot modify

the epidemiological impact of the disease.

Moreover, the cost of health systems grows

exponentially with the widespread use of complex,

and often inappropriate, diagnostic procedures, as

well as with population aging. At present, the

strategy of prevention, which attempts to modify

some pathophysiological factors related to the

genesis of the disease, seems to be the only way to

limit the epidemic growth of CVDs (Graham et al.,

2007). In a recent paper (Pandya et al., 2013) on

forecasting cardiovascular disease in the USA

through the year 2030, an inversion of the

epidemiologic trend was found, which predicts an

increase in the overall incidence of cardiovascular

disease. This trend is related to two independent

factors: the aging of the population and the

incidence of obesity and diabetes.

Cardio-metabolic risk is a cluster of risk factors

indicative of a patient's overall risk for CVD and

type-2 diabetes. These risk factors include: incorrect

dietary habits, physical inactivity, smoke, alcohol

abuse, abnormal lipid metabolism, hyper-glycaemia,

and arterial hypertension (Grundy et al., 2005,

National Cholesterol Education Program, 2002,

MovingMedicalSemeioticstotheDigitalRealm-SEMEOTICONSApproachtoFaceSignsofCardiometabolicRisk

607

Kahn, 2005). In particular, the metabolic syndrome

is characterized by groups of clinical and metabolic

features that include high triglycerides, low HDL

cholesterol, high blood pressure, high fasting

glucose, visceral adiposity (Alberti et al., 2006,

Lorenzo et al., 2007). Epidemiological studies have

shown that persons with metabolic syndrome have

morbidity and mortality for cardiovascular disease

3-4 times increased as compared to control

population (Isomaa et al., 2011).

The importance of primary prevention for the

decrease of cardiovascular epidemic is well

documented by epidemiological studies (Tunstall-

Pedoe, 2003) Moreover the impact on mortality of

prevention is judged higher in comparison with the

effects of evidence-based therapies such as medical

and interventional treatments (Ford, 2007; Di Chiara

et al., 2009). According to this observation some

clinical trials and observational studies have shown a

rapid decline in the risk for cardiovascular disease

mortality after individual or population-wide

changes in diet and/or smoking and in general

following a healthy life-style (Capewell et al., 2011).

In addition, the favourable impact of prevention on

human wellbeing and economics has been estimated

by committees of several countries (Capewell et al.,

2011, National Institute for Health and Clinical

Excellence 2010). Unfortunately, the adherence to

the recommended lifestyles and the proportions at

goal for blood pressure, lipids, and blood glucose in

patients at high risk resulted less than 50% in

European surveys (Kotseva et al., 2009; 2010).

Educational programs and lifestyle interventions

represent effective tools for reducing cardio-

metabolic risk profile and incidence of CVDs

(Laaksonen, 2002, Tuomilehto et al., 2001, Vale et

al., 2005). However, maintaining a healthy lifestyle

frequently needs the counselling and supervision of

various health professionals such as dieticians,

physical trainers, psychologists and behaviourists.

Such a prevention strategy is individually tuned and

requires an expensive organization of the health

systems.

This evidence suggests the need for new strategies

aiming to directly involve people and families in this

important task (Aktas, 2004). Self-monitoring is an

effective tool to stimulate individual awareness of

physical cues and/or behaviours and to identify the

barriers to changing behaviour. It may allow the

recognition of goals and may provide direct

feedbacks guarantying discretion and confidentiality.

At the same time, people may choose to activate

external communication with prompts such as

personal digital assistant or health care professionals

(Appel et al., 2003, Wing, 1999). Ward et al., (2010)

analysed clinical trials that used self-monitoring in

the area of cardiovascular risk management. They

indicated 4 major interventions obtainable with self-

monitoring strategies: a) education b) self-

measurement c) adjustment of (or adherence to)

behaviour d) contact with health professionals

(Figure 1).

Figure 1: Basic self-monitoring strategy.

As a matter of fact, a rationale alternative to the

intensive use of conventional individual coaching is

the exploitation of properly developed systems for

self-learning and self-monitoring. These systems are

expected to help people to change and maintain their

lifestyle providing tailored suggestions about

nutrition, weight, physical activity, fatigue, and

stress according to daily surveys. Moreover, data

collected by such coaching systems could be

analysed and interpreted by health care professionals

so as to support decision-making targeted to the

specific individual conditions. This approach has the

potential to result highly cost-effective and might

foster the diffusion of self-coaching systems with

favourable impact on social, physiological, and

environmental factors that, at present, remain

barriers for the success of large-scale preventive

intervention on CVD and diabetes.

4 CARDIO-METABOLIC RISK

AND FACE SIGNS

In Table 1 we list a set good indicators of patient’s

overall risk for CVD and type-2 diabetes. These risk

factors are classified in two groups: modifiable and

non-modifiable. The majority of them belong to the

modifiable category.

In SEMEOTICONS, we propose the adoption of

self‐

measuremen

coach

adjust

HEALTHINF2014-InternationalConferenceonHealthInformatics

608

a digital semeiotics of the face, that is the

computational evaluation of facial signs, focused on

those signs that relate to some widely recognized

risk factors of CVDs. Consequently, a semeiotic

model of the face can be defined taking into account

signs concerning: obesity, diabetes, hyper-

cholesterolemia, endothelial dysfunction, and

psychological status.

Other general semeiotic signs can be considered

to gain more general information about the overall

individual’s condition.

Each of the above-mentioned risk factors has

several observable manifestations on an individual’s

face.

Table 1: Main cardio-metabolic risk factors.

Modifiable factors

Non-modifiable

factors

Overweight/Obesity

High LDL

cholesterol

Low HDL

cholesterol

High triglycerides

hyper-coagulation

High blood glucose

Hypertension

Inflammation

Smoking

Unhealthy

eating

Health

disparities

Physical

inactivity

Psychosocial

issues

Age

Race/Ethnicity

Gender

Family history

Obesity: The general appearance and several

features of the face are relevant indicators of

overweight and obesity. The reconstruction of face

shape from images and videos and its

characterization through a detailed morphometric

analysis can, then, serve to localize and evaluate a

fatty physiognomy.

Diabetes: The metabolic alterations due to diabetes

favour the glycation of proteins and the

accumulation of Advanced Glycation End-products

(AGE) in the skin (Singh et al., 2001). According to

recent results in AGE detection and measurements,

autofluorescence stimulated by UV light is a viable

solution to detect AGE in sub-cutaneous layer.

Diagnostic instrument working in contact with

forearm skin are commercially available for clinical

use (Meerwaldt et al., 2004). Contact-less

measurement of AGEs looks a challenging task.

Using hyperspectral-imaging methods is expected to

provide a meaningful solution to the problem.

Hyper-cholesterolemia: High level of blood

cholesterol could result in some typical signs that

have been often related to hypercholesterolemia. In

particular, this may be the case of xanthelasmas in

the periocular region, as well as the arcus cornealis

in the iris border. Both these signs are detectable and

reasonably quantifiable in the images of the face

with proper morphological/structural analysis

methods. Furthermore, one expects that visible -

near infrared (VIS-NIR) spectroscopy and imaging

may be used to assess the accumulation of

cholesterol in the skin of the face and in the iris

(Ikawa et al., 2009). As in the case of AGEs, contact

measurement of skin cholesterol is described in

literature using different approaches (Zawydiwski et

al., 2001; Mancini et al., 2002), while contactless

measurements are still open to scientific

investigation. It must be also pointed out that a

deepened knowledge about skin cholesterol

accumulation processes and their relation to cardio-

metabolic risk would be an important achievement.

Endothelial Dysfunction: Endothelial dysfunction is

a major physio-pathological mechanism that leads

towards coronary artery disease. Broadly speaking,

the endothelium function can be seen as the

capability of the endothelium to balance between

vasodilating and vasoconstricting substances

produced by (or acting on) the endothelium.

Endothelial dysfunction can result from and/or

contribute to several disease processes (e.g.

hypertension, hypercholesterolemia, and diabetes)

and it can also result from environmental factors

(e.g. smoking tobacco, exposure to air pollution).

Endothelial dysfunction is therefore highly

informative about individuals’ health status. Non-

invasive assessment of endothelial dysfunction may

be obtained by per cent Flow Mediated Dilation

(FMD) measured by brachial artery ultrasound

imaging (Peretz et al., 2007). Another simple non-

invasive evaluation of endothelium function adopted

in clinical practice is based on peripheral artery

tonometry (PAT) (Kuvin et al., 2007), which works

by measuring Reactive Hyperaemia Index and has

been tested in several clinical trials at multiple

centres (including major cohort studies such as the

Framingham Heart Study, the Heart SCORE study,

and the Gutenberg Health Study). A simple

alternative to these methods is based on studying

microcirculatory blood flow after local heating

(Joyner et al., 2001). The key point in the latter

approach is that variations in blood supply to sub-

cutaneous districts result in variations of skin optical

properties (“colour”). Consequently, an accurate

analysis of temporal sequences of face images in the

visible and a near-infrared band is a practicable way

to evaluate elementary hemodynamic parameters.

These can relate to heart rate, oxygen saturation,

MovingMedicalSemeioticstotheDigitalRealm-SEMEOTICONSApproachtoFaceSignsofCardiometabolicRisk

609

Figure 2: Wize Mirror hardware scheme, including Multi Spectral Camera (MSC), standard lights (L) and multicolour lights

(SMLi, i=1, ...,4).

peripheral resistances and also endothelial function.

Actually, after providing a controlled thermal

stimulus to trigger a vasodilatation, the evaluation of

such hemodynamic response gives a measure that

can be correlated to the endothelial function.

Moreover, the analysis of heart rate variability can

serve to the characterization of the autonomous

nervous system.

Psychological Status: The morphological appearance

of human face can reveal useful information about

an individual’s mood, anxiety, and status of fatigue

(Pantic and Bartlett, 2007). Facial expression

recognition and facial biometrics in 3D space are

expected to serve this scope. Integration of these

descriptors with other well-established cardio

metabolic risk signs into self-monitoring systems is

worth of investigation. Correlation of facial

descriptors of psychological status with

questionnaires can help to better understand their

role to assess an individual’s wellness status.

Other Semeiotics: Among the other signs that can be

related to well being status and may potentially

relate to different individual conditions going from

fatigue/tiredness and stress to specific diseases, the

following are worth of consideration:

– Face colorimetric features, e.g. pallor, jaundice,

redness.

– Regional skin surface temperature in the face.

– Respiratory rate.

5 SEMEOTICONS APPROACH

A main technological objective in SEMEOTICONS

is the development and design of a multisensory

system hosted into a hardware platform having the

exterior aspect of a conventional mirror (Wize

Mirror). A conceptual drawing of the Wize mirror is

shown in Figure 2.

It will include cameras and depth sensors to

enable 3D reconstruction of the user face. This will

be used for morphological analysis (overweight and

obesity description, feature recognition for

psychological status evaluation) as well as to drive

other acquisitions phases (normalization of face

position and orientation). Properly designed

multispectral cameras will allow obtaining data on

the cardio-respiratory system (heart rate, blood-

oxygen saturation, endothelial function, respiratory

rate), on the presence of products of glucose and

lipid metabolism in the skin. The inclusion of other

sensors is also planned. In particular, the Wize

Sniffer, i.e. a gas sensor able to analyse exhaled

gases, will help to monitor lifestyle habits (smoking

and alcohol consumption). The Wize mirror will

include a lighting system for image acquisition, UV

light to stimulate fluorescence mechanisms and

HEALTHINF2014-InternationalConferenceonHealthInformatics

610

Figure 3: Wize Mirror Virtual Individual Model.

thermal lamps for heat testing of endothelial

function. User interaction will occur by an integrated

touch screen.

According to section 4, images and signals have

to be mapped into a set of descriptors, each of them

being related to some aspects of cardio-metabolic

risk. Afterwards, to obtain indication about the

individual’s health status and discover noxious

lifestyles, the computed descriptors, which are

intrinsically disaggregated entities, need to be

integrated into a high-level representation able to

capture the individual as a whole. For that aim, one

has to take into account individual’s peculiarities,

including both family and personal history. The

Virtual Individual Model (VIM) implemented in the

mirror will carry out the integration task. The major

requirement on VIM is the ability to detect and

signal significant changes of face semeiotics to assist

user behavioural change. Therefore, an absolute

assessment of individual wellness is not mandatory.

The model instantiation process shall include a

preliminary user-profiling phase to be accomplished

at the beginning of any self-monitoring activity. This

phase aims at defining the psychophysical

characteristics of the individual at baseline. These

should be derived on the ground of previous medical

examinations, personal history, and interactive

psychological questionnaires.

Therefore, the user profile will serve to define

the initial status of the VIM and may also provide

important constraints on the user-interface

preferences. According to this approach, the

temporal evolution of the model will describe the

changes of the user’s psychophysical status starting

form a known initial condition. In general, past VIM

history would enter the computation of the current

status and is expected to contain useful indications

for user advising and coaching. For these reasons, a

track of the VIM evolution will be retained in the

Individual Diary.

The internal VIM representations and the related

adaption rules are designed to operate at the level of

input data and would be, in general, unsuitable for

user communication and interaction. For this aim

the system would compute the WBI, which

represents a synthetic description of VIM state. The

WBI will be a multidimensional entity whose

components should measure the different aspects of

individual lifestyle appropriateness (e.g. food,

physical activity, psychological status, noxious

habits). According to the profiling phase, WBI

components would be properly presented to the user

along with educational and coaching messages.

To summarize, detection and integration of signs

derived from the semeiotics of the face should lead

to build sensitive equipment to self-monitor the

Initial

Status

VirtualIndividual’s

Model

UserInterface

Well‐BeingIndex

Face

Morphology

Breath

Analysis

Face

Expression

Multispectral

analysis

Iris

Analysis

Individual

Diary

MovingMedicalSemeioticstotheDigitalRealm-SEMEOTICONSApproachtoFaceSignsofCardiometabolicRisk

611

psychophysical state and, possibly, elaborate

suggestions useful for optimizing the personal life

style and to trap the major cardio-metabolic risk

factors.

In this perspective, SEMEOTICONS aims at

building an innovative virtual individual’s model,

which is based on a set of objective signs, closely

related to cardio-metabolic risk profile and derived

from external physical examination of the face. The

model should serve to evaluate a subject well-being

status over time and allow early detection of

improper lifestyles as well as potentially dangerous

conditions mainly related to cardio- metabolic risk.

6 CONCLUSIONS

Medical semeiotics, in the era of magnetic

resonance, computed tomography and molecular

biology, is still a valuable resource that may be

useful in every condition and location independently

of other structural facilities. The development of

advanced imaging and biological diagnostic

techniques has probably attenuated the interest for

developing technological tools based on data

obtained from simple semeiotics. Nevertheless, the

role recently acquired by self-monitoring and self-

training has opened new, significant perspectives to

this well-established branch of classical Medicine.

Semeiotics offers a sound methodological frame

to build new computational tools also exploiting

innovative multi-sensing devices. The adoption of a

digital-semeiotics approach is expected to ease the

implementation of virtual individual models able to

effectively assist people in keeping a healthy life-

style.

In this view, SEMEOTICONS Wize Mirror will

be a new kind of self-monitoring system. Its

architecture will be absolutely non-obtrusive and

will be able to use personal data for coaching and

learner-adaptive messaging.

ACKNOWLEDGEMENTS

This work was partly supported by the EU FP7

Project SEMEOTICONS - SEMEiotic Oriented

Technology for Individual’s CardiOmetabolic risk

self-assessmeNt and Self-monitoring (Grant

agreement no: 611516)

REFERENCES

Aktas MK, Ozduran V, Pothier CE, Lang R, Lauer MS.

2004, Global risk scores and exercise testing for

predicting all-cause mortality in a preventive medicine

program. JAMA, 292:1462 – 1468.

Alberti KG, Zimmet P, Shaw J. Metabolic syndrome—a

new world-wide definition. A Consensus Statement

from the International Diabetes Federation. Diabet

Med (2006) 23: 469–480.

Appel LJ, Champagne CM, Harsha DW, Cooper LS,

Obarzanek E, Elmer PJ, Stevens VJ, Vollmer WM,

Lin PH, Svetkey LP, Stedman SW, Young DR;,

2013.Writing Group of the PREMIER Collaborative

Research Group. Effects of comprehensive lifestyle

modification on blood pressure control: main results of

the PREMIER clinical trial. JAMA, 289:2083–2093.

Capewell S, O’Flaherty M, 2011. Rapid mortality falls

after risk-factor changes in populations. Lancet,

378:752–753.

Di Chiara A, Vanuzzo D. 2009. Does surveillance impact

on cardiovascular prevention? Eur Heart J, 30:1027–

1029.

Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR,

Kottke TE, Giles WH, Capewell S. 2007. Explaining

the decrease in U.S. deaths from coronary disease,

1980 – 2000. N Engl J Med, 356:2388 – 2398.

Graham I, Atar D, Borch-Johnsen K, et al. 2007. European

guidelines on cardiovascular disease prevention in

clinical practice: full text. Fourth Joint Task Force of

the European Society of Cardiology and other

societies on cardiovascular disease prevention in

clinical practice (constituted by representatives of nine

societies and by invited experts). Eur J Cardiovasc

Prev Rehabil, 14 Suppl 2:1–113.

Grundy SM et al. 2005. Diagnosis and management of the

metabolic syndrome. An American Heart

Association/National Heart, Lung, and Blood Institute

Scientific Statement. Executive Summary.

Circulation, 112:2735-2752.

Honka A, Kaipainen K, Hietala H, Saranummi N, 2011.

Rethinking Health: ICT-Enabled Services to Empower

People to Manage Their Health. IEEE Reviews in

Biomedical Engineering, 4:119-139.

Ikawa T, Ishii K, Awazu K, 2009. Hyperspectral imaging

of lipids using near-infrared super continuum light.

WC 2009, IFMBE Proceedings 25/II: 630-632.

Isomaa B, Almgren P, Tuomi T, Forsén B, Lahti K,

Nissén M, Taskinen MR, Groop L. 2001.

Cardiovascular morbidity and mortality associated

with the metabolic syndrome. Diabetes Care, 24:683-

689.

Joyner MJ et al, 2001. Nitric oxide and neurally mediated

regulation of skin blood flow during local heating. J

Appl Physiol, 91:1619-1626.

Kahn R, Buse J, Ferrannin E, Stern M, 2005. The

metabolic syndrome: Time for a critical appraisal.

Diabetes Care, 28:2289-2304.

Kotseva K, Wood D, De Backer G, De Bacquer D,

Pyorala K, Keil U. , 2009. EUROASPIRE III: a survey

HEALTHINF2014-InternationalConferenceonHealthInformatics

612

on the lifestyle, risk factors and use of cardioprotective

drug therapies in coronary patients from 22 European

countries. Eur J Cardiovasc Prev Rehabil 16:121 –

137.

Kotseva K, Wood D, De Backer G, De Bacquer D,

Pyorala K, Reiner Z, Keil U. 2010. EUROASPIRE III.

Management of cardiovascular risk factors in

asymptomatic high-risk patients in general practice:

cross-sectional survey in 12 European countries. Eur J

Cardiovasc Prev Rehabil, 17:530 – 540.

Kuvin JT, Mammen A, Mooney P, Alsheikh-Ali AA,

Karas RH, 2007. Assessment of peripheral vascular

endothelial function in the ambulatory setting. Vasc

Med. 12:13-6.

Laaksonen DE, Lakka H-M, Salonen JT, Niskanen LK,

Rauramaa R, Lakka TA, 2002. Low Levels of Leisure-

Time Physical Activity and Cardiorespiratory Fitness

Predict Development of the Metabolic Syndrome.

Diabetes Care, 25(9): 1612-1618.

Lorenzo CL, Williams K, Hunt KJ, Haffner SM, 2007.

National Cholesterol Education Program - Adult

Treatment Panel III, International Diabetes Federation,

and WHO Definitions of the Metabolic Syndrome as

Predictors of Cardiovascular Disease and Diabetes.

Diabetes Care, 1:8-13.

Mancini GJ, Chan S, Frohlich J, Kuramoto L, Sulzer M,

Abbott D. 2002. Association of skin cholesterol

content, measure by a novel non-invasive method,

with markers of inflammation and Framingham risk

prediction. Am J Cardiol, 89:1313–6.

Meerwaldt R, Graaff R, Oomen PH, Links TP, Jager JJ,

Alderson NL, Thorpe SR, Baynes JW, Gans RO, Smit

AJ. 2004. Simple non-invasive assessment of

advanced glycation endproducts accumulation.

Diabetologia. 47:1324-30.

National Cholesterol Education Program, 2002. Third

report of the National Cholesterol Education Program

(NCEP) Expert Panel on Detection, Evaluation, and

Treatment of High Blood Cholesterol in Adults (Adult

Treatment Panel III). Final report. Circulation,

106:3143-3421.

National Institute for Health and Clinical Excellence, 2010

Prevention of Cardiovascular Disease: Costing

Report. 2010. Nice Public Health Guidance 25.

Pandya A, Gaziano TA, Weinstein MC, Cutler D, 2013.

More Americans Living Longer With Cardiovascular

Disease Will Increase Costs While Lowering Quality

Of Life. Health Aff, 32:1706-1714.

Peretz A, Leotta DF, Sullivan JH, Trenga CA, Sands FN,

Aulet MR, Paun M, Gill EA, Kaufman JD, 2007.

Flow mediated dilation of the brachial artery: an

investigation of methods requiring further

standardization. BMC Cardiovascular Disorders,

7:11.

Singh R, Barden A, Mori T, Beilin L, 2001. Advanced

glycation end-products: a review. Diabetologia, 44:

129-146.

Tunstall-Pedoe H (World Health Organization). MONICA

Monograph and Multimedia Sourcebook. 2003.

Pantic, M, Bartlett M, 2007. Machine Analysis of Facial

Expressions, Face Recognition. In: I-Tech Education

and Publishing, Vienna, Austria, pp. 377-416.

Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT,

Hamalainen H, Ilanne-Parikka P, Keinanen-

Kiukaanniemi S, Laakso M, Louheranta A, Rastas M,

et al., 2001: Prevention of Type 2 Diabetes Mellitus by

Changes in Lifestyle among Subjects with Impaired

Glucose Tolerance. N Engl J Med, 344(18):1343-

1350.

Vale M, Jelinek M, Best J, 2005. Impact of coaching

patients on coro- nary risk factors – Lessons from the

COACH Program. Disease Management & Health

Outcomes 13(4):225-244.

Ward LM, Carl Heneghan C, Perera R, Dan Lasserson D,

Nunan D, Mant D, Glasziou P, 2010. What are the

basic self-monitoring components for cardiovascular

risk management? BMC Medical Research, 10:105.

Wing RR, Jeffery RW, 1999. Benefits of recruiting

participants with friends and increasing social support

for weight loss and maintenance. J Consult Clin

Psychol; 67:132–138.

World Health Organization, 2008. The Global Burden of

Disease: 2004 Update, Geneva, Switzerland: World

Health Organization.

Zawydiwski R, Sprecher DL, Evelegh MJ, Horse- wood P,

Carte C, Patterson M. 2001. A novel test for the

measurement of skin cholesterol. Clin Chem, 47:

1302–1304.

MovingMedicalSemeioticstotheDigitalRealm-SEMEOTICONSApproachtoFaceSignsofCardiometabolicRisk

613