Academic Edi o : An oni Mo ell
Recei ed: 13 Ma ch 2025
Re ised: 17 Ap il 2025
Accep ed: 25 Ap il 2025
Published: 19 May 2025
Ci a ion: Vondikakis, I.; Poli i, E.;
Goulis, D.; Dimi akopoulos, G.;
Geo goulis, M.; Sal aou as, G.;
Kon ogianni, M.; B isimi, T.;
Logo he is, M.; Kakoulidis, H.; e al.
In eg a ed F amewo k o Managing
Childhood Obesi y Based on Biobanks,
AI Tools and Me hods, and Se ious
Games. Elec onics 2025,14, 2053.
h ps://doi.o g/10.3390/
elec onics14102053
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
In eg a ed F amewo k o Managing Childhood Obesi y Based
on Biobanks, AI Tools and Me hods, and Se ious Games
Ioannis Vondikakis
1,
* , Elena Poli i
1
, Dimi ios Goulis
1
, Geo ge Dimi akopoulos
1
, Michael Geo goulis
2
,
Geo ge Sal aou as 2, Me opi Kon ogianni 2, Theodo a B isimi 3, Ma ios Logo he is 3, Ha y Kakoulidis 4,
Ma ios P asinos 4, A hanasios Anas asiou 5, Ioannis Kakkos 5, Ele he ia Vellidou 5, Geo ge Ma sopoulos 5
and Dimi is Kou sou is 5
1Depa men o In o ma ics and Telema ics, Ha okopio Uni e si y o A hens (HUA), 17778 A hens, G eece;
[email p o ec ed] (E.P.); [email p o ec ed] (D.G.); [email p o ec ed] (G.D.)
2Depa men o Nu i ion and Die e ics, Ha okopio Uni e si y o A hens (HUA), 17671 A hens, G eece;
[email p o ec ed] (M.G.); [email p o ec ed] (G.S.); [email p o ec ed] (M.K.)
3Ne company-In aso SA, L-1253 Luxembou g, Luxembou g; heodo a.b isimi@ne company.com (T.B.);
ma ios.logo he is@ne company.com (M.L.)
4Telema ic Medical Applica ions L d., 18533 Pi aeus, G eece; [email p o ec ed] (H.K.);
[email p o ec ed] (M.P.)
5Biomedical Enginee ing Labo a o y, School o Elec ical and Compu e Enginee ing, Ins i u e o
Communica ions and Compu e Sys ems, 15772 A hens, G eece; [email p o ec ed] (A.A.);
[email p o ec ed] (I.K.); [email p o ec ed] (E.V.); [email p o ec ed] (G.M.);
[email p o ec ed] (D.K.)
*Co espondence: [email p o ec ed]
Abs ac : The g owing epidemic o childhood obesi y is a majo h ea o hei o e all de-
elopmen and poses a numbe o challenges o heal h sys ems. We p opose an in eg a ed
amewo k o comp ehensi ely add ess childhood obesi y. The p oposed a chi ec u e ad-
d esses essen ial da a managemen and p e-p ocessing unc ionali ies o suppo scalable,
secu e, and p i acy-p ese ing da a p ocessing in dis ibu ed en i onmen s. We a e also
inco po a ing a heal h da a-d i en AI app oach o p edic i e analy ics and decision sup-
po . The e is addi ionally a Use Engagemen Laye , which se es as he main poin o
in e ac ion o use s. I connec s indi iduals o sys em capabili ies, acili a ing da a collec-
ion, p og ess moni o ing, and insigh s. Finally, we p esen ou se ious games designed
o add ess p o ec i e ac o s (such as physical ac i i y and heal hy ea ing) and mi iga e
isk ac o s (such as excessi e sc een ime and unheal hy ood choices). The iden i ied
educa ional objec i es we e ansla ed in o game elemen s including goal se ing, social
suppo , and posi i e ein o cemen . In o de o acili a e ou app oach, we ha e desc ibed
he essen ial da a lows and use in e ac ions wi hin ou Biobank a chi ec u e.
Keywo ds: childhood obesi y; se ious games; beha io change; public heal h; in e en ion
s a egies; amewo k; AI ools
1. In oduc ion
Childhood obesi y is a c i ical public heal h challenge wi h a - eaching consequences
o physical, men al, and social well-being [
1
]. As he p e alence o childhood obesi y
con inues o inc ease wo ldwide, pa icula ly in de eloped na ions, i has become a majo
ocus o esea che s and heal hca e p ac i ione s alike [
2
,
3
]. The oo o childhood obesi y
a e complex and mul i ac o ial, in ol ing gene ics, li es yle beha io s, en i onmen al in lu-
ences, and socioeconomic ac o s [
4
]. Add essing hese issues equi es comp ehensi e and
Elec onics 2025,14, 2053 h ps://doi.o g/10.3390/elec onics14102053
Elec onics 2025,14, 2053 2 o 23
inno a i e solu ions ha go beyond adi ional app oaches, inco po a ing echnological
ools o engage child en, amilies, and heal hca e p o ide s in meaning ul ways.
In ecen yea s, ad ances in a i icial in elligence (AI), da a analy ics, and digi al heal h
ools ha e in oduced new oppo uni ies o ackle childhood obesi y mo e e ec i ely [
5
].
AI can enhance ou unde s anding o he pa e ns and isk ac o s associa ed wi h child-
hood obesi y h ough p edic i e modeling and pe sonalized in e en ion. Se ious games,
which use game mechanics o mo i a e he change in heal hy beha io s, o e an engaging
app oach o educa ing child en and p omo ing heal hie li es yles. In addi ion, Biobanks,
which collec and s o e biological da a, p o ide in aluable esou ces o analyzing gene ic
and me abolic ac o s ha in luence he isk o obesi y [
6
]. In eg a ing hese elemen s in o
a uni ied amewo k could e olu ionize he p e en ion and managemen o childhood
obesi y, pa ing he way o pe sonalized, e ec i e in e en ions [7].
Al hough he e a e nume ous echnologies o comba childhood obesi y, hey a e
o en implemen ed in isola ion, limi ing hei po en ial impac . AI ools, se ious games,
and Biobanks each con ibu e uniquely o unde s anding and managing obesi y, bu lack
cohesi e in eg a ion ha could maximize hei bene i s. Cu en obesi y in e en ions also
ace challenges wi h sus ained engagemen , pa icula ly among child en, and he e a e
gaps in le e aging gene ic da a o ailo in e en ions. Fu he mo e, public accep ance and
da a p i acy conce ns add complexi y o implemen ing hese echnologies in heal hca e.
Thus, a comp ehensi e, in eg a ed amewo k ha uni es Biobanks, AI ools, and se ious
games is needed o enhance use engagemen , pe sonaliza ion, and clinical e icacy in
comba ing childhood obesi y.
To he bes o ou knowledge, we p emie e an a chi ec u e ha in eg a es Biobanks,
AI ools, and se ious games in a uni ied amewo k o childhood obesi y managemen .
The con ibu ion o his wo k is summa ized as ollows:
•
We p opose a modula a chi ec u e ha seamlessly in eg a es da a om Biobanks,
p ocesses i h ough AI-d i en analy ics o assess obesi y isk and ecommend in-
e en ions, and deploys inno a i e applica ions o ein o ce heal hy beha io s in
child en. This a chi ec u e allows o a comp ehensi e app oach, whe e heal h, beha -
io al, and en i onmen al ac o s a e all add essed wi hin a single cohesi e sys em.
•
We implemen a sui e o se ious games, which a e designed o inc ease child en’s
nu i ional knowledge and p omo e physical ac i i y in an in e ac i e and sus ain-
able manne .
•
We p esen a chi ec u al da a- low diag ams illus a ing he in o ma ion exchange
be ween sys em componen s. These diag ams p o ide insigh in o how da a and
compu a ional p ocesses mo e h ough he amewo k a chi ec u e and how use s
in e ac wi h di e en elemen s o he a chi ec u e.
The emainde o his pape is s uc u ed as ollows. Sec ion 2p o ides a e iew o
he cu en li e a u e on childhood obesi y in e en ions, Biobanks, AI ools and me hods,
and se ious games, ou lining he exis ing gaps and he po en ial o in eg a ed solu ions.
Sec ion 3in oduces he p oposed a chi ec u e, de ailing each componen and how hey
in e ac wi hin he sys em. In addi ion, we desc ibe he se ious games de eloped as pa
o he amewo k, including hei design p inciples, educa ional objec i es, and echnical
speci ica ions. In Sec ion 4, we p esen a se ies o sequen ial diag ams ha illus a e he
essen ial in o ma ion pa hways and use engagemen p ocesses embedded wi hin ou
Biobank amewo k. Finally, Sec ion 5concludes he pape wi h a summa y o key indings,
discusses limi a ions, and p oposes di ec ions o u u e esea ch.
Elec onics 2025,14, 2053 3 o 23
2. E idence-Based Beha iou al Modi ica ion o he P e en ion o
Childhood Obesi y
2.1. Rela ed Wo k
2.1.1. Childhood Obesi y
Childhood o e weigh /obesi y is a mul i ac o ial disease, caused by a dynamic in-
e play be ween se e al ac o s ela ed o he “speci ic ex e nal exposome” (die , physical
ac i i y and sleep), hose ela ed o he “gene al ex e nal exposome” (pe ina al expo-
su es and social/buil en i onmen ), and hose ela ed o he “in e nal exposome” (gene -
ics/epigene ics and me abolomics) [
8
–
10
]. All hese ac o s di ec ly o indi ec ly a ec he
ene gy equilib ium, de ined as he balance be ween ene gy in ake ( om oods/be e ages)
and ene gy expendi u e ( om basal me abolic a e, exe cise- ela ed ene gy expendi u e
and die -induced he mogenesis) [
11
]. Al hough he ene gy equilib ium mus be posi i e
o suppo no mal physical g ow h du ing childhood, a g ea imbalance in a ou o ene gy
in ake can lead o excess body weigh and he de elopmen o o e weigh /obesi y in he
long e m.
Among he a o emen ioned isk ac o s, li es yle habi s, mos impo an ly die and
physical ac i i y, a e la gely modi iable and hus a e key a ge s o in e en ions agains
childhood o e weigh /obesi y.
Va ious componen s o die a y habi s ha e been in es iga ed in he de elopmen
o childhood o e weigh /obesi y, in pa icula ood and ood-g oup in ake [
12
], as well
as adhe ence o die a y pa e ns (e.g., Medi e anean die , Wes e n die ) [
13
] and meal
pa e ns (e.g., ea ing b eak as , ha ing dinne , he equency and ype o snacking) [
14
].
Cu en e idence poin s owa ds a signi ican de imen al impac o suga -swee ened
be e ages and as oods on o e weigh /obesi y, whe eas he ole o e ined g ains and
mea p oduc s emains less e iden [
12
]. Mo eo e , a highe adhe ence o p uden /heal hy
die a y pa e ns, which consis o high in akes o ui s, ege ables, whole g ains, ish, nu s,
legumes, and yogu , as well as low in akes o suga and animal a , has been associa ed
wi h dec eased odds o o e weigh /obesi y, whe eas a Wes e n- ype die a y pa e n, which
is cha ac e ised by suga y oods and d inks, p ocessed oods, as ood, animal p oduc s,
and e ined g ains, has been posi i ely associa ed wi h ma ke s o obesi y [
13
]. O e all,
die quali y seems o be a signi ican de e minan o childhood o e weigh /obesi y [
15
].
In ela ion o meal pa e ns, egula b eak as consump ion has been associa ed wi h lowe
body mass index and educed odds o o e weigh /obesi y, whe eas b eak as skipping
may be a de imen al ac o [
16
]. Besides b eak as , he e is also some e idence o suppo
ha egula amily meals a e p o ec i e owa ds obesi y, highligh ing a po en ial ole o he
amily en i onmen in ea ing pa e ns [16].
In ligh o he abo e, he mos ecen guidelines om he Ame ican Academy o
Pedia ics (AAP) ecommend educing suga -swee ened be e ages, a oiding b eak as
skipping, and consuming ui s and ege ables o 5 days as he mos e ec i e beha io al
s a egies o ea ing childhood o e weigh /obesi y [17].
Besides die , physical ac i i y is ano he impo an de e minan o body-weigh s a us,
being a main con ibu o o ene gy expendi u e, and heal h in gene al. A high le el o phys-
ical ac i i y has been associa ed wi h imp o emen s in se e al heal h ou comes, including
bone heal h, i ness, ca diome abolic p o ile, cogni i e unc ion, and men al heal h, bo h in
adul and you h popula ions [
18
]. Focusing on body-weigh s a us, he a ailable epidemio-
logical esea ch has consis en ly shown ha engagemen in physical ac i i y, pa icula ly
o mode a e- o- igo ous in ensi y, is in e sely associa ed wi h a ious adiposi y- ela ed
ou comes [
19
], and con e sely ha seden a iness, pa icula ly inc eased ime spen in
on o sc eens o ec ea ion, is associa ed wi h inc eased adiposi y [
20
], in child en’s
coho s. The iso empo al subs i u ion o seden a y ime wi h physical ac i i y has also
Elec onics 2025,14, 2053 4 o 23
been associa ed wi h lowe adiposi y indices among he you h, sugges ing ha inc easing
ime spen in physical ac i i y a he expense o seden a y ac i i ies migh ep esen he
op imal app oach o p e en childhood o e weigh /obesi y [
21
]. Besides epidemiological
da a, ecen ly published sys ema ic e iews and me a-analyses o clinical ials, mos ly
implemen ed a he school se ing, ha e concluded ha physical ac i i y, ei he alone o as
pa o mul icomponen li es yle in e en ions, is an e ec i e s a egy o he p e en ion o
childhood o e weigh /obesi y [22–24].
In ligh o he a o emen ioned e idence, he 2020 guidelines o he Wo ld Heal h
O ganiza ion (WHO) ecommend a minimum o 60 min o mode a e- o- igo ous physical
ac i i y pe day o child en and adolescen s in o de o achie e hese heal h bene i s [
25
].
Acco dingly, he 2020 posi ion s a emen o he Eu opean Childhood Obesi y G oup and he
Eu opean Academy o Pedia ics acknowledges physical ac i i y as a signi ican de e minan
o body-weigh s a us and p o ides ecommenda ions o adop ing a physically ac i e
li es yle, simila o hose o WHO, ha can con ibu e o he p e en ion o excessi e body
weigh in childhood and adolescence [
26
]. Howe e , based on a 2020 syn hesis o 298 su eys
om 146 coun ies, 81.0% o 11–17-yea -olds (77.6% o boys and 84.7% o gi ls) do no mee
hese ecommenda ions, highligh ing he u gen need o na ional and global ac ions o
p omo e physical ac i i y and educe seden a iness among child en and adolescen s [27].
2.1.2. AI in Heal hca e
A i icial in elligence (AI) is inc easingly being in eg a ed in o heal hca e, imp o ing
p edic i e models and enabling pe sonalised in e en ions. AI echnologies can help
add ess issues such as childhood obesi y h ough eal- ime da a analysis, isk-assessmen
models and ailo ed heal hca e plans.
A key applica ion o AI in heal hca e is p edic i e modelling, whe e algo i hms analyse
his o ical heal hca e da a o p edic u u e ends o a pa ien ’s e olu ion. In he case o
childhood obesi y, he e a e mul iple ac o s such as gene ics, li es yle and socio-economic
indica o s [28] ha need o be aken in o accoun by hese models.
Va ious algo i hms ha e been used o p edic childhood obesi y, such as Decision T ees
o p edic obesi y o e he 2–10 yea age ange [
5
], Random Fo es and G adien Boos ing
o p edic malnu i ion isk [
29
], and Neu al Ne wo ks and Suppo Vec o Machines ha e
been in es iga ed o hei po en ial o handle complex, non-linea ela ionships in obesi y
da a [
30
]. Machine lea ning models can accu a ely p edic obesi y isk in child en as young
as wo yea s old [31].
AI can enable pe sonalized in e en ion s a egies ailo ed o indi idual gene ic p o-
iles, en i onmen al ac o s, and beha io al pa e ns [
32
]. I has he po en ial o iden i y
ea ly indica o s o obesi y and enable a ge ed in e en ions. The in eg a ion o AI wi h
EHRs p o ides clinicians wi h eal- ime in o ma ion on pa ien p og ess and enables he
de elopmen o ailo ed ea men plans [
33
]. Using p ognos ic da a, AI can ailo die a y
ecommenda ions [
29
], exe cise egimes and psychological suppo mechanisms, ensu ing
ha in e en ions a e closely ailo ed o indi idual needs and inc easing hei e ec i eness.
To suppo such an AI-d i en app oach, access o di e se and high-quali y da ase s
is essen ial. Real-wo ld heal h da a a e o en subjec o s ic p i acy p o ec ions and
egula o y es ic ions, which a e essen ial o p o ec pa ien s. In his con ex , he use o
syn he ic da a has eme ged as a powe ul al e na i e, enabling he aining and alida ion
o AI models wi hou comp omising pa ien con iden iali y [
34
]. Resea ch pape s ha e
demons a ed ha syn he ic da ase s can main ain s a is ical simila i y o sou ce da a while
e ec i ely elimina ing e-iden i ica ion isks h ough me hods such as di e en ial p i acy
and noise injec ion du ing da a gene a ion [
35
]. These da ase s ha e been success ully
Elec onics 2025,14, 2053 5 o 23
employed in model aining, wi h negligible pe o mance di e ences compa ed o models
ained on eal da a [36].
2.1.3. Biobanks
A Biobank is a eposi o y ha collec s, s o es, and manages biological samples, such
as blood, issue, and DNA, and associa ed heal h and gene ic da a [
37
]. The main pu pose
o Biobanks is o acili a e scien i ic esea ch, o help scien is s unde s and disease, o use
da a o unco e complex gene ic ela ionships o imp o e diagnosis, and o complemen
esea ch wi h p ac ical applica ions [38,39].
Mode n Biobanks depend on IT in as uc u es capable o hos ing la ge and he e oge-
neous da ase s. Fo example, he Ge man Biobank Alliance has implemen ed a ede a ed
IT in as uc u e o manage and ha monise Biobank da a ac oss ins i u ions [40].
Biobanks can be used o iden i y he ac o s ha lead o disease, bo h gene ic and ac-
qui ed. In he Ne he lands, a Biobank has been es ablished o p o ide ongoing in o ma ion
on he p og ession and ea men o newly diagnosed pa ien s wi h ype 2 diabe es, wi h a
ocus on pe sonalised ea men [41].
Despi e hei i al con ibu ion, Biobanks ace many challenges. Fi s , hey ope a e
unde di e en egula o y, e hical, and ope a ional amewo ks, making da a sha ing
and collabo a ion complex. Ano he challenge is he managemen and analysis o la ge,
he e ogeneous da ase s. This in ol es in eg a ing gene ic, pheno ypic, and clinical da a,
while ensu ing da a secu i y and he p i acy o pa icipan s. Building us is essen ial o
ensu e ha his is sha ed and suppo ed by he popula ion a la ge.
2.1.4. Se ious Games
Game-based heal h in e en ions p oduce small bu signi ican BMI educ ions in
o e weigh /obese you h [
42
], wi h mul icomponen app oaches showing g ea e e ec-
i eness han s andalone in e en ions. Se ious games show p omise as an educa ional
s a egy o imp o e knowledge and encou age beha io al changes in o e weigh o obese
child en [
43
]. Se ious games can ackle childhood obesi y h ough enjoymen , mo emen ,
nu i ion educa ion, and social elemen s like eam play [
44
]. P o iding educa ional con en
h ough hem his makes knowledge abou nu i ion and physical ac i i y mo e a ac i e
and accep able o child en [45].
2.2. O e iew o P oposed F amewo k
We p opose a ede a ed and decen alised amewo k o ackle childhood obe-
si y le e aging ad anced da a collec ion, AI-powe ed analy ics and use engagemen
Figu e 1
. The main use in e aces o he a chi ec u e a e he heal h applica ion and he
se ious games sui e. The heal h applica ion se es as a pla o m o moni o ing key heal h
me ics, such as physical ac i i ies, sc een ime, sleep, well-being and p o iding ailo ed
ecommenda ions. Meanwhile, he se ious games le e age gami ica ion echniques o en-
cou age child en o adop heal hie beha io s h ough engaging and educa ional gameplay.
The dashboa d helps pa en s and heal hca e p o ide s make e idence-based decisions.
The heal h app and he se ious games will also be used o encou age child en o adop
heal hie li es yles. A communi y ne wo k and knowledge hub will p omo e collabo a ion
and dissemina e e idence-based esou ces. The da a collec ed om bo h he heal h appli-
ca ion and se ious games a e secu ely p ocessed and s o ed wi hin independen Biobank
edges ha collec i ely o m he Biobank, ensu ing p i acy and egula o y compliance.
Fu he da a ha monisa ion and pseudonymisa ion is unde aken o imp o e da a quali y
and secu i y, and GANs will be used o c ea e syn he ic da a o suppo esea ch while
main aining anonymi y. In addi ion, AI ools, such as isk assessmen and ecommenda ion
engines, a e used o p o ide pe sonalized heal h insigh s ha complemen he dashboa d.
Elec onics 2025,14, 2053 6 o 23
Figu e 1. The a chi ec u e o he sys em.
2.3. F amewo k Ad an ages
Ou p oposed amewo k o e s se e al meaning ul imp o emen s o e exis ing solu-
ions in key a eas. Cu en da a-in eg a ion app oaches end o ely on siloed collec ion
me hods wi h limi ed c oss-sys em compa ibili y [
46
,
47
], whe eas ou amewo k in o-
duces a ede a ed Biobank ne wo k wi h ha monized da a lows o enhance da a ichness
while espec ing p i acy conce ns. Fo use engagemen , exis ing sys ems ypically employ
single-channel ools wi h sus ainabili y limi a ions [
48
,
49
], while ou mul i-channel ap-
p oach h ough heal h apps and se ious games aims o suppo be e long- e m adhe ence.
Rega ding p i acy p o ec ion, adi ional solu ions o en use cen alized s o age wi h con-
en ional anonymiza ion echniques [
47
], bu ou amewo k p oposes decen alized edge
Biobanks wi h ad anced pseudonymiza ion and GAN-based syn he ic da a gene a ion o
s eng hen p i acy while main aining da a u ili y. Finally, cu en s akeholde collabo a ion
shows limi ed in e ac ion be ween heal hca e p o ide s, esea che s, and amilies [
50
,
51
],
which ou amewo k add esses h ough an in eg a ed dashboa d and communi y knowl-
edge hub designed o acili a e knowledge ans e and collabo a i e decision making.
3. Desc ip ion o F amewo k
3.1. Biobank
The Biobank is a co ne s one o he ad ancemen o esea ch on childhood obesi y. I
can suppo imp o ed diagnos ic and he apeu ic app oaches, suppo he de elopmen o
p e en ion s a egies and also help o unde s and he ansi ion om me abolically heal hy
o unheal hy s a es.
The p oposed Biobank a chi ec u e is designed o add ess he challenges aced by
cen alized Biobanks, da a in eg a ion, complexi y o owne ship, and compliance wi h
p i acy egula ions, such as GTPR, o example. The sys em le e ages a ede a ed s uc u e
ha enables decen alized da a managemen while ensu ing accessibili y, in e ope abili y
and secu i y. Each pa icipa ing si e main ains an independen Biobank edge ha ac s as a
local node o da a collec ion and s o age. This app oach elimina es he need o agg ega e
sensi i e heal h da a, he eby main aining da a so e eign y and ensu ing compliance
wi h egional p i acy amewo ks. The Biobank edge in eg a es se e al key componen s
o manage he da a li ecycle. Da a cu a ion ensu es he quali y and ep esen a i eness
o he da ase s, while echniques a e applied o p o ec pa ien p i acy by ans o ming
Elec onics 2025,14, 2053 7 o 23
iden i iable da a in o secu e, non- aceable o ma s. The backend o he Biobank edge is
buil on a high-pe o mance a chi ec u e u ilizing Fas API [
52
] o p ocessing logic and
MongoDB [
53
] o scalable da a s o age, complemen ed by a dedica ed Que y Handle ha
op imizes da a access and e ie al p ocesses.
3.2. Da a Managemen and P ep ocessing
The p oposed a chi ec u e add esses essen ial da a managemen and p ep ocessing
unc ionali ies o suppo scalable, secu e, and p i acy-p ese ing da a p ocessing ac oss
dis ibu ed en i onmen s. Da a ha moniza ion esol es inconsis encies ha a ise om in-
eg a ing di e se da ase s o igina ing om mul iple sou ces [
54
]. Di e ences in o ma s,
s anda ds, and seman ics a e esol ed using s anda dized amewo ks, such as CDISC o
OMOP, e . [
55
] and ad anced on ology-d i en seman ic analysis ools. This alignmen
ensu es ha he esul ing da ase s comply wi h he FAIR p inciples (Findable, Accessible,
In e ope able, and Reusable), enhancing da a usabili y and in e ope abili y [
56
]. Sensi i e
in o ma ion is ans o med o p e en unau ho ised e-iden i ica ion [
57
]. C yp og aphic hash
unc ions, de e minis ic pseudonymisa ion and ad anced app oaches, Me kle ees, secu e
mul i-pa y compu a ions, a e implemen ed o achie e da a pseudonymisa ion [
58
]. Commu-
nica ion wi h he dis ibu ed nodes is achie ed h ough REST ul APIs o e secu e channels,
using SSH-based connec ions o s eamline access while main aining obus secu i y [59].
The a chi ec u e implemen s an au ho iza ion amewo k cen e ed a ound a node
ga eway, which ac s as a secu e in e media y be ween local Biobank nodes and cen alized
se ices. OAu h2-based au hen ica ion h ough a cen al Keycloak se e ensu es ha only
p ope ly au hen ica ed use s can access da a, wi h pe missions managed h ough ole-based
con ols. The node ga eway p o ides applica ion-speci ic and use -le el au ho iza ion,
ensu ing use s can only access pe mi ed da a.
Da a gene a ed by applica ions emains decen alized and is s o ed di ec ly in Biobank
nodes a hei espec i e clinical si es, main aining da a so e eign y by ensu ing sensi i e
in o ma ion s ays wi hin app op ia e ju isdic ions. Each Biobank node inco po a es p o-
cessing pipelines o ha moniza ion, cu a ion, and pseudonymiza ion be o e local s o age
o ansmission.
Beyond SSH connec ions, all da a ansmission be ween componen s u ilizes TLS/SSL
enc yp ion as de ailed in he in eg a ion ma ix. REST APIs inco po a e comp ehensi e
secu i y con ols and a e documen ed using OpenAPI speci ica ions. The pla o m includes
a Dis ibu ed Log Se e and Secu i y Moni o ing Se ice ha ack all da a ans e s and
API eques s o secu i y audi ing.
We also use Gene a i e Ad e sa ial Ne wo ks (GANs) [
60
], o gene a e syn he ic
da ase s ha p ese e he s a is ical p ope ies o eal da a wi hou exposing sensi i e
in o ma ion [
61
]. This educes dependency on eal-wo ld da ase s and suppo ing ex ensi e
esea ch and de elopmen ini ia i es.
3.3. AI Tools
The AI ools in eg a e p edic i e modeling, beha io al da a p ocessing, and pe son-
alized ecommenda ion capabili ies wi hin he sys em a chi ec u e. The sys em ensu es
comp ehensi e acking o use engagemen me ics [
62
], while main aining p i acy s an-
da ds, and analyzes pa e ns o op imize in e en ions. In addi ion, i e alua es indi idual
isk ac o s ela ed o me abolic condi ions h ough models ained on eal- ime and his-
o ical da ase s, enabling he accu a e iden i ica ion o po en ial heal h isks. Ou goal is
o ha e collec ed a b oad da a se o mo e han 10,000 samples om hospi al eco ds and
di ec use in e ac ions in he inal phase o he aining wi h ou heal h and se ious games
app. By p ocessing use -speci ic heal h da a and inco po a ing es ablished guidelines, i
Elec onics 2025,14, 2053 8 o 23
p o ides ailo ed eedback and ac ionable sugges ions, such as die a y adjus men s, physi-
cal ac i i y goals, o adhe ence p o ocols, ha suppo e idence-based decision making o
heal hca e p o essionals and p omo e heal hie beha io s o use s, ul ima ely ensu ing
ha in e en ions emain a ge ed, p ac ical, and pe sonalized o indi idual isk p o iles.
Fo pe o mance e alua ion, we employ a combina ion o me ics including a ea unde
he ROC cu e (AUC), p ecision– ecall cu es, and Cohen’s kappa coe icien , which ep e-
sen he cu en s a e-o - he-a o imbalanced heal hca e da ase s. We u ilize a ans e
lea ning app oach, s a ing wi h a p e- ained ounda ion model ha is hen ine- uned on
ou domain-speci ic da a o maximize bo h e iciency and accu acy. C oss- alida ion will be
implemen ed using a s a i ied 10- old app oach o main ain class dis ibu ion ac oss olds,
wi h addi ional empo al alida ion o accoun o po en ial concep d i in longi udinal
heal h da a. Da a imbalance and e hnic di e si y conside a ions a e add essed h ough
s a egic sampling echniques and demog aphic weigh ing. Speci ically, we will ake in o
accoun a ia ions in disease p e alence ac oss di e en popula ions by implemen ing
adap i e boos ing o unde ep esen ed g oups and ensu ing p opo ional ep esen a ion
ac oss majo e hnic ca ego ies in ou aining da a. All model ou pu is in ended exclusi ely
as a ool o clinicians, who e ain ull au ho i y o inal ecommenda ions and ea men
decisions. In addi ion, all ecommenda ions gene a ed by he AI a e alida ed by heal hca e
p o essionals who e iew sample ecommenda ions and inco po a e a eedback loop whe e
use esul s a e acked o con inuously imp o e he accu acy o he model.
3.4. Use In e ace and Engagemen
The Use In e ace and Engagemen Laye o ms a c i ical on -end componen o
ou p oposed a chi ec u e, ac ing as he p ima y in e ac ion poin o child en, pa en s,
and heal hca e p o ide s.
Fo child en, he pla o m ea u es an in e ac i e mobile applica ion ha p omo es
heal hy li ing h ough gami ied expe iences. This app suppo s he collec ion o key be-
ha io al da a, including physical ac i i y, die a y habi s, and li es yle me ics, encou aging
ac i e pa icipa ion. Elemen s such as pe sonalized challenges, ask-based ewa ds, and i-
sual p og ess acking os e engagemen , ans o ming heal h moni o ing in o an enjoyable
and educa ional expe ience, mo i a ing child en o adop and sus ain heal hy habi s. Fo
ins ance, a 9-yea -old s uden plays Food Ninja on he mobile app, ea ning poin s by
iden i ying ood g oups and lea ning he basics o a balanced die . The app keeps ack o
he ime spen in he app, he sco e achie ed, and he p og ess made.
Pa en s bene i om a dashboa d ha p o ides a clea and comp ehensi e iew o
hei child’s p og ess and heal h engagemen . This dashboa d agg ega e beha io al and
heal h da a, p esen ing ends, me ics, and insigh s in an accessible o ma . Pa en s inpu
daily in o ma ion abou hei child’s ea ing habi s, he ype and du a ion o physical ac i i y,
and ha e access o his o ical cha s and clinical ecommenda ions. The da a is sen o a
clinician, who, wi h he help o AI ools, can p o ide pe sonalized die a y guidelines and
amily ac i i ies ailo ed o hei p e e ences. Pa en s can ecei e ale s, iew pe sonalized
ecommenda ions, and gain a deepe unde s anding o hei child’s heal h s a us while
main aining secu e, ole-based access o ensu e da a p i acy and ele ance.
Heal hca e p o ide s in e ac wi h he pla o m h ough he dashboa d designed o
acili a e clinical decision-making and in e en ion s a egies. This dashboa d in eg a es
ools like isk-assessmen modules, enabling p o ide s also o moni o child en’s heal h
da a, iden i y po en ial isks, and deli e pe sonalized ecommenda ions o heal hy li ing.
Fo example, a pedia ician e iews a pa ien ’s consolida ed heal h da a be o e hei annual
check-up, no ing ends in BMI and ac i i y le els. The dashboa d lags po en ial ea ly
wa ning signs and sugges s e idence-based in e en ion app oaches based on he child’s
Elec onics 2025,14, 2053 9 o 23
speci ic engagemen pa e ns and p e e ences. Visual analy ics powe ed by Ma omo Analy -
ics and P ome heus ans o m complex heal h me ics in o ac ionable insigh s, suppo ing
heal hca e p o essionals in making da a-d i en decisions o imp o e ou comes.
This laye also se es as a p ima y in e ace o da a collec ion and in eg a ion, en-
abling eal- ime inpu o beha io al and heal h me ics h ough mobile and web channels.
Complemen ing he Use In e ace and Engagemen Laye , he communi y ne wo k
and knowledge hub u he enhance knowledge dissemina ion, and collabo a ion among
s akeholde s. The communi y ne wo k ac s as a i ual ecosys em whe e child en, pa en s,
and heal hca e p o ide s connec , sha e expe iences, and wo k owa d common heal h
goals. Child en a e encou aged o engage in pee -d i en ac i i ies, such as communi y
i ness challenges. These ac i i ies p omo e eamwo k and heal hy habi s. Addi ionally,
pa en s can pa icipa e in suppo g oups and educa ional campaigns o sha e insigh s
and s a egies o be e manage hei child’s heal h. Fo heal hca e p o ide s, p o essional
communi ies enable he exchange o bes p ac ices and collabo a i e app oaches o obesi y
p e en ion, while eedback channels p o ide di ec communica ion pa hways wi h amilies
o o e ailo ed ad ice and suppo .
The knowledge hub unc ions as a cen alized eposi o y o esou ces and ools,
empowe ing all s akeholde s wi h e idence-based in o ma ion and ac ionable insigh s.
I p o ides access o educa ional ma e ials, guidelines, clinical da ase s, and he la es
esea ch, which in o m e ec i e in e en ion s a egies.
3.5. Se ious Games
This sec ion ou lines he me hodology and de elopmen o he se ious games p oposed
ha a e designed o suppo he p e en ion and managemen o obesi y among young
popula ions. Games ha e been deployed as mobile apps and web-based applica ions, allowing
o easy access. Table 1p esen s a compa a i e o e iew o key cha ac e is ics in ou se ious
games designed o p e en obesi y. Following his able, we p o ide an analy ical p esen a ion
o each game, ocusing on i s objec i es, gameplay mechanics, and educa ional impac .
Table 1. Compa ison o se ious games o obesi y p e en ion.
Cha ac e is ics Food Ninja Food Quiz Food T easu e Le ’s Mo e
P ima y
Objec i e Food-g oup iden i ica ion
and ca ego iza ion
Heal h and nu i ion
li e acy
Combine physical ac i i y
wi h nu i ion educa ion
Es ablish egula
physical ac i i y habi s
Ta ge Use s 6–12 yea s 8–16 yea s 8–14 yea s 6–14 yea s
Co e Mechanics Tapping/sc olling i ems in
ca ego ies
Mul iple-choice
ques ions wi h aids
AR scanning o hidden
i ems
Guided exe cises and
dance ou ines
Lea ning Focus Food g oups and balanced
die
Meal pa e ns, nu i ion
basics, die a y pa e ns
Nu i ional in o ma ion
abou speci ic oods
Exe cise echniques
and mo emen pa e ns
Social Elemen s Indi idual play Mul iplaye op ion Pa en –child in e ac ion Family pa icipa ion
Physical
Ac i i y None None Mode a e High
Technology Basic ouchsc een Basic de ice Sma phone wi h AR
capabili y
Basic de ice wi h ideo
playback
En i onmen Indoo sc een-based Indoo sc een-based Indoo /ou doo
explo a ion
Indoo o ou doo
space o mo emen
Feedback Immedia e eedback wi h
educa ional messages
Explana ions o
inco ec answe s AR in o ma ion displays Visual guidance and
achie emen acking
P o essional
inpu Nu i ional guidelines E idence-based
ques ions Nu i ional in o ma ion Pedia ic consul a ion
o exe cises
Elec onics 2025,14, 2053 16 o 23
Table 2. Con .
E alua ion C i e ion Ou Fede a ed A chi ec u e T adi ional Cen alized App oaches E idence o Imp o emen
S akeholde
Collabo a ion
In eg a ed dashboa d and
communi y knowledge hub
Limi ed in e ac ion be ween
p o ide s, esea che s, and amilies
Facili a es knowledge ans e
and collabo a i e
decision-making; c ea es
eedback loops be ween
s akeholde s
Adap abili y
Flexible implemen a ion
op ions o a ying esou ce
se ings and cul u al
backg ounds
O en equi es s anda dized
implemen a ion
Componen s can be adop ed
based on local cons ain s;
cul u ally adap able con en
and ecommenda ions
Technical
Implemen a ion
REST ul APIs wi h
s anda dized documen a ion;
modula a chi ec u e
P op ie a y in e aces,
monoli hic sys ems
Easie in eg a ion wi h
exis ing heal hca e sys ems;
s anda ds-based app oach
educes implemen a ion
ba ie s
E hical Conside a ions
Ve i iable pa en al consen
p ocess; age-based access
con ols; op -ou mechanisms
O en limi ed p i acy con ols
Enhanced p o ec ion o
mino s’ da a; clea
go e nance s uc u e o
sensi i e in o ma ion
The e alua ion demons a es se e al key ad an ages o ou p oposed a chi ec u e.
Fi s , he decen alized Biobank edges p o ide enhanced da a p i acy while s ill enabling
comp ehensi e analysis h ough ha monized da a lows. This di ec ly add esses limi a ions
o cen alized app oaches [
46
,
47
], ha s uggle o balance da a u ili y wi h p i acy p o ec-
ion. Second, ou mul i-channel use engagemen s a egy le e ages bo h heal h applica-
ions and se ious games o c ea e sus ainable beha io change, o e coming he engagemen
limi a ions obse ed in single-channel in e en ions [
48
,
49
]. Thi d, he amewo k acili-
a es imp o ed s akeholde collabo a ion h ough in eg a ed dashboa ds and knowledge
sha ing pla o ms, enhancing communica ion be ween heal hca e p o ide s, esea che s,
and amilies compa ed o adi ional siloed app oaches [
50
,
51
]. Finally, he modula na u e
o ou a chi ec u e enables con ex ual adap a ion ac oss di e se implemen a ion se ings,
allowing componen s o be adop ed based on local esou ces, cul u al conside a ions,
and echnical capabili ies.
Mapping A chi ec u al Componen s o Requi emen s
Ou a chi ec u e ul ills c i ical equi emen s h ough a ge ed design choices (
Table 3
).
The decen alized Biobank s uc u e wi h independen edges di ec ly add esses p i acy
conce ns by ensu ing sensi i e heal h da a emains wi hin app op ia e ju isdic ions while
s ill enabling secu e analy ics. The node ga eway componen unc ions as a secu e in e -
media y ha enables con olled access o da a h ough applica ion-speci ic and use -le el
au ho iza ion p o ocols, ensu ing use s can only access pe mi ed da a. To acili a e com-
p ehensi e da a collec ion while main aining use engagemen , we implemen ed a dual
app oach h ough he heal h applica ion and se ious games sui e. These componen s
wo k syne gis ically— he heal h applica ion cap u es s uc u ed heal h me ics while he
se ious games p o ide an engaging con ex o beha io modi ica ion and addi ional da a
collec ion. The dashboa d’s isualiza ion capabili ies ans o m complex heal h me ics
in o in ui i e insigh s, allowing pa en s and heal hca e p o ide s o iden i y ends and
make e idence-based decisions.
Elec onics 2025,14, 2053 17 o 23
Table 3. A chi ec u e componen s ul illing obesi y managemen equi emen s.
Requi emen A chi ec u al Componen Enabling Fea u es
Da a P i acy P o ec ion Fede a ed Biobank Edges
• Decen alized s o age keeping da a a o igina ing si es
• Ad anced pseudonymiza ion p ocessing
• GAN-based syn he ic da a gene a ion
Comp ehensi e Da a
In eg a ion Da a Ha moniza ion Laye
• S anda dized amewo ks (CDISC/OMOP)
• On ology-d i en seman ic analysis
• REST ul APIs o e secu e channels
Sus ainable Use
Engagemen Mul i-channel Use In e ace
• Mobile heal h applica ion
• Se ious games sui e
• Age-app op ia e gami ica ion elemen s
E idence-based In e en ion
AI Tools and Knowledge Hub
• Risk-assessmen models
• Pe sonalized ecommenda ion engine
• Knowledge eposi o y o heal hca e p o essionals
Mul i-s akeholde
Collabo a ion
Communi y Ne wo k and
Dashboa d
• Role-based in e aces o child en, pa en s, and clinicians
• Sha ed isualiza ion o p og ess
• Collabo a i e decision-making ools
4. Sys em In e ac ion Wo k lows
This sec ion p esen s sequence diag ams depic ing he c i ical da a lows and use
in e ac ions wi hin ou Biobank a chi ec u e.
The sequence diag am in Figu e 6illus a es da a analy ics a chi ec u e in which mul-
iple applica ions sha e anonymous use IDs wi hin a uni ied sys em. The i e applica ions
(Ac i eHeal hApp and ou se ious games) a e connec ed o a cen alized Ma omo Analy -
ics pla o m. The Ac i eHeal hApp uses se e -side da a cap u e o ansmi in o ma ion
o he analy ics pla o m, while he ou se ious games communica e ia REST API. This en-
ables consolida ed acking and analysis o use beha io ac oss all applica ions. The sys em
is designed o collec and analyze use engagemen da a in a p i acy-conscious manne .
Figu e 6. Analy ics se ice in e nal a chi ec u e.
Elec onics 2025,14, 2053 18 o 23
Use in e ac ions wi h he knowledge hub a e ep esen ed in Figu e 7, which shows he
communica ion low be ween use s, he Ac i eHeal h App, and he knowledge hub. When
a use s a s he app, one o wo pa hs occu s based on au hen ica ion s a us: au hen ica ed
use s na iga e o a home page and choose a websi e om na iga ion op ions, while
unau hen ica ed use s iew a welcome page wi h a slide om which hey selec a websi e.
In bo h scena ios, a e he use makes hei selec ion, he Ac i eHeal h App launches
he chosen websi e by communica ing wi h he knowledge hub, which hen displays he
ele an websi e con en back o he use h ough he app. This s eamlined p ocess ensu es
all use s can access app op ia e con en ega dless o hei au hen ica ion s a us, wi h he
knowledge hub se ing as he cen al con en eposi o y ha deli e s in o ma ion back
h ough he Ac i eHeal h App in e ace.
Figu e 7. a sequence diag am desc ibing he use in e ac ion wi h THE knowledge hub, communi y
ne wo k, and ma ke place.
As shown in Figu e 8, he se ious game sys em ollows a s uc u ed in e ac ion pa h-
way beginning wi h use au hen ica ion. The sequence s a s wi h use au hen ica ion
h ough he AUTH se ice o access he dashboa d. Once au hen ica ed, use s can in e ac
wi h he game h ough s a and esume ac ions handled by he SERG (se ious game) com-
ponen . Du ing gameplay, use in e ac ion da a a e empo a ily s o ed on he use ’s de ice.
Upon game comple ion, a e p ocessing, he collec ed da a a e ansmi ed o he Biobank
edge h ough he node ga eway (NGW) se ice. When he game da a each he node
bundle, i unde goes p ocessing h ough ha moniza ion, cu a ion, and pseudonymiza ion
componen s be o e being secu ely s o ed. Fo AI model p ocessing, au ho ized heal hca e
p o essionals can access hese da a h ough he dashboa d, using secu ed REST APIs. This
allows AI componen s o analyze he game da a when igge ed by an au ho ized use
h ough he dashboa d in e ace. The en i e da a low is p o ec ed by OAu h2 au hen-
ica ion ia he cen al Keycloak se e , ensu ing ha only p ope ly au hen ica ed and
au ho ized use s can access sensi i e in o ma ion h oughou he sys em.
Elec onics 2025,14, 2053 19 o 23
Figu e 8. A sequence diag am desc ibing he use in e ac ion wi h he se ious game.
5. Conclusions and Fu u e Wo k
The challenge o childhood obesi y equi es inno a i e app oaches ha ake ad an-
age o echnology while main aining p i acy, secu i y, and engagemen . Ou p oposed
ede a ed and decen alised amewo k ep esen s a comp ehensi e solu ion o his com-
plex public heal h issue. By in eg a ing independen Biobank edges o localized da a
collec ion wi h edges in schools and communi ies, we c ea e a obus da a ecosys em ha
espec s p i acy conce ns h ough ca e ul da a ha moniza ion, pseudonymisa ion, and syn-
he ic da a gene a ion ia GANs. The s eng h o he amewo k lies in i s mul i ace ed
app oach, combining AI-powe ed analy ics o pe sonalized isk assessmen wi h use -
iendly dashboa ds ha enable e idence-based decision making by pa en s and heal hca e
p o ide s. The inco po a ion o se ious games and heal h applica ions engages child en
di ec ly, encou aging heal hie li es yle choices h ough in e ac i e expe iences ha a ge
bo h p o ec i e ac o s and isk ac o s. Fu he mo e, he communi y ne wo k and knowl-
edge hub os e collabo a ion and knowledge sha ing among s akeholde s. Economic
sus ainabili y emains cen al o ou app oach, wi h cos assessmen ools ensu ing he
iabili y o implemen a ion ac oss di e se se ings.
As a c i ical nex phase in ou esea ch, we plan o subjec he en i e sys em o a
comp ehensi e e hical e iew h ough an Ins i u ional Re iew Boa d (IRB). This o mal
e alua ion will ensu e ha all aspec s o da a collec ion, s o age, p ocessing, and usage
comply wi h e hical s anda ds and egula o y equi emen s.
We plan o implemen a comp ehensi e alida ion s a egy h ough wo kshops and
pilo s. As a i s s ep, we will conduc a se ies o wo kshops wi h a leas 200 pa ien s
and s uden s o e ine he games based on eal-wo ld needs and con ex ual ac o s. These
Elec onics 2025,14, 2053 20 o 23
wo kshops will in o m he inal p o ocol design by inco po a ing cul u al conside a ions
and esea ch- ela ed ac o s. This pa icipa o y app oach will ensu e ha he p oposed
a chi ec u e mee s eal use needs be o e o mal es ing.
The nex s ep will be o implemen a mixed-me hods alida ion s udy o e one school
yea wi h a p e- es /pos - es expe imen al design compa ing con ol and in e en ion
g oups. The in e en ion will span 6 mon hs and include ou join pa en –child class oom
wo kshops ocusing on heal hy ea ing and physical ac i i y o p omo e p e en i e beha io .
Key ou come measu es will include changes in nu i ion knowledge sco es, daily physical
ac i i y le els measu ed ia alida ed ins umen s, ood-choice beha io s in con olled
se ings, and use engagemen me ics. The pilo s will be conduc ed in collabo a ion wi h
es ablished obesi y ca e cen e s whe e mul idisciplina y eams including child psychia is s,
psychologis s, endoc inologis s, diabe ologis s, su geons, and nu i ionis s will suppo
he implemen a ion and assessmen . These pilo s aim o demons a e he impac o he
isk assessmen amewo k on clinical ou ine and guidelines, he e ec i eness o he
ecommenda ion sys em on pa ien empowe men and engagemen , and he po en ial o
unde s anding how mul iple ac o s (gene ic, epigene ic, en i onmen al, socioeconomic,
and li es yle) in e ac in he de elopmen o childhood obesi y.
Au ho Con ibu ions: Concep ualiza ion, A.A. and M.K.; Me hodology, T.B. and M.G.; So wa e,
M.L. and D.G.; Valida ion, M.P.; Fo mal analysis, E.P. and G.S.; In es iga ion, M.K.; Resou ces, G.M.;
Da a cu a ion, H.K.; W i ing—o iginal d a , I.V.; Supe ision, G.D., G.M. and D.K.; P ojec adminis-
a ion, E.V. and I.K. All au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding: This pape has been conduc ed wi hin he BIO-STREAMS p ojec , which has ecei ed
unding om he Eu opean Union’s HORIZON 2022 esea ch and inno a ion p og am unde g an
ag eemen No. 101080718.
Da a A ailabili y S a emen : The o iginal con ibu ions p esen ed in his s udy a e included in he
a icle. Fu he inqui ies can be di ec ed o he co esponding au ho .
Con lic s o In e es : Au ho s Theodo a B isimi and Ma ios Logo he is we e employed by he
company Ne company-In aso SA. Au ho s Ha y Kakoulidis and Ma ios P asinos we e employed
by he company Telema ic Medical Applica ions L d. The emaining au ho s decla e ha he esea ch
was conduc ed in he absence o any comme cial o inancial ela ionships ha could be cons ued as
a po en ial con lic o in e es .
Re e ences
1.
Ma cus, C.; Danielsson, P.; Hagman, E. Pedia ic obesi y—Long- e m consequences and e ec o weigh loss. J. In e n. Med. 2022,
292, 870–891. [C ossRe ] [PubMed]
2.
Ka sa o a, I. Sky ocke ing Obesi y in Child en: Why E e ybody Should Be Conce ned? Technical Repo ; Eu opean Pa liamen a y
Resea ch Se ice: B ussels, Belgium, 2024.
3.
Tsoi, M.F.; Li, H.L.; Feng, Q.; Cheung, C.L.; Cheung, T.T.; Cheung, B.M. P e alence o childhood obesi y in he Uni ed S a es in
1999–2018: A 20-yea analysis. Obes. Fac s 2022,15, 560–569. [C ossRe ]
4. Masood, B.; Moo hy, M. Causes o obesi y: A e iew. Clin. Med. 2023,23, 284–291. [C ossRe ] [PubMed]
5.
LeC oy, M.N.; Kim, R.S.; S e ens, J.; Hanna, D.B.; Isasi, C.R. Iden i ying key de e minan s o childhood obesi y: A na a i e
e iew o machine lea ning s udies. Child. Obes. 2021,17, 153–159. [C ossRe ]
6.
Hea h, L.; Jebb, S.A.; A eya d, P.; Pie nas, C. Obesi y, me abolic isk and adhe ence o heal hy li es yle beha iou s: P ospec i e
coho s udy in he UK Biobank. BMC Med. 2022,20, 65. [C ossRe ]
7.
Lamas, S.; Rebelo, S.; da Cos a, S.; Sousa, H.; Zagalo, N.; Pin o, E. The in luence o se ious games in he p omo ion o heal hy die
and physical ac i i y heal h: A sys ema ic e iew. Nu ien s 2023,15, 1399. [C ossRe ]
8.
Vou doumpa, A.; Pal oglou, G.; Cha manda i, E. The gene ic basis o childhood obesi y: A sys ema ic e iew. Nu ien s 2023,
15, 1416. [C ossRe ] [PubMed]
9.
Wójcik, M.; Al a ez-Pi i, J.; Kozioł-Kozakowska, A.; B zezi´nski, M.; Gabbianelli, R.; He ceg- ˇ
Ca ak, V.; Wühl, E.; Lucas, I.;
Rado ano i´c, D.; Melk, A.; e al. Psychosocial and en i onmen al isk ac o s o obesi y and hype ension in child en and
adolescen s—A li e a u e o e iew. F on . Ca dio asc. Med. 2023,10, 1268364. [C ossRe ] [PubMed]
Elec onics 2025,14, 2053 21 o 23
10.
Alba aineh, S.R.; Bad an, E.F.; Tayyem, R.F. O e weigh and obesi y in childhood: Die a y, biochemical, in lamma o y and
li es yle isk ac o s. Obes. Med. 2019,15, 100112. [C ossRe ]
11.
Hall, K.D.; Heyms ield, S.B.; Kemni z, J.W.; Klein, S.; Schoelle , D.A.; Speakman, J.R. Ene gy balance and i s componen s:
Implica ions o body weigh egula ion. Am. J. Clin. Nu . 2012,95, 989–994. [C ossRe ]
12.
Jakobsen, D.D.; B ade , L.; B uun, J.M. Associa ion be ween ood, be e ages and o e weigh /obesi y in child en and
adolescen s—A sys ema ic e iew and me a-analysis o obse a ional s udies. Nu ien s 2023,15, 764. [C ossRe ] [PubMed]
13.
Libe ali, R.; Kupek, E.; Assis, M.A.A.d. Die a y pa e ns and childhood obesi y isk: A sys ema ic e iew. Child. Obes. 2020,16, 70–85.
[C ossRe ] [PubMed]
14.
Dallacke , M.; He wig, R.; Ma a, J. The equency o amily meals and nu i ional heal h in child en: A me a-analysis. Obes. Re .
2018,19, 638–653. [C ossRe ] [PubMed]
15.
Zheng, X.; Wang, H.; Wu, H. Associa ion be ween die quali y sco es and isk o o e weigh and obesi y in child en and
adolescen s. BMC Pedia . 2023,23, 169. [C ossRe ]
16.
Sal aou as, G.; Ky kili, A.; Ba h ellou, E.; Geo goulis, M.; Yannakoulia, M.; Boun ziouka, V.; Sm ke, U.; Dimi akopoulos, G.;
Kon ogianni, M.D. Associa ions be ween Meal Pa e ns and Risk o O e weigh /Obesi y in Child en and Adolescen s in Wes e n
Coun ies: A Sys ema ic Re iew o Longi udinal S udies and Randomised Con olled T ials. Child en 2024,11, 1100. [C ossRe ]
17.
Hampl, S.E.; Hassink, S.G.; Skinne , A.C.; A ms ong, S.C.; Ba low, S.E.; Bolling, C.F.; A ila Edwa ds, K.C.; Eneli, I.; Ham e, R.;
Joseph, M.M.; e al. Execu i e summa y: Clinical p ac ice guideline o he e alua ion and ea men o child en and adolescen s
wi h obesi y. Pedia ics 2023,151. [C ossRe ]
18.
DiPie o, L.; Buchne , D.M.; Ma quez, D.X.; Pa e, R.R.; Pesca ello, L.S.; Whi -Glo e , M.C. New scien i ic basis o he 2018 US
Physical Ac i i y Guidelines. J. Spo Heal h Sci. 2019,8, 197. [C ossRe ]
19. Bou die , P.; Simon, C.; Bessesen, D.H.; Blanc, S.; Be gouignan, A. The ole o physical ac i i y in he egula ion o body weigh :
The o e looked con ibu ion o ligh physical ac i i y and seden a y beha io s. Obes. Re . 2023,24, e13528. [C ossRe ]
20.
Biddle, S.J.H.; Ga cía Bengoechea, E.; Wiesne , G. Seden a y beha iou and adiposi y in you h: A sys ema ic e iew o e iews
and analysis o causali y. In . J. Beha . Nu . Phys. Ac . 2017,14, 43. [C ossRe ]
21.
Ga cía-He moso, A.; Saa ed a, J.M.; Ramí ez-Vélez, R.; Ekelund, U.; Del Pozo-C uz, B. Realloca ing seden a y ime o mode a e-
o- igo ous physical ac i i y bu no o ligh -in ensi y physical ac i i y is e ec i e o educe adiposi y among you hs: A sys ema ic
e iew and me a-analysis. Obes. Re . 2017,18, 1088–1095. [C ossRe ]
22.
Kobes, A.; K e schme , T.; Timme man, G.; Sch eude , P. In e en ions aimed a p e en ing and educing o e weigh /obesi y
among child en and adolescen s: A me a-syn hesis. Obes. Re . 2018,19, 1065–1079. [C ossRe ]
23.
Deno a-Gu ie ez, E.; Gonzalez-Rocha, A.; Mendez-Sanchez, L.; A aiza-Na a, B.; Balde as, N.; Lopez, G.; Tolen ino-Mayo, L.;
Jau egui, A.; He nandez, L.; Unikel, C.; e al. O e iew o sys ema ic e iews o heal h in e en ions o he p e en ion and
ea men o o e weigh and obesi y in child en. Nu ien s 2023,15, 773. [C ossRe ] [PubMed]
24.
Podna , H.; Ju i´c, P.; Ka uc, J.; Saez, M.; Ba celó, M.A.; Radman, I.; S a c, G.; Ju ak, G.; Ðu i´c, S.; Po oˇcnik, Ž.L.; e al. Compa a i e
e ec i eness o school-based in e en ions a ge ing physical ac i i y, physical i ness o seden a y beha iou on obesi y
p e en ion in 6- o 12-yea -old child en: A sys ema ic e iew and me a-analysis. Obes. Re . 2021,22, e13160. [C ossRe ] [PubMed]
25.
Chapu , J.P.; Willumsen, J.; Bull, F.; Chou, R.; Ekelund, U.; Fi h, J.; Jago, R.; O ega, F.B.; Ka zma zyk, P.T. 2020 WHO guidelines
on physical ac i i y and seden a y beha iou o child en and adolescen s aged 5–17 yea s: Summa y o he e idence. In . J.
Beha . Nu . Phys. Ac . 2020,17, 141. [C ossRe ]
26.
Wyszy´nska, J.; Ring-Dimi iou, S.; Thi el, D.; Weghube , D.; Hadjipanayis, A.; G ossman, Z.; Ross-Russell, R.; De e´n, K.; Mazu ,
A. Physical ac i i y in he p e en ion o childhood obesi y: The posi ion o he Eu opean Childhood Obesi y G oup and he
Eu opean Academy o Pedia ics. F on . Pedia . 2020,8, 535705. [C ossRe ]
27.
Gu hold, R.; S e ens, G.A.; Riley, L.M.; Bull, F.C. Global ends in insu icien physical ac i i y among adolescen s: A pooled
analysis o 298 popula ion-based su eys wi h 1· 6 million pa icipan s. Lance Child Adolesc. Heal h 2020,4, 23–35. [C ossRe ]
[PubMed]
28.
Campbell, M.K. Biological, en i onmen al, and social in luences on childhood obesi y. Pedia . Res. 2016,79, 205–211. [C ossRe ]
29.
Alghalyini, B. Applica ions o a i icial in elligence in he managemen o childhood obesi y. J. Fam. Med. P im. Ca e 2023,12,
2558–2564. [C ossRe ]
30.
Singh, B.; Go benko, A.; Palczewska, A.; Taw ik, H. Applica ion o Machine Lea ning Techniques o P edic Teenage Obesi y
Using Ea lie Childhood Measu emen s om Millennium Coho S udy. In P oceedings o he 2023 7 h In e na ional Con e ence
on Cloud and Big Da a Compu ing, Chongqing, China, 2–5 June 2023; pp. 55–60.
31.
Dugan, T.M.; Mukhopadhyay, S.; Ca oll, A.; Downs, S. Machine lea ning echniques o p edic ion o ea ly childhood obesi y.
Appl. Clin. In o m. 2015,6, 506–520.
32.
Bays, H.E.; Fi ch, A.; Cuda, S.; Gonsahn-Bollie, S.; Rickey, E.; Hablu zel, J.; Coy, R.; Censani, M. A i icial in elligence and obesi y
managemen : An obesi y medicine associa ion (OMA) clinical p ac ice s a emen (CPS) 2023. Obes. Pilla s 2023,6, 100065.
[C ossRe ]
Elec onics 2025,14, 2053 22 o 23
33.
Bond, A.; Mccay, K.; Lal, S. A i icial in elligence & clinical nu i ion: Wha he u u e migh ha e in s o e. Clin. Nu . ESPEN
2023,57, 542–549.
34.
Guillaudeux, M.; Rousseau, O.; Pe o , J.; Bennis, Z.; Dein, C.A.; Go on lo , T.; Vince, N.; Limou, S.; Ka akacho , M.;
Wa gny, M.; e al.
Pa ien -cen ic syn he ic da a gene a ion, no eason o isk e-iden i ica ion in biomedical da a analysis.
NPJ Digi . Med. 2023,6, 37. [C ossRe ] [PubMed]
35.
Mendes, J.M.; Ba ba , A.; Re aie, M. Syn he ic da a gene a ion: A p i acy-p ese ing app oach o accele a e a e disease esea ch.
F on . Digi . Heal h 2025,7, 1563991. [C ossRe ] [PubMed]
36.
Qian, Z.; Callende , T.; Cebe e, B.; Janes, S.M.; Na ani, N.; an de Schaa , M. Syn he ic da a o p i acy-p ese ing clinical isk
p edic ion. Sci. Rep. 2024,14, 25676. [C ossRe ]
37. Hewi , R.; Wa son, P. De ining biobank. Biop ese . Biobank. 2013,11, 309–315. [C ossRe ] [PubMed]
38.
Kinko o á, J.; Topolˇcan, O. Biobanks in Ho izon 2020: Sus ainabili y and a ac i e pe spec i es. Epma J. 2018,9, 345–353.
[C ossRe ]
39.
Laugesen, K.; Mengel-F om, J.; Ch is ensen, K.; Olsen, J.; Hougaa d, D.M.; Boding, L.; Olsen, A.; E iks up, C.; He land, M.L.;
Høgdall, E.; e al.
A e iew o majo danish biobanks: Ad an ages and possibili ies o heal h esea ch in Denma k. Clin. Epidemiol.
2023,15, 213–239. [C ossRe ]
40.
Schü le , C.; P okosch, H.U.; Hummel, M.; Lablans, M.; K oll, B.; Engels, C.; de elopmen eam, G.B.A.I. The jou ney o
es ablishing an IT-in as uc u e wi hin he Ge man Biobank Alliance. PLoS ONE 2021,16, e0257632. [C ossRe ]
41.
an’ Rie , E.; Sch am, M.T.; Abbink, E.J.; Admi aal, W.M.; Dijk-Schaap, M.W.; Holleman, F.; Nijpels, G.; Özcan, B.; Pijl, H.;
Schape , N.C.; e al. The diabe es pea l: Diabe es biobanking in The Ne he lands. BMC Public Heal h 2012,12, 949. [C ossRe ]
42.
Ame youn, A.; Sanaeinasab, H.; Sa a i, M.; Koenig, H.G. Impac o game-based heal h p omo ion p og ams on body mass index
in o e weigh /obese child en and adolescen s: A sys ema ic e iew and me a-analysis o andomized con olled ials. Child.
Obes. 2018,14, 67–80. [C ossRe ]
43.
Dias, J.D.; Domingues, A.N.; Tibes, C.M.; Zem-Masca enhas, S.H.; Fonseca, L.M.M. Se ious games as an educa ional s a egy o
con ol childhood obesi y: A sys ema ic li e a u e e iew. Re . La .-Am. En e m. 2018,26, e3036. [C ossRe ] [PubMed]
44.
Belghali, M.; S a senko, Y.; Al-Za’abi, A. Imp o ing se ious games o ackle childhood obesi y. F on . Psychol. 2021,12, 657289.
[C ossRe ]
45.
Mack, I.; Baye , C.; Schae ele , N.; Reiband, N.; B oelz, E.; Zu s iege, G.; Fe nandez-A anda, F.; Gaw ilow, C.; Zip el, S. Chances
and limi a ions o ideo games in he igh agains childhood obesi y—A sys ema ic e iew. Eu . Ea . Diso d. Re . 2017,25, 237–267.
[C ossRe ] [PubMed]
46.
Asiimwe, R.; Lam, S.; Leung, S.; Wang, S.; Wan, R.; Tinke , A.; McAlpine, J.N.; Woo, M.M.; Hun sman, D.G.; Talhouk, A. F om
biobank and da a silos in o a da a commons: Con e gence o suppo ansla ional medicine. J. T ansl. Med. 2021,19, 493.
[C ossRe ]
47.
Alkha ib, R.; Gaede, K.I. Da a managemen in biobanking: S a egies, challenges, and u u e di ec ions. BioTech 2024,13, 34.
[C ossRe ]
48.
Mu phy, M.; Ga e , S.B.; Boyd, E.; D y, S.; Dohan, D. Engaging di e se s akeholde s o in o m biobank go e nance. Biop ese .
Biobank. 2017,15, 393–395. [C ossRe ] [PubMed]
49.
Mi chell, D.; Geissle , J.; Pa y-Jones, A.; Keulen, H.; Schmi , D.C.; Va asso i, R.; Ma ha oo-Ball, B. Biobanking om he pa ien
pe spec i e. Res. In ol . Engagem. 2015,1, 4. [C ossRe ]
50.
S aun on, C.; Tindana, P.; Hend icks, M.; Moodley, K. Rules o engagemen : Pe spec i es on s akeholde engagemen o genomic
biobanking esea ch in Sou h A ica. BMC Med. E hics 2018,19, 13. [C ossRe ]
51.
D y, S.M.; Ga e , S.B.; Koenig, B.A.; B own, A.F.; Bu gess, M.M.; Hul , J.R.; Longs a , H.; Wilcox, E.S.; Mad igal Con e as, S.K.;
Ma inez, A.; e al. Communi y ecommenda ions on biobank go e nance: Resul s om a delibe a i e communi y engagemen
in Cali o nia. PLoS ONE 2017,12, e0172582. [C ossRe ]
52.
Bansal, P.; Ouda, A. S udy on in eg a ion o Fas API and machine lea ning o con inuous au hen ica ion o beha io al biome ics.
In P oceedings o he 2022 In e na ional Symposium on Ne wo ks, Compu e s and Communica ions (ISNCC), Shenzhen, China,
19–22 July 2022; pp. 1–6.
53.
B adshaw, S.; B azil, E.; Chodo ow, K. MongoDB: The De ini i e Guide: Powe ul and Scalable Da a S o age; O’Reilly Media:
Sebas opol, CA, USA, 2019.
54.
Nan, Y.; Del Se , J.; Walsh, S.; Schönlieb, C.; Robe s, M.; Selby, I.; Howa d, K.; Owen, J.; Ne ille, J.; Guio , J.; e al. Da a
ha monisa ion o in o ma ion usion in digi al heal hca e: A s a e-o - he-a sys ema ic e iew, me a-analysis and u u e esea ch
di ec ions. In . Fusion 2022,82, 99–122. [C ossRe ]
55.
Hu s edle , H.; Roell, Y.; Peña, A.; K ishnan, A.; G een, I.; Ba bosa-Sil a, A.; K eme , A.; Blacke e , C.; Fo ie , I.; Howa d, K.;
e al. Na iga ing da a s anda ds in public heal h: A b ie epo om a da a-s anda ds mee ing. J. Glob. Heal h 2024,14, 03024.
[C ossRe ] [PubMed]
Elec onics 2025,14, 2053 23 o 23
56.
Inau, E.T.; Sack, J.; Wal ema h, D.; Zeleke, A.A. Ini ia i es, concep s, and implemen a ion p ac ices o he indable, accessible,
in e ope able, and eusable da a p inciples in heal h da a s ewa dship: Scoping e iew. J. Med. In e ne Res. 2023,25, e45013.
[C ossRe ]
57.
Rak, R. Anonymisa ion, Pseudonymisa ion and Secu e P ocessing En i onmen s Rela ing o he Seconda y Use o Elec onic
Heal h Da a in he Eu opean Heal h Da a Space (EHDS). Eu . J. Risk Regul. 2024,15, 928–938. [C ossRe ]
58.
Sepas, A.; Bangash, A.H.; Al aoui, O.; El Emam, K.; El-Hussuna, A. Algo i hms o anonymize s uc u ed medical and heal hca e
da a: A sys ema ic e iew. F on . Bioin o m. 2022,2, 984807. [C ossRe ] [PubMed]
59.
Ehsan, A.; Abuhaliqa, M.A.M.; Ca al, C.; Mish a, D. REST ul API es ing me hodologies: Ra ionale, challenges, and solu ion
di ec ions. Appl. Sci. 2022,12, 4369. [C ossRe ]
60.
C eswell, A.; Whi e, T.; Dumoulin, V.; A ulkuma an, K.; Sengup a, B.; Bha a h, A.A. Gene a i e ad e sa ial ne wo ks:
An o e iew. IEEE Signal P ocess. Mag. 2018,35, 53–65. [C ossRe ]
61.
Figuei a, A.; Vaz, B. Su ey on syn he ic da a gene a ion, e alua ion me hods and GANs. Ma hema ics 2022,10, 2733. [C ossRe ]
62.
Gamalielsson, J.; Lundell, B.; Bu le , S.; B ax, C.; Pe sson, T.; Ma sson, A.; Gus a sson, T.; Feis , J.; Lön o h, E. Towa ds open
go e nmen h ough open sou ce so wa e o web analy ics: The case o Ma omo. JeDEM-eJou nal eDemoc acy Open Go . 2021,
13, 133–153. [C ossRe ]
63.
Johannesson, P.; Pe jons, E. E alua e A e ac . In An In oduc ion o Design Science; Sp inge In e na ional Publishing: Cham,
Swi e land, 2021; pp. 141–152.
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .
