*Co esponding au ho : Benjamin I ebhude Esangbedo
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.
Machine Lea ning Applica ions in Mala ia Elimina ion P og ams: Compa ing Vec o
Con ol S a egies Ac oss Wes A ica and Fo me Endemic Regions in he Sou he n
Uni ed S a es
Benjamin I ebhude Esangbedo 1, , Des iny E uemulo Nwanjo 2, Benedic a Esangbedo 3 , Dolly Nwanjo 4 and
G ace Nwachukwu 5
1 Depa men o Public heal h, Mon oe Uni e si y, New Rochelle, New Yo k, USA.
2 Depa men o Nu sing and Heal h P o essions, Geo gia S a e Uni e si y, A lan a, Geo gia, USA.
3 Depa men o Nu sing Educa ion, Massasoi Communi y College, B ock on, MA, USA.
4 Depa men o Nu sing, Beacon Hill Academy, Fo Laude dale, Flo ida, USA.
5 G adua e School o Geog aphy, Cla k Uni e si y, Wo ces e , MA, USA.
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
Publica ion his o y: Recei ed on 18 Augus 2025; e ised on 23 Sep embe 2025; accep ed on 26 Sep embe 2025
A icle DOI: h ps://doi.o g/10.30574/gscbps.2025.32.3.0371
Abs ac
The applica ion o machine lea ning (ML) echnologies in mala ia elimina ion p og ams ep esen s a pa adigm shi in
ec o -bo ne disease con ol s a egies. This e iew examines he compa a i e implemen a ion o ML-based
app oaches in ec o con ol p og ams ac oss Wes A ica and his o ically endemic egions in he Sou he n Uni ed
S a es. Th ough sys ema ic analysis o ecen li e a u e, we e alua e he e ec i eness o ML algo i hms including
suppo ec o machines, andom o es s, deep lea ning models, and p edic i e analy ics in mala ia ec o su eillance
and con ol. Ou indings e eal ha while Wes A ican p og ams le e age ML p ima ily o ou b eak p edic ion and
ec o habi a mapping using d one image y and en i onmen al da a, he his o ical elimina ion success in he Sou he n
Uni ed S a es p o ides aluable lessons o con empo a y ML-enhanced p og ams. The e iew demons a es ha
machine lea ning models such as suppo ec o machines, decision ees, andom o es s, Ex eme G adien Boos ing,
logis ic eg ession, K-Nea es Neighbou s, Naï e Bayes, and mul ilaye pe cep on ha e been g ea ly used o p edic
mala ia using socioeconomic and en i onmen al a iables. Cu en applica ions show p omise in d one image y and
deep lea ning analysis o a ge ed ec o su eillance, enabling mo e p ecise iden i ica ion o mosqui o b eeding si es.
This compa a i e analysis highligh s he e olu ion om adi ional ec o con ol me hods o sophis ica ed ML-d i en
app oaches, o e ing insigh s o op imizing u u e mala ia elimina ion s a egies in endemic egions.
Keywo ds: Machine lea ning; Mala ia elimina ion; Vec o con ol; Wes A ica; Sou he n Uni ed S a es; Anopheles
su eillance; P edic i e modelling
1. In oduc ion
Mala ia con inues o pose one o he mos p essing public heal h challenges globally, wi h Sub-Saha an A ica
shoulde ing he g ea es bu den o disease ansmission (Sak i e al., 2025). The Wo ld Heal h O ganiza ion's global
mala ia elimina ion ini ia i e has accele a ed e o s o explo e inno a i e ec o con ol me hods, wi h machine
lea ning eme ging as a game-changing echnology o augmen adi ional in e en ion s a egies (T ujillano e al.,
2023). The in eg a ion o a i icial in elligence (AI) and machine lea ning algo i hms in o mala ia con ol p og ams
ma ks a pi o al shi om adi ional su eillance echniques o da a-d i en, p edic i e models capable o op imizing
esou ce alloca ion and imp o ing he iming o in e en ions (Ha dy e al., 2022; Golumbeanu e al., 2021).
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
269
Figu e 1 Global mala ia incidence ends by egion om 2000 o 2020. The cha shows a signi ican educ ion in
cases, pa icula ly in A ica, while egions such as he Ame icas, Wes e n Paci ic, Eas e n Medi e anean, and Sou h-
Eas Asia ha e also seen s eady declines
The his o ical con ex o mala ia elimina ion o e s aluable insigh s in o he de elopmen o ec o con ol s a egies.
In he Uni ed S a es, he Na ional Mala ia E adica ion P og am (NMEP) was ini ia ed in July 1947. By 1951, his ede al
p og am, wi h con ibu ions om s a e and local en i ies, had success ully educed mala ia incidence o he ex en ha
he p og am was concluded. This signi ican achie emen was p ima ily due o he sys ema ic applica ion o DDT
sp aying, d ainage ini ia i es, and obus su eillance sys ems (Johnson, 1965; Raj anshi e al., 2019). The success o
he NMEP, pa icula ly in he Sou he n s a es whe e mala ia had been his o ically endemic, p o ides c ucial lessons o
mode n machine lea ning-enhanced p og ams aimed a mala ia con ol in Wes A ica (Gueye e al., 2016).
Figu e 2 Global mala ia dea h a e pe 100,000 popula ion as o he WHO Wo ld Mala ia Repo 2012. The map
highligh s he highes mo ali y a es in sub-Saha an A ica, wi h a ying le els ac oss o he egions, e lec ing he
une en bu den o mala ia globally
Con empo a y mala ia con ol p og ams in Wes A ica ace a ange o challenges dis inc om hose encoun e ed
du ing he mid-20 h cen u y e o s in he Uni ed S a es. Vec o con ol emains he co ne s one o mala ia p e en ion,
con ibu ing o a no able 65% educ ion in mala ia cases be ween 2000 and 2015 (Tacone e al., 2021). Howe e , he
eme gence o insec icide esis ance, he impac o clima e change on ec o b eeding pa e ns, and he complexi ies o
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
270
socioeconomic ac o s necessi a e mo e ad anced analy ical me hods han we e a ailable du ing he Ame ican mala ia
elimina ion campaign (Ib ahim e al., 2024).
Machine lea ning (ML) applica ions in mala ia con ol ha e apidly e ol ed, encompassing p edic i e modeling o
ou b eak o ecas ing, compu e ision o ec o iden i ica ion, geospa ial analysis o habi a mapping, and
op imiza ion algo i hms o in e en ion planning (Singh and Sa an, 2024). Recen s udies ha e demons a ed he
abili y o ML algo i hms o p ocess complex, mul i-dimensional da ase s, inco po a ing a iables such as clima e,
socioeconomic indica o s, ec o su eillance da a, and epidemiological pa e ns o gene a e ac ionable insigh s o
mala ia con ol p og ams (Hancock e al., 2020; Wamalwa e al., 2024).
The compa a i e analysis be ween con empo a y Wes A ican p og ams and his o ical mala ia con ol e o s in he
Sou he n Uni ed S a es e eals bo h con inui ies and inno a ions. While he undamen al p inciples o in eg a ed ec o
managemen emain unchanged, he applica ion o ML echnologies now enables unp eceden ed p ecision in a ge ing
in e en ions, p edic ing ou b eak isks, and moni o ing p og am e ec i eness in eal- ime (Mbunge and Sibiya, 2023).
This e iew syn hesizes cu en e idence on ML applica ions in mala ia elimina ion p og ams, compa ing
con empo a y app oaches in Wes A ica wi h lessons d awn om he his o ical success o he Sou he n Uni ed S a es'
elimina ion e o s. The analysis aims o iden i y bes p ac ices, highligh echnological inno a ions, and p o ide
ecommenda ions o op imizing ML-enhanced ec o con ol s a egies in mala ia elimina ion p og ams.
2. Me hods
2.1. Li e a u e Sea ch S a egy
A comp ehensi e li e a u e sea ch was conduc ed using mul iple academic da abases, including PubMed, Google
Schola , IEEE Xplo e, and specialized public heal h eposi o ies. The sea ch s a egy employed bo h Medical Subjec
Headings (MeSH) e ms and ee- ex keywo ds o cap u e ele an publica ions on machine lea ning applica ions in
mala ia ec o con ol. P ima y sea ch e ms included: "machine lea ning," "mala ia elimina ion," " ec o con ol,"
"Anopheles su eillance," "Wes A ica mala ia," "p edic i e modelling," and "Sou he n Uni ed S a es mala ia his o y"
(Egbuna e al., 2025; Nki uka e al., 2021; Hancock e al., 2020; Mbunge and Sibiya, 2023).
2.2. Inclusion and Exclusion C i e ia
S udies we e included i hey: (a) desc ibed machine lea ning applica ions in mala ia ec o con ol o elimina ion
p og ams; (b) p o ided da a on ec o su eillance echnologies; (c) examined mala ia elimina ion s a egies in Wes
A ica o his o ical p og ams in he Sou he n Uni ed S a es; (d) we e published in pee - e iewed jou nals o epu able
academic sou ces be ween 2015-2025; and (e) we e a ailable in English. S udies we e excluded i hey ocused solely
on clinical mala ia diagnosis wi hou ec o con ol componen s, add essed non-Anopheles ec o species, o lacked
su icien me hodological de ail o analysis (Phoobane e al., 2022; Sa i, 2022; Po ami is, 2025).
2.3. Da a Ex ac ion and Analysis
Da a ex ac ion ocused on machine lea ning (ML) algo i hm ypes and pe o mance me ics, ec o con ol
in e en ion s a egies, geog aphical implemen a ion con ex s, p og am ou comes and e ec i eness measu es,
echnological in as uc u e equi emen s, and compa a i e analysis amewo ks. S udies we e ca ego ized based on
geog aphical ocus (Wes A ica s. Sou he n Uni ed S a es), me hodological app oach (p edic i e modelling, compu e
ision, op imiza ion), and implemen a ion scale (local, na ional, egional) (Egbuna e al., 2025; Tembine e al., 2024;
Mosugu, 2024).
2.4. Quali y Assessmen
S udy quali y was assessed using adap ed c i e ia o e iew a icles, including me hodological igo , sample size
adequacy, s a is ical analysis app op ia eness, ep oducibili y o indings, and ele ance o mala ia elimina ion
p og ams (Akinwale e al., 2025; Dhlamini e al., 2025). His o ical analyses we e e alua ed based on documen a ion
quali y, da a a ailabili y, and compa a i e ele ance o con empo a y p og ams (Bu ne e al., 2023; Pa son e al.,
2020).
2.5. Syn hesis App oach
A na a i e syn hesis app oach was employed due o he e ogenei y in s udy designs, machine lea ning (ML) algo i hms,
and ou come measu es ac oss included s udies. Findings we e o ganized hema ically a ound key ML applica ion a eas:
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
271
ou b eak p edic ion, ec o habi a mapping, su eillance op imiza ion, and in e en ion planning. Compa a i e
analysis be ween Wes A ican and Sou he n Uni ed S a es app oaches was s uc u ed a ound simila i ies, di e ences,
and lessons lea ned (Egbuna e al., 2025; Njo oge, 2022; Nkya, 2025).
3. Resul s
3.1. Machine Lea ning Algo i hm Applica ions in Wes A ican Mala ia P og ams
Con empo a y mala ia elimina ion p og ams in Wes A ica ha e inc easingly adop ed di e se machine lea ning (ML)
algo i hms o a ious aspec s o ec o con ol and disease p edic ion. Machine lea ning models such as suppo ec o
machines, decision ees, andom o es s, Ex eme G adien Boos ing, logis ic eg ession, K-Nea es Neighbou s, Naï e
Bayes, and mul ilaye pe cep on ha e been ex ensi ely used o p edic mala ia ac oss mul iple Wes A ican coun ies
including Bu kina Faso, Cô e d'I oi e, Ghana, and Mali (Tshimula e al., 2024).
3.2. P edic i e Modelling o Ou b eak Fo ecas ing
Recen implemen a ions in Wes A ica ha e demons a ed signi ican success in ou b eak p edic ion capabili ies.
Mala ia con inues o pose a g owing h ea o he public heal h and economic g ow h o na ions in he opical and
sub opical pa s o he wo ld, necessi a ing ad anced p edic i e app oaches. S udies om The Gambia and
su ounding egions ha e shown ha ensemble me hods combining mul iple machine lea ning (ML) algo i hms achie e
p edic ion accu acies exceeding 85% o seasonal mala ia ou b eaks when inco po a ing me eo ological,
socioeconomic, and his o ical epidemiological da a (Kapwa a and Geb eslasie, 2015; Dhuguma e al., 2025).
Random Fo es algo i hms ha e p o en pa icula ly e ec i e in Wes A ican con ex s due o hei abili y o handle
missing da a and cap u e non-linea ela ionships be ween en i onmen al a iables and mala ia ansmission isk.
Suppo Vec o Machines ha e shown supe io pe o mance in egions wi h limi ed aining da a, while deep lea ning
app oaches using mul ilaye pe cep ons demons a e he highes accu acy in a eas wi h comp ehensi e su eillance
sys ems and la ge da ase s (Tshimula e al., 2024; Ezugwu and Oyelade, 2023).
3.3. Compu e Vision and Remo e Sensing Applica ions
A signi ican ad ancemen in Wes A ican mala ia p og ams in ol es he in eg a ion o d one image y wi h deep
lea ning algo i hms o ec o habi a iden i ica ion. Disease con ol p og ams a e needed o iden i y he b eeding si es
o mosqui oes, which ansmi mala ia and o he diseases, o a ge in e en ions and iden i y en i onmen al isk
ac o s. The inc easing a ailabili y o e y-high- esolu ion d one da a p o ides new oppo uni ies o ind and
cha ac e ize hese ec o b eeding si es.
Implemen a ion in Bu kina Faso and Cô e d'I oi e has demons a ed ha con olu ional neu al ne wo ks (CNNs) can
achie e o e 90% accu acy in iden i ying po en ial Anopheles b eeding si es om high- esolu ion d one image y. These
sys ems p ocess mul ispec al da a o iden i y wa e bodies, ege a ion pa e ns, and human se lemen cha ac e is ics
ha co ela e wi h mosqui o b eeding habi a sui abili y. The echnology enables a ge ed la icide applica ions and
en i onmen al managemen in e en ions wi h unp eceden ed p ecision (T ujillano e al., 2023; Ca asco-Escoba e
al., 2022).
3.4. Genomic Da a Analysis and Vec o Popula ion Dynamics
Recen ad ances in Wes A ican p og ams include he applica ion o gene a i e machine lea ning models o analysing
mosqui o popula ion gene ics. E o s o con ol he sp ead o mala ia ha e o en ocused on hese ec o s, bu ela i ely
li le is known abou he ela ionships be ween popula ions and species in he Anopheles complex. Unsupe ised
lea ning algo i hms ha e been success ully applied o quan i y gene ic s uc u e o Anopheles gambiae complex
popula ions ac oss Guinea and Bu kina Faso, p o iding insigh s in o ec o mig a ion pa e ns and insec icide
esis ance gene low (Pe ez e al., 2025; Redmond, 2015).
3.5. His o ical Vec o Con ol S a egies in he Sou he n Uni ed S a es
The success ul elimina ion o mala ia om he Sou he n Uni ed S a es p o ides impo an baseline compa isons o
e alua ing con empo a y ML-enhanced p og ams. The Mala ia Con ol in Wa A eas (MCWA) p og am was es ablished
in 1942 o con ol mala ia nea mili a y aining bases in he sou he n Uni ed S a es and i s e i o ies, whe e mala ia
was s ill p oblema ic and posed a h ea o mili a y ec ui s. This ini ia i e ep esen ed one o he mos sys ema ic and
la ge-scale ec o con ol campaigns in mode n his o y, laying he ounda ion o pos -wa public heal h in as uc u e
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
272
and o e ing c ucial lessons o p esen -day p og ams inco po a ing machine lea ning and da a-d i en su eillance
(Caldwell e al., 2021; Ba a a, 2022; Snowden and Bucala, 2014).
3.6. T adi ional Su eillance and In e en ion Me hods
The Ame ican mala ia elimina ion p og am elied on h ee co e s a egies: sys ema ic DDT sp aying campaigns,
en i onmen al managemen h ough d ainage and modi ica ion o wa e bodies, and a obus case su eillance sys em
wi h apid ea men p o ocols. These in eg a ed app oaches led o d ama ic educ ions in mala ia, wi h he disease
e ec i ely elimina ed om he Sou he n Uni ed S a es by he la e 1940s (Humph eys, 1996; Snowden and Bucala,
2014).
The su eillance sys em de eloped du ing he 1940s and 1950s emphasized s anda dized da a collec ion p ocedu es,
consis en moni o ing o ec o popula ions, sys ema ic case in es iga ions, and s ong coo dina ion ac oss local, s a e,
and ede al public heal h agencies. These o ganiza ional and ope a ional p o ocols emain highly ele an oday,
especially as con empo a y machine lea ning–enhanced mala ia su eillance sys ems seek o emula e his s uc u ed,
scalable amewo k in esou ce-limi ed se ings (Tozan e al., 2007; Webb, 2014).
3.7. In as uc u e and Resou ce Requi emen s
The success ul Ame ican mala ia elimina ion p og am equi ed subs an ial in as uc u e in es men s, including
ained pe sonnel o su eillance and in e en ion ac i i ies, eliable anspo a ion ne wo ks o accessing emo e
a eas, labo a o y acili ies o species iden i ica ion and suscep ibili y es ing, and communica ion sys ems o
coo dina ing mul i-ju isdic ional esponses. These in as uc u e equi emen s emain ele an o con empo a y ML
implemen a ions, hough echnological ad ances ha e educed some ba ie s while c ea ing new equi emen s o
compu a ional esou ces and echnical expe ise (Ali, 2024; Okoye, 2024; Tshimula, 2024).
3.8. Compa a i e Analysis o ML Applica ions s. T adi ional App oaches
3.8.1. Accu acy and P ecision Imp o emen s
Con empo a y machine lea ning (ML) applica ions in Wes A ica demons a e signi ican imp o emen s in p edic ion
accu acy compa ed o adi ional s a is ical app oaches. Ensemble ML models consis en ly achie e 80-95% accu acy in
mala ia isk p edic ion, compa ed o 60-75% accu acy using con en ional epidemiological models. Compu e ision
sys ems o habi a iden i ica ion show o e 90% accu acy, compa ed o 70-80% accu acy o adi ional ield-based
assessmen s (Rahman e al., 2023; Mosugu, 2024; Nkya, 2025).
Figu e 3 Accu acy compa ison be ween machine lea ning algo i hms (80-95%) and adi ional models (60-75%) in
p edic ing mala ia isk. The cha demons a es he supe io p edic ion accu acy o ML-based me hods o e
con en ional app oaches
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
273
3.8.2. Resou ce E iciency and Scalabili y
ML-enhanced p og ams demons a e supe io esou ce e iciency in se e al key a eas. P edic i e models enable
p oac i e esou ce alloca ion, educing eme gency esponse cos s by 30-40% compa ed o eac i e app oaches.
Au oma ed habi a iden i ica ion educes ield su ey equi emen s by up o 60% while main aining highe accu acy
le els. Remo e sensing applica ions enable moni o ing o la ge geog aphical a eas wi h ewe pe sonnel equi emen s
(Ja ed e al., 2025; Ayoka and Nnadi, 2025; Al es e al., 2025).
3.8.3. Technological Dependencies and Limi a ions
Wes A ican machine lea ning (ML) implemen a ions ace se e al cons ain s no p esen in his o ical Ame ican
p og ams: dependence on eliable in e ne connec i i y o cloud-based p ocessing, equi emen s o echnical
expe ise in ML algo i hm implemen a ion and main enance, highe ini ial capi al cos s o equipmen and aining, and
challenges wi h da a quali y and s anda diza ion ac oss di e en su eillance sys ems (Kuponiyi and Akomola e, 2024;
Mbunge and Sibiya, 2025). These ac o s c ea e signi ican ba ie s o success ul implemen a ion in esou ce-limi ed
se ings, which mus be add essed o maximize he po en ial o ML in mala ia con ol.
3.8.4. In eg a ion wi h Exis ing Heal h Sys ems
Success ul machine lea ning (ML) implemen a ion equi es ca e ul in eg a ion wi h exis ing heal h sys em
in as uc u e. Wes A ican p og ams ha e achie ed he bes esul s when ML sys ems complemen a he han eplace
adi ional su eillance me hods, pa icula ly in a eas wi h limi ed echnological in as uc u e. Hyb id app oaches ha
combine au oma ed ML analysis wi h human expe alida ion show highe accep ance a es and be e sus ainabili y
compa ed o ully au oma ed sys ems (Folasole, 2023; De ine e al., 2022; Tshimula e al., 2024).
3.9. Pe o mance Me ics and E ec i eness Measu es
3.9.1. Ou b eak P e en ion and Response Time
ML-enhanced p og ams in Wes A ica demons a e signi ican imp o emen s in ou b eak p e en ion capabili ies. Ea ly
wa ning sys ems using ensemble machine lea ning (ML) algo i hms p o ide 2–4-week ad ance no ice o po en ial
ou b eaks compa ed o 1–2-week no ice using adi ional indica o s. Response ime imp o emen s o 40-60% ha e
been documen ed in a eas wi h in eg a ed ML su eillance sys ems (Singh e al., 2025; Ko u e al., 2025; Bayliss e al.,
2023).
Figu e 4 Compa ison o esponse imes be ween ML-enhanced and adi ional sys ems o mala ia ou b eak
p edic ion. The cha shows ha ML sys ems p o ide as e esponse imes, imp o ing ea ly wa ning capabili ies
3.9.2. Vec o Con ol Ta ge ing P ecision
Compu e ision and geospa ial machine lea ning (ML) applica ions enable mo e p ecise ec o con ol a ge ing.
D one-based habi a iden i ica ion sys ems educe unnecessa y la icide applica ions by 50-70%, while main aining
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
274
equi alen o supe io ec o popula ion con ol. GPS-guided in e en ion sys ems using ML isk p edic ions achie e
25-35% be e co e age o high- isk a eas compa ed o ou ine sp aying schedules (Ja ed e al., 2025; Nkya, 2025;
Tembine e al., 2024).
3.9.3. Cos -E ec i eness Analysis
Economic analysis o ML-enhanced p og ams shows mixed esul s depending on implemen a ion scale and
echnological in as uc u e. La ge-scale implemen a ions ( egional o na ional le el) demons a e cos sa ings o 20-
30% compa ed o adi ional p og ams a e ini ial 2–3-yea in es men pe iods. Smalle scale implemen a ions may
equi e 5-7 yea s o achie e cos neu ali y due o highe pe -uni echnology cos s (Golds ein e al., 2023; Tembine e
al., 2024).
Figu e 5 Compa ison o esponse imes be ween ML-enhanced and adi ional sys ems o mala ia ou b eak
p edic ion. The cha shows ha ML sys ems p o ide as e esponse imes, imp o ing ea ly wa ning capabili ies
3.10. Challenges and Limi a ions in Cu en Implemen a ions
3.10.1. Da a Quali y and S anda diza ion Issues
Wes A ican machine lea ning (ML) p og ams ace signi ican challenges ela ed o da a quali y and s anda diza ion.
Inconsis en su eillance p o ocols ac oss di e en heal h acili ies esul in he e ogeneous da ase s ha can educe
ML algo i hm pe o mance. Missing da a a es o 15-30% a e common in ou ine su eillance sys ems, equi ing
sophis ica ed impu a ion me hods ha may in oduce bias (Yasin e al., 2025; Mille e al., 2023).
3.10.2. Technical In as uc u e Requi emen s
Reliable in e ne connec i i y emains a signi ican cons ain o cloud-based ML sys ems in u al Wes A ican
se ings. Powe in as uc u e limi a ions a ec he con inuous ope a ion o su eillance equipmen and da a collec ion
sys ems. Technical expe ise o ML sys em main enance and oubleshoo ing is limi ed in many a eas, c ea ing a need
o u he in es men in human esou ces and echnical aining (Singh and Singh, 2025; Cudjoe and Vi le , 2023).
3.10.3. Cul u al and Social Accep ance Fac o s
Communi y accep ance o d one-based su eillance and au oma ed decision-making sys ems a ies signi ican ly ac oss
di e en cul u al con ex s. Success ul p og ams equi e ex ensi e communi y engagemen and educa ion abou he
bene i s o ML sys ems and p i acy p o ec ions. T adi ional au ho i y s uc u es and decision-making p ocesses mus
be inco po a ed in o ML sys em implemen a ion s a egies o ensu e highe accep ance (Fo nace e al., 2023; Yasin e
al., 2025).
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
275
4. Discussion
The compa a i e analysis o machine lea ning (ML) applica ions in Wes A ican mala ia elimina ion p og ams e sus
his o ical app oaches in he Sou he n Uni ed S a es e eals bo h signi ican ad ances and pe sis en challenges in ec o
con ol s a egies. The e olu ion om adi ional su eillance me hods o sophis ica ed ML-d i en sys ems ep esen s
a undamen al shi in how public heal h p og ams app oach mala ia elimina ion, o e ing unp eceden ed capabili ies
o p edic ion, a ge ing, and op imiza ion o in e en ions (Egbuna e al., 2025; Njo oge, 2022; Nkya, 2025).
4.1. Technological Ad ances and Capabili ies
Con empo a y machine lea ning (ML) applica ions demons a e ema kable imp o emen s in se e al key a eas o
mala ia con ol. The in eg a ion o d one image y and deep lea ning analysis o a ge ed ec o su eillance enables
he iden i ica ion and mapping o mosqui o b eeding si es wi h accu acy le els ha would ha e been impossible using
adi ional ield su ey me hods (Cha u, 2021). This echnological capabili y add esses one o he undamen al
challenges in ec o con ol: he p ecise iden i ica ion and a ge ing o in e en ion si es.
The di e si y o ML algo i hms now applied o mala ia con ol e lec s he sophis ica ion o con empo a y app oaches.
Machine lea ning models such as suppo ec o machines, decision ees, andom o es s, Ex eme G adien Boos ing,
logis ic eg ession, K-Nea es Neighbou s, Naï e Bayes, and mul ilaye pe cep on ha e been ex ensi ely used o p edic
mala ia, each o e ing speci ic ad an ages o di e en aspec s o p og am implemen a ion. This algo i hmic di e si y
enables p og ams o op imize hei app oaches based on local da a a ailabili y, in as uc u e cons ain s, and speci ic
in e en ion objec i es (Tshimula, 2024; Njo oge, 2022).
4.2. Lessons om His o ical Elimina ion Success
The success ul elimina ion o mala ia om he Sou he n Uni ed S a es p o ides impo an insigh s o con empo a y
ML-enhanced p og ams. The Na ional Mala ia E adica ion P og am (NMEP) was launched in July 1947. By 1951, his
ede al p og am, wi h pa icipa ion om s a e and local agencies, had educed he incidence o mala ia in he Uni ed
S a es o he poin ha he p og am was o icially ended. This achie emen was buil on he sys ema ic applica ion o
a ailable echnologies, comp ehensi e su eillance sys ems, and coo dina ed mul i-ju isdic ional implemen a ion
(Alilio e al., 2004; Aje unmobi e al., 2025).
The Ame ican elimina ion p og am's emphasis on sys ema ic da a collec ion, s anda dized p o ocols, and in e -agency
coo dina ion emains highly ele an o con empo a y ML implemen a ions. While he echnological ools ha e
ad anced d ama ically, he undamen al o ganiza ional and ope a ional p inciples es ablished du ing he 1940s-1950s
elimina ion campaign con inue o p o ide a ounda ion o success ul p og am implemen a ion (Wood e al., 2016;
Sebuabe e al., 2024).
4.3. In eg a ion Challenges and Oppo uni ies
Con empo a y ML applica ions ace in eg a ion challenges ha di e signi ican ly om hose encoun e ed in his o ical
p og ams. Wes A ican implemen a ions mus na iga e complex echnological dependencies, including eliable
in e ne connec i i y, powe in as uc u e, and echnical expe ise equi emen s (Sebuabe e al., 2024). Howe e , hese
challenges a e o se by unp eceden ed capabili ies o da a in eg a ion, eal- ime analysis, and adap i e p og am
managemen (Ekundayo, 2025).
The mos success ul con empo a y p og ams demons a e hyb id app oaches ha combine ML capabili ies wi h
adi ional su eillance and in e en ion me hods. This in eg a ion s a egy add esses bo h echnological limi a ions
and communi y accep ance ac o s while maximizing he bene i s o ad anced analy ical capabili ies (Mosugu, 2024;
Wesonga e al., 2020).
4.4. Scalabili y and Sus ainabili y Conside a ions
Scalabili y ep esen s bo h an oppo uni y and a challenge o ML-enhanced mala ia p og ams. While cloud-based ML
sys ems can heo e ically scale o co e la ge geog aphical a eas wi h minimal addi ional in as uc u e, p ac ical
implemen a ion equi es subs an ial in es men s in aining, equipmen , and ongoing echnical suppo (Ekundayo,
2024). The expe ience om his o ical Ame ican p og ams sugges s ha sus ainable elimina ion equi es long- e m
commi men o sys ema ic implemen a ion a he han sho - e m echnological solu ions (Kuponiyi and Akomola e,
2024).
GSC Biological and Pha maceu ical Sciences, 2025, 32(03), 268-280
276
Economic sus ainabili y o ML-enhanced p og ams depends c i ically on he implemen a ion scale and echnological
in as uc u e de elopmen . La ge-scale egional implemen a ions demons a e be e cos -e ec i eness p o iles
compa ed o smalle pilo p ojec s, sugges ing ha coo dina ion ac oss mul iple coun ies o egions may be necessa y
o op imal p og am sus ainabili y (Ogwu and Izah, 2025; Tshimula e al., 2024).
4.5. Fu u e Di ec ions and Inno a ions
The apid e olu ion o machine lea ning (ML) echnologies sugges s se e al p omising di ec ions o u u e mala ia
elimina ion p og ams. Ad ances in edge compu ing may add ess in e ne connec i i y cons ain s by enabling local
p ocessing o su eillance da a (Tshimula e al., 2024). Imp o ed senso echnologies and sa elli e image y could educe
dependence on d one-based da a collec ion while main aining high accu acy le els o habi a iden i ica ion (De Ma co,
2025).
In eg a ion o genomic da a analysis wi h adi ional su eillance sys ems ep esen s a pa icula ly p omising a ea o
u u e de elopmen . ML algo i hms ha e been employed o in e he join e olu iona y his o y o popula ions sampled
in Guinea and Bu kina Faso, Wes A ica, demons a ing he po en ial o unde s and ec o popula ion dynamics and
insec icide esis ance pa e ns a unp eceden ed esolu ion (Ib ahim e al., 2024; Nkya, 2025).
5. Conclusion
This e iew highligh s he signi ican ad ancemen s machine lea ning (ML) b ings o mala ia elimina ion p og ams,
enhancing ec o con ol, ou b eak p edic ion, and esou ce op imiza ion. ML ools, including suppo ec o machines,
decision ees, and deep lea ning, a e e ec i ely in eg a ed wi h d one image y o p ecise su eillance. The success o
he 1947 Na ional Mala ia E adica ion P og am (NMEP) in he Sou he n U.S. p o ides aluable lessons o cu en ML-
enhanced s a egies. Con empo a y p og ams in Wes A ica show p omise when ML complemen s exis ing heal h
sys ems. Key indings include imp o ed p edic ion accu acy, he need o in as uc u e and aining in es men s, and
he impo ance o communi y engagemen . Fu u e esea ch should ocus on o e coming in as uc u e cons ain s and
in eg a ing ML wi h adi ional me hods. ML-enhanced p og ams o e subs an ial imp o emen s, bu ca e ul
in eg a ion and a en ion o local needs a e essen ial o long- e m success.
Compliance wi h e hical s anda ds
The au ho s decla e ha all p ocedu es ollowed in his s udy we e in acco dance wi h he e hical
s anda ds o he ele an ins i u ional and/o na ional esea ch commi ees.
Disclosu e o con lic o in e es
The au ho s decla e ha hey ha e no con lic s o in e es ele an o his s udy.
Re e ences
[1] Abbasi, E. (2025). Eme ging and ansbounda y a bo i al diseases: The ole o insec ec o s in sp ead. Resea ch
Squa e. h ps://www. esea chsqua e.com/a icle/ s-6918720/la es
[2] Aje unmobi, O. A., Abdul-Rahman, T., & Wyss, K. (2025). Imp o ing diagnos ics and su eillance o mala ia among
displaced people in A ica. In e na ional Jou nal o Equi y in Heal h, 9(30).
h ps://link.sp inge .com/a icle/10.1186/s12939-025-02378-6
[3] Akinwale, O. D., Oke eke, P. U., & Akinbowale, B. T. (2025). Assessing he e ec i eness o cu en mala ia con ol
s a egies in Nige ia: A sys ema ic e iew and me a-analysis. In e na ional Jou nal o Medicine and Heal h
De elopmen . h ps://www. esea chga e.ne /p o ile/Akinwale-Damilola-2/publica ion/392658346
[4] Ali, H. (2024). AI o pandemic p epa edness and in ec ious disease su eillance: P edic ing ou b eaks, modeling
ansmission, and op imizing public heal h in e en ions. In e na ional Jou nal o Resea ch and Public Heal h,
15(2), 259. h ps://www. esea chga e.ne /p o ile/Hassan-Ali-293/publica ion/389138161
[5] Alilio, M. S., Bygbje g, I. C., & B eman, J. G. (2004). In ole able Bu den o Mala ia II. Na ional Ins i u es o Heal h.
h ps://www.ncbi.nlm.nih.go /books/NBK3776/? epo = eade
[6] Al es, H., Mahmood, N. H., & López, O. L. A. (2025). A deep di e in o AI in eg a ion and ad anced nanobiosenso
echnologies o enhanced bac e ial in ec ion moni o ing. a Xi p ep in . h ps://a xi .o g/pd /2509.09005
