Co esponding au ho : Anil Kuma
Copy igh © 2025 Au ho (s) e ain he copy igh o his a icle. This a icle is published unde he e ms o he C ea i e Commons A ibu ion Liscense 4.0.
Pos -quan um c yp og aphy combined wi h neu o-symbolic AI o sa egua d sensi i e
psychia ic he apy models agains u u e cybe h ea s
Anil Kuma *
Depa men o Compu e Science, Maha ishi In e na ional Uni e si y, USA.
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
Publica ion his o y: Recei ed on 21 Augus 2025; e ised on 01 Sep embe 2025; accep ed on 09 Oc obe 2025
A icle DOI: h ps://doi.o g/10.30574/gscbps.2025.33.1.0379
Abs ac
The p o ec ion o psychia ic he apy models p esen s a c i ical challenge as heal hca e inc easingly adop s digi al
pla o ms o diagnosis, ea men pe sonaliza ion, and p edic i e analy ics. These models con ain highly sensi i e
pa ien da a and he apeu ic s a egies ha mus emain secu e agains bo h cu en and u u e cybe h ea s.
T adi ional c yp og aphic sa egua ds, while e ec i e oday, a e ulne able o he eme ging capabili ies o quan um
compu ing, which h ea ens o unde mine widely deployed enc yp ion algo i hms. F om a b oade pe spec i e, pos -
quan um c yp og aphy o e s ma hema ically esilien enc yp ion schemes ha a e designed o wi hs and a acks om
quan um ad e sa ies. Simul aneously, neu o-symbolic a i icial in elligence (AI) combines he adap i e powe o neu al
ne wo ks wi h he in e p e abili y and easoning s eng hs o symbolic sys ems, enabling in elligen de ec ion o
anomalous beha io s and policy-d i en access con ol. Na owing he ocus, his s udy explo es a secu i y a chi ec u e
ha in eg a es pos -quan um c yp og aphic algo i hms wi h neu o-symbolic AI o p o ec psychia ic he apy models.
The p oposed amewo k uses la ice-based c yp og aphy o da a con iden iali y, signa u e schemes o secu e model
p o enance, and hyb id AI sys ems o con inuous moni o ing o access and h ea pa e ns. By embedding easoning-
d i en sa egua ds alongside machine lea ning de ec ion, he amewo k o e s bo h p oac i e and explainable de ense
mechanisms. Fu he mo e, compliance wi h p i acy egula ions is suppo ed h ough au oma ed e i ica ion o access
policies and anspa en audi ails. This dual app oach no only shields he apy models om quan um-e a dec yp ion
isks bu also ensu es e hical and us wo hy deploymen in sensi i e men al heal h domains. Ul ima ely, sa egua ding
psychia ic he apy models h ough pos -quan um and neu o-symbolic me hods con ibu es o u u e- eady heal hca e
sys ems ha uphold bo h da a in eg i y and pa ien us .
Keywo ds: Pos -Quan um C yp og aphy; Neu o-Symbolic AI; Psychia ic The apy Models; Cybe secu i y In
Heal hca e; Quan um-Resilien Enc yp ion; P i acy P o ec ion
1. In oduc ion
1.1. Con ex : Rising eliance on psychia ic he apy models in digi al heal hca e
The las decade has wi nessed a signi ican ise in digi al heal hca e pla o ms ha in eg a e psychia ic he apy models
o expand access, educe s igma, and suppo pa ien s ou side con en ional clinics [1]. This e olu ion is p opelled by
he g owing bu den o men al heal h diso de s, pa icula ly dep ession and anxie y, which emain leading con ibu o s
o global disabili y [2]. Wi hin he Uni ed S a es, elepsychia y has seen accele a ed adop ion ollowing ede al
encou agemen o eleheal h eimbu semen pa i y, ensu ing ha psychia ic he apy models a e no longe con ined
o me opoli an cen e s [3].
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
2
Cen al o his expansion is he embedding o cogni i e beha io al he apy and dialec ical beha io al he apy in o
mobile applica ions and AI-assis ed pla o ms. Such ools enable he apis s o scale hei each while p o iding pa ien s
wi h consis en , s uc u ed in e en ions. The e ec i eness o hese digi al psychia ic models has been demons a ed
in andomized clinical ials, showing ou comes compa able o in-pe son sessions [4].
Howe e , eliance on digi al psychia ic he apy in oduces challenges in sa egua ding sensi i e heal h da a. As
p o ide s in eg a e AI-d i en sys ems in o ca e deli e y, he impe a i e o add ess p i acy, da a so e eign y, and
e hical use becomes p essing [2]. Figu e 1 highligh s g ow h p ojec ions o digi al psychia ic he apy adop ion, while
Table 1 compa es cos -e ec i eness be ween con en ional and digi al app oaches.
1.2. Cybe secu i y isks in psychia ic AI sys ems
While digi al psychia y enhances accessibili y, i c ea es ulne abili ies when in eg a ed wi h AI-based diagnos ic and
he apeu ic models [6]. Psychia ic da a a e pa icula ly sensi i e, and b eaches can lead no only o inancial ha m bu
also o signi ican epu a ional damage and s igma o a ec ed indi iduals [5]. Unlike con en ional heal h sys ems,
psychia ic AI pla o ms o en ely on eal- ime moni o ing, con e sa ional agen s, and emo ion-de ec ion algo i hms,
which in oduce unique a ack su aces.
Cybe c iminals inc easingly a ge heal hca e sys ems because o he high alue o medical eco ds on unde g ound
ma ke s. In psychia ic con ex s, ad e sa ial machine lea ning a acks may co up he apy cha bo s, causing ha m ul
o misleading ad ice [6]. Fu he mo e, quan um compu ing, hough no ye ully ealized, poses a looming h ea by
unde mining widely used c yp og aphic p o ocols ha cu en ly secu e pa ien -p o ide communica ion [7].
Regula o y amewo ks like HIPAA p o ide pa ial sa egua ds, bu hey we e no designed o AI-d i en psychia ic
applica ions ope a ing ac oss cloud in as uc u es and mobile de ices. Wi hou ailo ed p o ec ions, us in
psychia ic AI pla o ms isks e osion [2]. Figu e 1 unde sco es how ising adop ion inc eases exposu e o cybe h ea s,
and Table 1 e eals ha p o ide s alloca ing esou ces o cybe secu i y epo be e pa ien e en ion a es.
Add essing hese isks is essen ial o sus aining digi al psychia y as a iable heal hca e deli e y model.
1.3. Need o u u e-p oo solu ions: pos -quan um c yp og aphy and neu o-symbolic AI
Fu u e-p oo ing psychia ic AI equi es in eg a ing bo h ad anced c yp og aphic p o ec ions and hyb id AI models
capable o explaining hei easoning. Pos -quan um c yp og aphy, cu en ly unde e alua ion by he Na ional Ins i u e
o S anda ds and Technology, aims o c ea e enc yp ion me hods esis an o quan um dec yp ion capabili ies [5]. This
is pa icula ly ele an o psychia ic da a, which equi e long- e m con iden iali y gi en hei li elong sensi i i y [3].
In pa allel, neu o-symbolic AI app oaches o e a pa hway o combine he pa e n ecogni ion s eng h o neu al
ne wo ks wi h he in e p e abili y o symbolic easoning [6]. Fo psychia ic he apy models, his is c ucial: pa ien s
and clinicians mus us AI ecommenda ions, and anspa en decision-making helps mi iga e conce ns abou bias o
inapp op ia e guidance [8]. By enhancing explainabili y, neu o-symbolic AI aligns wi h e hical impe a i es in men al
heal h ca e [1].
In eg a ing pos -quan um c yp og aphy wi h neu o-symbolic AI can c ea e psychia ic sys ems ha a e simul aneously
secu e and in e p e able. Such sys ems would be esilien agains bo h nea - e m cybe a acks and long- e m
echnological h ea s [4]. Table 1 highligh s compa a i e amewo ks o esilience planning, and Figu e 1 illus a es
he u gency by co ela ing adop ion cu es wi h p ojec ed ulne abili y windows. Ensu ing he U.S. leads in his
in eg a ion no only secu es psychia ic ca e bu also se s a global p eceden in esponsible digi al heal hca e.
2. Backg ound and li e a u e e iew
2.1. E olu ion o Cybe secu i y in Heal hca e Sys ems
Cybe secu i y wi hin heal hca e has unde gone a p o ound ans o ma ion, e lec ing bo h ad ances in digi al
in as uc u e and he g owing sophis ica ion o cybe h ea s. Ea ly digi iza ion e o s p ima ily cen e ed on elec onic
heal h eco ds (EHRs), whe e secu i y measu es consis ed o access con ols, passwo ds, and pe ime e de enses [7].
These mechanisms we e udimen a y, o en assuming ha h ea s o igina ed ex e nally and unde es ima ing he isks
posed by inside misuse.
The subsequen in eg a ion o in e connec ed de ices spanning eleheal h, mobile apps, and In e ne o Things (IoT)-
enabled medical de ices exponen ially expanded a ack su aces [8]. Ransomwa e a acks a ge ing hospi als
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
3
demons a ed how ulne abili ies in ou da ed in as uc u e could pa alyze en i e sys ems, comp omise pa ien sa e y,
and gene a e se e e inancial losses [9]. Heal hca e hus eme ged as one o he mos luc a i e a ge s o cybe c iminals
due o i s combina ion o sensi i e da a and ope a ional u gency.
F om 2010 onwa d, egula o y amewo ks such as HIPAA in he U.S. and GDPR in Eu ope manda ed s ic e p o ec ions
and epo ing mechanisms, emphasizing enc yp ion, audi ails, and access go e nance [10]. Ye , compliance-d i en
app oaches o en lagged behind dynamic h ea e olu ion. The eme gence o ad anced pe sis en h ea s (APTs)
e ealed ha heal hca e ins i u ions lacked esilience agains long- e m, s eal hy a acks exploi ing sys emic
weaknesses [11].
In ecen yea s, a i icial in elligence (AI)-d i en analy ics, anomaly de ec ion, and ze o- us a chi ec u es ha e become
inc easingly p ominen in heal h cybe secu i y. These ools enable eal- ime moni o ing and esponse a he han
eac i e eco e y [12]. Howe e , as psychia ic he apy pla o ms inco po a e pe sonalized algo i hms and sensi i e
beha io al da a, he consequences o b eaches now ex end beyond inancial o clinical ha m o include long-las ing
psychosocial impac s on ulne able popula ions [13].
This his o ical p og ession unde sco es a ecu ing heme: heal hca e cybe secu i y e ol es eac i ely, chasing
inno a ions a e adop ion a he han embedding p o ec ions du ing sys em design. Add essing his lag is pa icula ly
c i ical in psychia ic he apy sys ems whe e pa ien us is cen al [14].
2.2. Psychia ic The apy Models: Da a Sensi i i y and AI-D i en Pe sonaliza ion
Psychia ic he apy models ha e unique cybe secu i y demands because o he in ima e and s igma ized na u e o he
da a hey gene a e. Unlike gene al heal h in o ma ion, psychia ic eco ds o en include na a i es abou auma,
subs ance use, o social unc ioning de ails ha , i disclosed, could se e ely ha m pa ien s’ epu a ions and men al well-
being [15]. The high sensi i i y o his da a makes psychia ic pla o ms p ime a ge s o iden i y he , ex o ion, o
manipula ion [9].
AI-d i en pe sonaliza ion adds bo h alue and complexi y o hese he apy models. Algo i hms inc easingly analyze
pa ien speech, sen imen , and beha io al pa e ns o ecommend ailo ed in e en ions, o e ing scalable suppo in
con ex s o global men al heal h sho ages [8]. While his pe sonaliza ion enhances he apeu ic ou comes, i also
mul iplies he numbe o da a s eams anging om biome ic senso s o na u al language inpu s equi ing obus
in eg a ion [16].
Fu he mo e, psychia ic sys ems inc easingly inco po a e mobile apps, cha bo s, and elepsychia y pla o ms. These
mediums in oduce da a collec ion h ough geoloca ion, eal- ime oice analysis, and e en social media ac i i y, he eby
widening he exposu e o sensi i e da ase s [11]. In cases whe e AI models ely on cloud s o age and hi d-pa y
analy ics, isks a e exace ba ed by he po en ial o c oss-bo de da a ans e s ha may no align wi h local p i acy
laws [7].
The e hical implica ions o b eaches in his con ex a e se e e. Pa ien s may a oid seeking ea men i hey pe cei e
digi al pla o ms as unsa e, ul ima ely wo sening public heal h ou comes [13]. Mo eo e , he s igma iza ion o men al
illness ampli ies he impac o disclosu e compa ed o b eaches o non-psychia ic da a.
Thus, psychia ic he apy models p esen a pa adox: hei eliance on AI pe sonaliza ion p omises unp eceden ed
accessibili y and p ecision in men al heal h ca e, bu i simul aneously in oduces ulne abili ies ha demand equally
ad anced cybe secu i y mechanisms [10].
2.3. Cu en C yp og aphic P o ec ions and Vulne abili ies
C yp og aphy emains he co ne s one o cybe secu i y in psychia ic digi al heal h sys ems. S anda d p o ec ions
include symme ic enc yp ion o da a-a - es and asymme ic enc yp ion o secu e communica ions be ween pa ien s,
p o ide s, and cloud se e s [9]. P o ocols such as AES and RSA domina e, supplemen ed by secu e key exchange
mechanisms like Di ie-Hellman [14]. Mul i- ac o au hen ica ion adds a u he laye o esilience agains c eden ial
he .
Howe e , he psychia ic con ex exposes he limi s o hese sa egua ds. Fo ins ance, enc yp ed communica ions may
s ill be in e cep ed h ough side-channel a acks o imp ope ly con igu ed sys ems [7]. Mo eo e , eliance on public-
key in as uc u e (PKI) can become cumbe some in esou ce-limi ed se ings, whe e key managemen is e o -p one
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
4
[15]. As psychia ic he apy o en in ol es con inuous moni o ing, la ency in oduced by c yp og aphic p o ocols can
hinde eal- ime esponsi eness, unde mining pa ien expe ience [16].
A c i ical ulne abili y lies in he s o age o longi udinal pa ien his o ies. E en i enc yp ed, cen alized da abases
emain a ac i e o a acke s capable o b u e- o ce dec yp ion o exploi ing inside p i ileges [13]. The high esea ch
alue o psychia ic da ase s use ul o pha maceu ical de elopmen and beha io al s udies u he incen i izes
a ge ed in usions [17].
Recen de elopmen s such as homomo phic enc yp ion and blockchain-based audi ails o e p omising a enues,
enabling compu a ion on enc yp ed da a and immu able access logs [8]. Ye hese inno a ions emain expe imen al and
esou ce-in ensi e, limi ing scalabili y in mains eam psychia ic ca e.
Figu e 1 Concep ual o e iew o psychia ic he apy model ulne abili ies and a ack su aces in digi al heal h
pla o ms
As illus a ed in Figu e 1, ulne abili ies a e mul i ace ed, encompassing communica ion channels, s o age sys ems, and
algo i hmic laye s. While c yp og aphic echniques mi iga e many isks, hey canno ully add ess inside h ea s,
ad e sa ial machine lea ning, o sys em miscon igu a ions. A laye ed de ense s a egy combining c yp og aphy wi h
con inuous moni o ing and AI-d i en anomaly de ec ion is he e o e indispensable [12].
2.4. Limi a ions o Con en ional Machine Lea ning o Secu i y Moni o ing
Con en ional machine lea ning (ML) models ha e been widely deployed in heal hca e secu i y moni o ing, pa icula ly
o anomaly de ec ion, in usion de ec ion sys ems (IDS), and aud p e en ion [10]. These models ely on his o ical
da a o classi y no mal e sus suspicious ac i i y, using s a is ical baselines o lag de ia ions [9]. While e ec i e in
s uc u ed en i onmen s, psychia ic he apy pla o ms p esen unique complexi ies ha con en ional ML s uggles o
cap u e.
Fi s , psychia ic da a is o en uns uc u ed, comp ising ee- ex session ansc ip s, oice eco dings, and beha io al
logs. Linea classi ie s and s a ic anomaly de ec ion models ail o accommoda e he e ol ing, non-linea na u e o
pa ien – he apis in e ac ions [15]. Fo example, unusual login pa e ns may e lec legi ima e he apeu ic needs such
as la e-nigh c ises a he han malicious ac i i y [13]. S a ic ML models isk gene a ing alse posi i es, unde mining
clinician us in moni o ing sys ems [7].
Second, ad e sa ial a acks pose a signi ican challenge. Malicious ac o s can delibe a ely c a inpu s o decei e ML
classi ie s, c ea ing blind spo s in secu i y moni o ing [16]. Fo psychia ic pla o ms, ad e sa ial manipula ions could
disguise unau ho ized access o co up he apy algo i hms. T adi ional ML, which assumes s a iona y da a
dis ibu ions, is ill-equipped o handle such dynamic h ea s [14].
Thi d, con en ional ML o en lacks explainabili y. In high-s akes domains like men al heal h, opaque de ec ion
mechanisms p e en clinicians and egula o s om unde s anding why ce ain beha io s we e lagged [11]. This
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
5
unde mines accoun abili y, pa icula ly in cases whe e in e en ions based on alse ale s dis up he apeu ic
con inui y [8].
These limi a ions highligh he need o adap i e, con ex -awa e, and explainable AI app oaches. Inco po a ing deep
lea ning, ede a ed lea ning, and eal- ime ein o cemen models may o e pa hways owa d mo e esilien psychia ic
cybe secu i y [17]. Howe e , un il such sys ems ma u e, eliance on con en ional ML alone lea es psychia ic pla o ms
exposed o c i ical ulne abili ies ha di ec ly impac pa ien us and ca e ou comes [12].
3. Pos -quan um c yp og aphy ounda ions
3.1. Th ea o Quan um Compu ing o Classical C yp og aphy
The looming ad en o quan um compu ing p esen s one o he mos signi ican h ea s o classical c yp og aphic
me hods unde pinning heal hca e sys ems. Classical enc yp ion schemes such as RSA and ellip ic-cu e c yp og aphy
(ECC) ely on he compu a ional ha dness o ac o ing la ge in ege s o sol ing disc e e loga i hm p oblems. Quan um
algo i hms, pa icula ly Sho ’s algo i hm, could sol e hese p oblems in polynomial ime, ende ing such schemes
obsole e [16].
Heal hca e sys ems, including psychia ic he apy pla o ms, a e pa icula ly exposed because o hei eliance on
enc yp ed communica ion o ansmi ing highly sensi i e eco ds [18]. Once a su icien ly powe ul quan um
compu e becomes ope a ional, ad e sa ies could e ospec i ely dec yp p e iously cap u ed ciphe ex s, leading o
ca as ophic b eaches o longi udinal pa ien his o ies [17]. This “ha es now, dec yp la e ” model is especially
oubling in psychia y, whe e con iden ial eco ds ha e endu ing sensi i i y ha may a ec pa ien s’ decades in o he
u u e [20].
Mo eo e , quan um capabili ies ex end beyond b eaking enc yp ion o accele a ing b u e- o ce a acks agains
symme ic schemes. While algo i hms like AES-256 emain ela i ely secu e due o hei key leng h, G o e ’s algo i hm
e ec i ely hal es he secu i y ma gin by educing he b u e- o ce complexi y [22]. Fo sys ems al eady cons ained by
la ency in eal- ime psychia ic moni o ing, compensa ing wi h longe key leng hs imposes pe o mance ade-o s ha
deg ade pa ien - acing se ices [19].
The heal hca e sec o is u he disad an aged by i s ypically slow adop ion o cu ing-edge echnologies due o
egula o y bu dens and cos cons ain s. Thus, as quan um capabili ies eme ge, hospi als and psychia ic pla o ms may
ace an asymme y whe e a acke s le e age quan um ools long be o e de ende s ha e ansi ioned o pos -quan um
sa egua ds [23]. Recognizing his looming h ea , in e na ional s anda diza ion bodies such as NIST ha e accele a ed
e o s o inalize pos -quan um c yp og aphic ecommenda ions, signaling an u gen need o heal hca e eadiness
[24].
3.2. Key Pos -Quan um C yp og aphic Families: La ice-Based, Code-Based, Mul i a ia e, Hash-Based, and
Isogeny-Based
Pos -quan um c yp og aphy (PQC) encompasses se e al amilies o algo i hms designed o esis quan um a acks
while main aining easible e iciency o eal-wo ld sys ems. La ice-based c yp og aphy, pa icula ly schemes such as
Lea ning wi h E o s (LWE), is conside ed he mos p omising due o i s e sa ili y and balance o e iciency and
secu i y [17]. These sys ems ely on he ha dness o la ice p oblems, which emain in ac able o bo h classical and
quan um algo i hms.
Code-based app oaches, pionee ed by he McElwee c yp osys em, de i e secu i y om he di icul y o decoding gene al
linea codes [16]. While obus agains quan um a acks, hei la ge key sizes ha e his o ically hinde ed widesp ead
deploymen . None heless, hei esilience makes hem appealing o heal hca e pla o ms whe e du abili y and long-
e m secu i y ou weigh s o age cos s [18].
Mul i a ia e polynomial schemes ely on sol ing sys ems o non-linea equa ions o e ini e ields. Al hough hey o e
e icien signa u e gene a ion, ce ain designs ha e been ulne able o algeb aic a acks, limi ing hei adop ion o
heal hca e-g ade applica ions [20]. Hash-based signa u es, such as XMSS, o e quan um esis ance oo ed in he
p eimage esis ance o hash unc ions. They a e especially a ac i e in elemedicine se ings equi ing ligh weigh , one-
ime secu e au hen ica ion [22].
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
6
Isogeny-based schemes, le e aging he ma hema ical complexi y o ellip ic cu e isogenies, p o ide ela i ely compac
key sizes and enc yp ion signa u es. Howe e , hey a e s ill unde sc u iny due o eme ging a acks ha ha e weakened
some candida e cons uc ions [23].
Each amily ca ies ade-o s in e ms o e iciency, key size, and implemen a ion complexi y, making hei deploymen
con ex -speci ic. In heal hca e, whe e esou ces a y ac oss egions and compliance amewo ks impose s ic
cons ain s, choosing he app op ia e scheme equi es balancing heo e ical s eng h agains ope a ional easibili y
[24].
3.3. Applicabili y o Pos -Quan um Schemes o Heal hca e Models
The applicabili y o pos -quan um c yp og aphy in heal hca e, and speci ically psychia ic he apy pla o ms, lies in i s
abili y o p o ec long- e m sensi i e da a while suppo ing eal- ime clinical wo k lows. La ice-based sys ems a e
pa icula ly p omising because hey can be adap ed o bo h enc yp ion and digi al signa u es, secu ing pa ien –
p o ide communica ions while au hen ica ing access o psychia ic he apy eco ds [16].
Code-based schemes, despi e hei s o age demands, may be app op ia e o cen al eposi o ies o psychia ic heal h
da a main ained by na ional ins i u ions. Thei esilience agains bo h classical and quan um ad e sa ies ensu es ha
once eco ds a e enc yp ed, hey emain secu e o decades, a necessi y in psychia y gi en he endu ing sensi i i y o
men al heal h da a [18].
Hash-based signa u es ha e p ac ical alue o session-based eleheal h pla o ms, whe e sho -li ed bu equen
au hen ica ion is necessa y. Thei low compu a ional o e head enables secu e logins and digi al p esc ip ions wi hou
dis up ing use expe ience [20]. Meanwhile, mul i a ia e schemes may se e in niche applica ions such as ligh weigh
de ices o emo e psychia ic moni o ing, hough hei ulne abili y o algeb aic c yp analysis limi s long- e m
adop ion [22].
Isogeny-based app oaches, while compac and bandwid h-e icien , equi e u he ma u a ion be o e being in eg a ed
in o mains eam psychia ic heal h pla o ms [23]. Howe e , hei po en ial compa ibili y wi h esou ce-cons ained
de ices makes hem candida es o wea able heal h moni o ing sys ems in he u u e.
Table 1 Compa ison o classical s. pos -quan um c yp og aphic me hods wi h s eng hs, weaknesses, and heal hca e
applicabili y
C yp og aphic
Family
S eng hs
Weaknesses
Heal hca e Applicabili y
RSA / ECC
(Classical Public-
Key)
Widely adop ed,
s anda dized, e icien o
cu en sys ems
Vulne able o Sho ’s
algo i hm unde quan um
compu ing; educed u u e
iabili y
Secu e elec onic heal h eco ds
(EHRs) and au hen ica ion
oday bu no sus ainable long-
e m
La ice-Based
(PQC)
S ong secu i y
assump ions, e icien key
exchange, scalable o high
da a olumes
La ge key sizes, po en ial
pe o mance o e head
Sui able o p o ec ing
psychia ic he apy models and
ede a ed lea ning pipelines
wi h la ge da ase s
Code-Based (PQC)
High esilience o quan um
a acks, well-s udied (e.g.,
McElwee)
Ex emely la ge public keys
make deploymen
challenging
Use ul o a chi ing long- e m
psychia ic eco ds whe e
s o age space is less cons ained
Mul i a ia e
Polynomial (PQC)
Fas signa u e gene a ion,
low compu a ional cos
La ge signa u e sizes, less
ma u e s anda diza ion
Sui able o ligh weigh heal h
applica ions like mobile
psychia ic moni o ing ools
Hash-Based (PQC)
Secu i y oo ed in well-
unde s ood c yp og aphic
hashes, simple and obus
Typically limi ed o signa u e
schemes, may equi e s a e
managemen
Ideal o ensu ing in eg i y o
he apy da a logs and audi
ails
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
7
Isogeny-Based
(PQC)
Small key sizes, e icien o
cons ained de ices
Less ma u e, ewe secu i y
p oo s compa ed o
la ice/code-based schemes
P omising o wea able
heal hca e de ices and IoT-
enabled psychia ic moni o ing
pla o ms
As summa ized in Table 1, classical c yp og aphic me hods like RSA and ECC a e e icien oday bu ace obsolescence
unde quan um h ea s, while pos -quan um schemes o e a ying ade-o s in key size, speed, and adap abili y [24].
Heal hca e sys ems mus begin es ing hese schemes in pa allel o ensu e con inui y and esilience, a oiding he isk
o u u e da a exposu e h ough delayed adop ion [19].
4. Neu o-symbolic ai o secu e and explainable sa egua ds
4.1. O e iew: Me ging Neu al Ne wo ks wi h Symbolic Reasoning
The con e gence o neu al ne wo ks and symbolic easoning has eme ged as a compelling pa adigm in a i icial
in elligence (AI), aiming o b idge he gap be ween s a is ical lea ning and logical in e ence. Neu al a chi ec u es,
pa icula ly deep lea ning models, excel a ecognizing pa e ns wi hin high-dimensional da a such as pa ien eco ds
o clinical na a i es. Howe e , hey s uggle wi h anspa ency, logical consis ency, and he abili y o en o ce
s uc u ed ules c i ical in sensi i e en i onmen s like heal hca e [23]. Symbolic easoning, in con as , le e ages
explici ules and logical ope a o s, enabling aceable decision-making p ocesses bu lacking adap abili y o noisy,
uns uc u ed da a [26].
By me ging hese app oaches, neu o-symbolic AI combines he scalabili y and accu acy o neu al ne wo ks wi h he
in e p e abili y and igo o symbolic logic [22]. This usion is especially ele an o psychia ic he apy models, whe e
da a complexi y spans ex , audio, biome ic signals, and beha io al pa e ns ha canno be cap u ed by one pa adigm
alone [25]. In eg a ing symbolic laye s allows easoning o e high-le el cons uc s, such as compliance policies, e hical
cons ain s, o p i acy ules embedded wi hin he apy pla o ms.
Mo eo e , neu o-symbolic amewo ks enhance esilience agains ad e sa ial h ea s, as symbolic cons ain s can ac
as sa egua ds agains model d i o unin ended in e ences [29]. Fo example, while neu al ne wo ks may iden i y
anomalous biome ic spikes, symbolic easoning ensu es hese de ec ions align wi h clinically alida ed ules. This dual
mechanism o e s no only accu acy bu also accoun abili y, c ea ing an AI laye o us wo hiness ha is cen al o
psychia ic heal h applica ions [28].
Thus, he syn hesis o symbolic and neu al pa adigms c ea es a new on ie whe e compu a ional in elligence can
e ol e in o con ex -sensi i e, egula ion-complian , and clinically us wo hy sys ems [30]. This es ablishes he
ounda ion o p ac ical applica ions in anomaly de ec ion and policy en o cemen .
4.2. Applica ions in Anomaly De ec ion and Policy En o cemen
Anomaly de ec ion is undamen al o p o ec ing psychia ic he apy models because malicious in usions o en mani es
as sub le i egula i ies wi hin sensi i e da ase s [27]. Con en ional machine lea ning me hods can lag de ia ions bu
s uggle o con ex ualize hem. Neu o-symbolic AI add esses his by embedding domain-speci ic knowledge wi hin he
anomaly de ec ion pipeline. Fo example, de ia ions in session equency o a ypical biome ic pa e ns can be
in e p e ed agains es ablished psychia ic he apy schedules, imp o ing he speci ici y o de ec ion [24].
This in eg a ion also enhances explainabili y. Ins ead o simply labeling a da a poin as anomalous, he sys em can
p o ide symbolic easoning ou pu s ha a icula e why he de ia ion con lic s wi h no ma i e ules [26]. Such
in e p e abili y is i al o men al heal h p o essionals who mus alida e ale s wi hou unde mining clinical
wo k lows. Addi ionally, he dual amewo k educes alse posi i es a common challenge in anomaly de ec ion by
combining s a is ical h esholds wi h symbolic easoning il e s [22].
Beyond anomaly de ec ion, neu o-symbolic models con ibu e signi ican ly o policy en o cemen in digi al psychia ic
pla o ms. Regula ions such as GDPR and HIPAA manda e s ingen da a p o ec ion p o ocols, which equi e con inuous
moni o ing o da a access, sha ing, and usage [23]. Neu al models can moni o a ic a scale, while symbolic ules
ensu e compliance wi h access hie a chies and consen amewo ks [28]. Fo ins ance, a sys em could au oma ically
block unau ho ized c oss-bo de ans e s o he apy da a while logging he decision in compliance- iendly symbolic
e ms.
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
8
By embedding bo h clinical knowledge and legal cons ain s wi hin AI wo k lows, neu o-symbolic sys ems ex end
beyond adi ional moni o ing owa d p oac i e en o cemen mechanisms. This ensu es ha psychia ic he apy
pla o ms main ain no only esilience agains echnical in usions bu also adhe ence o e ol ing e hical and legal
s anda ds [29].
4.3. Rele ance o Psychia ic The apy Model P o ec ion
Psychia ic he apy models ace a dual challenge: sa egua ding sensi i e pa ien in o ma ion and ensu ing he apeu ic
p ocesses emain us wo hy in digi al ecosys ems. Neu o-symbolic AI di ec ly add esses bo h dimensions by o e ing
in e p e abili y, esilience, and compliance wi hin a single amewo k [25]. In p ac ice, he apy da a pipelines in ol e
mul imodal inpu s elec onic heal h eco ds, oice da a om he apy sessions, and wea able de ice ou pu s. Neu al
ne wo ks excel a p ocessing such he e ogeneous da a, bu wi hou symbolic easoning, hey isk p oducing opaque
ou pu s wi h limi ed accoun abili y [22].
Figu e 2 Wo k low o neu o-symbolic AI o con inuous moni o ing and in e p e abili y in psychia ic he apy da a
pipelines illus a es how neu al laye s p ocess mul imodal inpu s, while symbolic componen s en o ce compliance
and in e p e anomalies. This laye ed s uc u e ensu es he apy sessions lagged as unusual a e no me ely ea ed as
s a is ical ou lie s bu a e c oss- alida ed agains logical ules de i ed om clinical guidelines [30]
F om a cybe secu i y pe spec i e, neu o-symbolic amewo ks p o ide a unique de ense agains ad e sa ial a acks
aimed a exploi ing model weaknesses [23]. By in eg a ing symbolic easoning, a acke s canno as easily manipula e
he sys em in o alse p edic ions wi hou also b eaking high-le el logical ules [28]. This c ea es an added laye o
p o ec ion compa ed o pu ely neu al app oaches.
The app oach also has e hical signi icance. Pa ien s and he apis s alike mus us ha sensi i e he apy da a will no
be misused o misin e p e ed. The symbolic easoning laye p o ides anspa en jus i ica ions o sys em ac ions,
ensu ing s akeholde s a e in o med and con iden in AI-d i en moni o ing ou comes [27].
Ul ima ely, by embedding policy awa eness, anomaly de ec ion, and in e p e abili y, neu o-symbolic AI aligns
seamlessly wi h he p o ec ion needs o psychia ic he apy models. I es ablishes a amewo k whe e clinical ca e,
cybe secu i y, and legal compliance con e ge in sa egua ding ulne able popula ions wi hin digi al heal h ecosys ems
[29].
5. In eg a ed secu i y amewo k: PQC + neu o-symbolic ai
5.1. A chi ec u e Design o Psychia ic The apy Model P o ec ion
The a chi ec u e o secu ing psychia ic he apy models in digi al heal hca e pla o ms equi es a mul i-laye ed design
ha accoun s o con iden iali y, in eg i y, a ailabili y, and compliance simul aneously. Unlike con en ional heal hca e
IT sys ems, he apy models handle ex emely sensi i e, mul imodal inpu s such as speech, biome ic s eams, and
beha io al ma ke s, all o which necessi a e obus secu i y con ols [29].
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
9
The ounda ion o his a chi ec u e in eg a es pos -quan um c yp og aphy (PQC) o da a con iden iali y and
au hen ici y, neu o-symbolic AI o in elligen moni o ing, and s anda dized compliance modules o legal en o cemen .
A he lowes laye , c yp og aphic mechanisms secu e aw da a collec ion, ensu ing ha he apy ansc ip s, biome ic
signals, and me ada a emain p o ec ed om in e cep ion [31]. Abo e his, s o age and p ocessing nodes u ilize hyb id
enc yp ion o p o ec in e media e model ou pu s, while in eg i y checks con i m esis ance o ampe ing.
The middle laye embeds AI-d i en moni o ing sys ems ha blend anomaly de ec ion wi h symbolic easoning,
enabling eal- ime policy en o cemen . Fo ins ance, access anomalies a e lagged agains p ede ined symbolic
compliance ules while neu al submodules e alua e usage pa e ns o hidden ad e sa ial in usions [34]. This ensu es
ha pa ien da a lows emain consis en wi h bo h cybe secu i y and e hical equi emen s.
The uppe laye o he a chi ec u e ocuses on go e nance, ensu ing alignmen wi h HIPAA, GDPR, and eme ging
egional digi al heal h s anda ds [33]. He e, explainable decision ou pu s a e logged and documen ed, c ea ing audi able
ails o heal hca e egula o s. The laye ed in eg a ion educes single poin s o ailu e by dis ibu ing p o ec ion ac oss
c yp og aphic, easoning, and compliance dimensions [30].
Ul ima ely, his a chi ec u e does no me ely sa egua d he apy models om ex e nal b eaches bu embeds
us wo hiness in o he pla o m i sel . By syne gizing PQC wi h neu o-symbolic easoning, psychia ic he apy
sys ems e ol e om eac i e o p oac i ely esilien amewo ks ha can adap o eme ging h ea s in eal ime [37].
5.2. C yp og aphic Modules: Enc yp ion, Digi al Signa u es, Key Managemen
C yp og aphic modules o m he backbone o da a p o ec ion wi hin psychia ic he apy sys ems. Enc yp ion
mechanisms, pa icula ly la ice-based and code-based PQC algo i hms, sa egua d sensi i e inpu s agains he looming
h ea o quan um dec yp ion [32]. Unlike classical schemes ulne able o Sho ’s algo i hm, PQC ensu es ha
psychia ic da a e ains con iden iali y in long- e m s o age and ansmission [29].
Digi al signa u es se e as he assu ance o au hen ici y and non- epudia ion. In psychia ic pla o ms, hey e i y no
only clinician iden i y bu also model-gene a ed ecommenda ions, educing isks o o ge y o manipula ion [35].
Symbolic easoning modules u he e i y whe he hese c yp og aphic signa u es align wi h au ho ized wo k lows,
ensu ing ha only alida ed ansac ions and he apeu ic ecommenda ions en e he sys em.
Key managemen eme ges as a c i ical ulne abili y in his con ex . T adi ional symme ic and asymme ic amewo ks
o en ely on cen alized au ho i ies, which c ea e po en ial single poin s o ailu e [34]. To mi iga e his, decen alized
PQC-based key exchange p o ocols dis ibu e us ac oss mul iple nodes, enhancing esilience agains inside and
ex e nal a acks. Neu o-symbolic modules can o e see key o a ion schedules and en o ce access ules dynamically,
ensu ing c yp og aphic hygiene aligns wi h clinical go e nance [31].
By embedding hese modules wi hin a laye ed a chi ec u e, he apy sys ems achie e con iden iali y h ough enc yp ion,
in eg i y ia signa u es, and esilience h ough adap i e key managemen . Toge he , hey p o ide a baseline o
ma hema ical obus ness upon which highe -o de AI moni o ing laye s ope a e [36].
5.3. AI-D i en Moni o ing and Reasoning o Adap i e De ense
AI-d i en moni o ing in oduces adap abili y in o psychia ic he apy model p o ec ion, complemen ing he s a ic
gua an ees o c yp og aphy. Neu al ne wo ks excel a iden i ying de ia ions in high-dimensional da a s eams, such as
sudden spikes in he apy session equency, i egula biome ic pa e ns, o unusual c oss-pla o m da a access [30].
Howe e , elying solely on neu al moni o ing isks gene a ing opaque esul s, unde mining us . This is whe e
symbolic easoning modules ein o ce in e p e abili y.
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
16
[9] Ra hna S. Enhancing So wa e De elopmen wi h Adap i e AI h ough MASRL, Bayesian Op imiza ion, and GNNs
o Ad anced Code In elligence. In e na ional Jou nal. 2024;9(5):1-8.
[10] Thelma Chibueze, Taiwo Adeshina, Linda Uzoamaka Ch is ophe , S ephanie Dolapo Ewubajo, Lisa Ebe e. Access
o c edi and inancial inclusion o MSMEs in sub-Saha an A ica: Challenges and oppo uni ies. In J Finance
Manage Econ 2025;8(2):861-872. DOI: 10.33545/26179210.2025. 8.i2.609
[11] Mahama T. Gene alized addi i e model using ma ginal in eg a ion es ima ion echniques wi h in e ac ions.
In e na ional Jou nal o Science Academic Resea ch. 2023;4(5):5548-5560.
[12] Ojo OR, Elesho OE, Adeniyi A, Oluwadamilola OJ. Applying machine lea ning models o eal- ime p ocess
moni o ing and anomaly de ec ion in pha ma manu ac u ing. GSC Biol Pha m Sci. 2024;27(1):315-41.
doi:10.30574/gscbps.2024.27.1.0153.
[13] Alozie M. Sus ainable p ocu emen p ac ices in cons uc ion p ojec s d i ing eco- iendly in as uc u e, e hical
con ac ing, and long- e m esilience in u ban de elopmen . In J Eng Technol Res Manag (IJETRM). 2022
Dec;6(12).
[14] Ade emi Bunmi Ku elu, and Baba unde Ib ahim Ojoawo. 2025. “Digi al S a egies o Sus ainable Ag icul u al
Ou each: A Model o Food Secu i y Ad ocacy”. Cu en Jou nal o Applied Science and Technology 44 (7):104–
113. h ps://doi.o g/10.9734/cjas /2025/ 44i74577.
[15] Ukaoha C. De e minan s o adop ion and echnical e iciency o bio o i ied c ops among smallholde a me s in
No h-Cen al Nige ia. Magna Scien ia Ad anced Resea ch and Re iews. 2021;3(2):108-121. doi:
h ps://doi.o g/10.30574/msa .2021.3.2.0091
[16] Mahama T. Bayesian hie a chical modeling o small-a ea es ima ion o disease bu den. In e na ional Jou nal o
Science and Resea ch A chi e. 2022;7(2):807-827. doi: h ps://doi.o g/10.30574/ijs a.2022.7.2.0295
[17] Abi R. E hical and explainable AI in da a science o anspa en decision-making ac oss c i ical business
ope a ions. In e na ional Jou nal o Ad ance Resea ch Publica ion and Re iews. 2025;2(6):50-72. doi:
h ps://doi.o g/10.55248/gengpi.6.0625.2126
[18] Akangbe BO, Akinwumi FE, Adekunle DO, Tijani AA, Aneke OB, Anukam S, Akangbe B, Adekunle D, Tijani A, Aneke
O. Como bidi y o Anxie y and Dep ession Wi h Hype ension Among Young Adul s in he Uni ed S a es: A
Sys ema ic Re iew o Bidi ec ional Associa ions and Implica ions o Blood P essu e Con ol. Cu eus. 2025 Jul
22;17(7). doi:10.7759/cu eus.88532.
[19] Ukaoha C. Economic impac o poul y supply chain dis up ions on ood secu i y: E idence om pos -pandemic
ma ke ola ili y in Wes A ica. Wo ld J Ad Res Re . 2023;20(3):2380-94. doi:
h ps://doi.o g/10.30574/wja .2023.20.3.2507
[20] Mammah CU. Digi al T ans o ma ion in A ican Re ail Banking: Adop ion Ba ie s and S a egic Enable s. In J
Ad Mul idisc Res S ud. 2024;4(2):1578-84. doi: h ps://doi.o g/10.62225/2583049X.2024.4.2.4824.
[21] Ase on, T. I. (2025). U ilizing Chlo ide and B omide Le els as an Indica o o Wa e Quali y in he Mahoning Ri e
Wa e shed [Mas e 's hesis, Youngs own S a e Uni e si y]. OhioLINK Elec onic Theses and Disse a ions Cen e .
h p:// a e.ohiolink.edu/e dc/ iew?acc_num=ysu175612989060847
[22] O oko J. Op imizing cos , ime, and con amina ion con ol in clean oom cons uc ion using ad anced BIM, digi al
win, and AI-d i en p ojec managemen solu ions. Wo ld J Ad Res Re . 2023;19(02):1623-38. doi:
h ps://doi.o g/10.30574/wja .2023.19.2.1570
[23] Boullosa Dapena Ó. Design o a Quan um-Awa e Embedded Language o Au onomous Cybe de ense o Sa elli e
Cons ella ions in Hos ile En i onmen s. A ailable a SSRN 5362121. 2025 Jul 22.
[24] Jemimah O oko. MULTI OBJECTIVE OPTIMIZATION OF COST, CONTAMINATION CONTROL, AND
SUSTAINABILITY IN CLEANROOM CONSTRUCTION: A DECISIONSUPPORT MODEL INTEGRATING LEAN SIX
SIGMA, MONTE CARLO SIMULATION, AND COMPUTATIONAL FLUID DYNAMICS (CFD). In e na ional Jou nal o
Enginee ing Technology Resea ch and Managemen (ije m). 2023Jan21;07(01).
[25] Raza A. Scalable A chi ec u es o Dis ibu ed Commonsense Knowledge Bases wi h Real-Time Synch oniza ion
and Faul Tole ance. Open Jou nal o Robo ics, Au onomous Decision-Making, and Human-Machine In e ac ion.
2022 Dec 4;7(12):1-6.
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
17
[26] O oko J. Economic impac o clean oom in es men s: s eng hening U.S. ad anced manu ac u ing, job g ow h,
and echnological leade ship in global ma ke s. In J Res Publ Re . 2025;6(2):1289-1304. doi:
h ps://doi.o g/10.55248/gengpi.6.0225.0750
[27] Tang X, Li Q. Logical Gene Encoding: a Bio-Inspi ed App oach o Ene gy-E icien Au oma ed Reasoning. In5 h
In e na ional Con e ence on In e ne , Educa ion and In o ma ion Technology (IEIT 2025) 2025 Jul 31 (pp. 830-
844). A lan is P ess.
[28] Samson-Onuo ah CI. AI-d i en c edi isk modeling: Le e aging big da a analy ics o imp o e inancial s abili y
and lending e iciency in banks. In J Sci Eng Appl. 2025;14(10):57-70. doi:10.7753/IJSEA1410.100925 ci a ion
[29] O oko J, O oko GA. Clean oom-d i en ae ospace and de ense manu ac u ing: enabling p ecision enginee ing,
mili a y eadiness, and economic g ow h. In J Compu Appl Technol Res. 2023;12(11):42-56.
doi:10.7753/IJCATR1211.1007
[30] Okuwobi FA, Akomola e OO, Majebi NL. Neu odi e si y and equi y: Designing cul u ally esponsi e ABA ools o
di e se popula ions. In J Appl Res Soc Sci. 2025;7(9):553-81.
[31] Mammah CU. The Role o Women in Execu i e Banking Posi ions: Challenges and Success S a egies in Sub-
Saha an A ica. In J Ad Mul idisc Res S ud. 2023;3(2):1230-8.
[32] Jü ise M. " The TeO Ene gy Re olu ion: Mehhano onika™ and he Cosmic Symbiosis o Powe "; AI and he
Bounda ies o T u h–Scien i ic Tool o Specula i e Aid?. " The TeO Ene gy Re olu ion: Mehhano onika™ and he
Cosmic Symbiosis o Powe "; AI and he Bounda ies o T u h–Scien i ic Tool o Specula i e Aid?. 2025 Jan 1.
[33] Umako MF. Enhancing cloud secu i y pos u es: a mul i-laye ed amewo k o de ec ing and mi iga ing
eme ging cybe h ea s in hyb id cloud en i onmen s. In J Compu Appl Technol Res. 2020;9(12):438-51.
[34] Kumbankye J. Checks and balances: he ul ima e guide o in e nal con ol sys ems. Feb ua y 2025. ISBN:
9798311897655.
[35] O oko J. Mic oelec onics clean oom design: p ecision ab ica ion o semiconduc o inno a ion, AI, and na ional
secu i y in he U.S. ech sec o . In Res J Mod Eng Technol Sci. 2025;7(2)
[36] Kalejaiye AN. Ad e sa ial machine lea ning o obus cybe secu i y: s eng hening deep neu al a chi ec u es
agains e asion, poisoning, and model-in e ence a acks. In e na ional Jou nal o Compu e Applica ions
Technology and Resea ch. 2024;13(12):72-95.
[37] Ibi oye JS. Modeling Th ea Vec o s in Real-Time Using AI-Enhanced Su eillance Analy ics Ac oss Cybe , Land,
Ai , and Ma i ime Domains. In J Ad Res Publ Re . 2025 Jun;2(6):440-63. doi:
h ps://doi.o g/10.55248/gengpi.6.0625.22102.
[38] Lukyanenko R. Nex Da a Pa adigm: Using AI o Manage All Human Da a Founda ions, A chi ec u e, and
Challenges in Using a Uni e sal AI Da a Manage . A chi ec u e, and Challenges in Using a Uni e sal AI Da a
Manage (Ap il 08, 2025). 2025 Ap 8.
[39] Mammah CU. Risk Asse Po olio Managemen and i s In luence on B anch Pe o mance: E idence om Nige ian
Banks. In J Ad Mul idisc Res S ud. 2023;3(3):1137-45
[40] Skopik F, Naessens V, De Su e B, edi o s. A ailabili y, Reliabili y and Secu i y: ARES 2025 EU P ojec s
Symposium Wo kshops, Ghen , Belgium, Augus 11–14, 2025, P oceedings, Pa I. Sp inge Na u e; 2025 Aug 9.
[41] Ade emi Bunmi Ku elu. 2025. “B idging Ag onomy and Public Heal h: The Role o C op Quali y in Nu i ional
Secu i y”. Cu en Jou nal o Applied Science and Technology 44 (7):114–122.
h ps://doi.o g/10.9734/cjas /2025/ 44i74578.
[42] Tang X, Li Q. Logical Gene Encoding: a Bio-Inspi ed App oach o Ene gy-E icien Au oma ed Reasoning. In5 h
In e na ional Con e ence on In e ne , Educa ion and In o ma ion Technology (IEIT 2025) 2025 Jul 31 (pp. 830-
844). A lan is P ess.
[43] Umako MF. Th ea modelling o a i icial in elligence go e nance: in eg a ing e hical conside a ions in o
ad e sa ial a ack simula ions o c i ical in as uc u e using gene a i e AI. Wo ld J Ad Res Re .
2022;15(2):873-90. doi:10.30574/wja .2022.15.2.0829.
[44] Okuwobi FA, Akomola e OO, Majebi NL. F om Agile Sys ems o Beha io al Heal h: Le e aging Tech Leade ship o
Build Scalable Ca e Models o Child en wi h Au ism. In J Sci Res Compu Sci Eng In Technol. 2023;893. doi:
h ps://doi.o g/10.32628/IJSRCSEIT
GSC Biological and Pha maceu ical Sciences, 2025, 33(01), 001-018
18
[45] Mahama T. S a is ical app oaches o iden i ying eQTLs (exp ession quan i a i e ai loci) in plan and human
genomes. In e na ional Jou nal o Science and Resea ch A chi e. 2023;10(2):1429-1437. doi:
h ps://doi.o g/10.30574/ijs a.2023.10.2.0998
[46] Ka hick M. Secu e Heal hca e Analy ics h ough Fede a ed Cloud In elligence. In e na ional Jou nal.
2024;9(1):1-8.
[47] Ukaoha C. Ta i Policies, Animal Disease Risks, and Food Secu i y: A Compa a i e Simula ion o Wes A ican and
U.S. Ag icul u al Sys ems. GSC Biol Pha m Sci. 2024;29(3):411-27. doi:
h ps://doi.o g/10.30574/gscbps.2024.29.3.0507
[48] Edwa ds DJ. A unc ional con ex ual, obse e -cen ic, quan um mechanical, and neu o-symbolic app oach o
sol ing he alignmen p oblem o a i icial gene al in elligence: sa e AI h ough in e sec ing compu a ional
psychological neu oscience and LLM a chi ec u e o eme gen heo y o mind. F on ie s in Compu a ional
Neu oscience. 2024 Aug 8;18:1395901.
[49] Ibi oye JS. Mul i-Agen AI Sys ems o Secu e, T anspa en , and Complian F aud Su eillance in C oss-Bo de
FinTech Ope a ions. In J Res Publ Re . 2025 Jun;6(6):9724-40. doi:
h ps://doi.o g/10.55248/gengpi.6.0625.22103.
[50] Alwakeel M. Neu o-D i en Agen -Based Secu i y o Quan um-Sa e 6G Ne wo ks. Ma hema ics. 2025 Jun
23;13(13):2074.
[51] Nso M. P edic i e main enance using machine lea ning o enginee ing sys ems h ough eal- ime senso da a
and anomaly de ec ion models. In J Res Publ Re . 2024 Oc ;5(10):5167-83. doi:
h ps://doi.o g/10.55248/gengpi.6.0725.2541
[52] Okuwobi FA, Akomola e OO, Majebi NL. B idging he au ism ca e gap: How echnology can expand access o ABA
he apy in unde se ed communi ies. Gyanshau yam In Sci Re Res J. 2024;7(5):103-40.
