Title
Multiscale model for thrombosis / atherosclerotic plaque formation and progression: doctoral dissertation
Creator
Spahić, Lemana, 1996-
CONOR:
130878729
Copyright date
2024
Object Links
Select license
Autorstvo-Nekomercijalno-Bez prerade 3.0 Srbija (CC BY-NC-ND 3.0)
License description
Dozvoljavate samo preuzimanje i distribuciju dela, ako/dok se pravilno naznačava ime autora, bez ikakvih promena dela i bez prava komercijalnog korišćenja dela. Ova licenca je najstroža CC licenca. Osnovni opis Licence: http://creativecommons.org/licenses/by-nc-nd/3.0/rs/deed.sr_LATN. Sadržaj ugovora u celini: http://creativecommons.org/licenses/by-nc-nd/3.0/rs/legalcode.sr-Latn
Language
Serbian
Cobiss-ID
Inventory ID
3822
Theses Type
Doktorska disertacija
description
Datum odbrane: 05.06.2025.
Other responsibilities
Academic Expertise
Multidisciplinarne i interdisciplinarne naučne oblasti
University
Univerzitet u Kragujevcu
Faculty
Fakultet inženjerskih nauka
Publisher
[L. S. Spahić]
Format
119 listova
Abstract (en)
Cardiovascular diseases, particularly atherosclerosis, remain leading causes of morbidity
and mortality worldwide, necessitating innovative approaches for early detection, risk
stratification, and management. This research explores the application of advanced
computational techniques—artificial intelligence (AI) and agent-based modeling
(ABM)—to address the complexities of atherosclerosis progression. AI, leveraging
machine learning and deep learning algorithms, has demonstrated significant potential in
analyzing large-scale datasets, including electronic health records, medical imaging, and
genetic profiles, to predict disease onset and progression with greater accuracy than
traditional methods. Concurrently, ABM offers insights into the intricate biological
interactions within the cardiovascular system by simulating the behaviors of individual
agents, such as cells and tissues, in response to various stimuli. However, both
methodologies present limitations, including challenges related to data quality, model
interpretability, and the complexity of biological systems.
This research underscores the need for interdisciplinary collaboration between
computational scientists, clinicians, and engineers to refine these models and facilitate
their integration into clinical practice. Sensitivity analysis was conducted on the
developed ABM model and a virtual population was created from the data in order to
develop as surrogate model based on AI. The dataset captured a landscape of patientspecific
variability and provided significant variation for the model to learn. The
surrogate model for atherosclerotic plaque progression was based on artificial neural
networks and deep learning and performed with 95.4% accuracy and congruency with
the ABM indicating its strong potential to be used in practice.
By addressing their inherent limitations, AI and ABM hold the potential to revolutionize
cardiovascular medicine, leading to more personalized and effective treatments. Future
research directions include improving data integration, enhancing model transparency,
and conducting real-world validation studies to translate computational insights into
meaningful clinical outcomes. The findings of this study contribute to the growing body
of evidence supporting the role of ABM andAI surrogate modeling in advancing our
understanding of cardiovascular diseases. The potential of ABM modeling backed with
decreasing of computational resources necessary and enhanced speed of decission
making ensured by surrogate modeling offers promising pathways for better patient care
and disease management.
Abstract (sr)
Kardiovaskularne bolesti, naročito ateroskleroza, i dalje su vodeći uzroci
morbiditeta i mortaliteta širom sveta, što zahteva inovativne pristupe za
rano otkrivanje, stratifikaciju rizika i upravljanje bolestima. Ovo
istraživanje istražuje primenu naprednih računarskih tehnika—veštačke
inteligencije (AI) i modeliranja zasnovanog na agentima (ABM)—u rešavanju
kompleksnosti progresije ateroskleroze. Veštačka inteligencija, koristeći
algoritme mašinskog i dubokog učenja, pokazala je značajan potencijal u analizi
velikih skupova podataka, uključujući elektronske zdravstvene zapise,
medicinske slike i genetske profile, za preciznije predviđanje pojave i
progresije bolesti od tradicionalnih metoda. Istovremeno, ABM pruža uvid u
složene biološke interakcije unutar kardiovaskularnog sistema simulirajući
ponašanje pojedinačnih agenata, poput ćelija i tkiva, kao odgovor na različite
stimuluse. Međutim, obe metodologije imaju ograničenja, uključujući izazove
vezane za kvalitet podataka, interpretabilnost modela i složenost bioloških
sistema.
Ovo istraživanje ističe potrebu za interdisciplinarnom saradnjom između
računarskih naučnika, kliničara i inženjera radi unapređenja ovih modela i
njihove integracije u kliničku praksu. Sprovedena je analiza osetljivosti na
razvijenom ABM modelu, a iz podataka je kreirana virtuelna populacija radi
razvoja surogat modela zasnovanog na veštačkoj inteligenciji. Skup podataka je
obuhvatio spektar varijabilnosti specifične za pacijente i obezbedio značajnu
varijaciju za učenje modela. Surogat model za progresiju aterosklerotičnih
plakova zasnovan je na veštačkim neuronskim mrežama i dubokom učenju i
postigao je tačnost od 95.4% i usklađenost sa ABM, što ukazuje na njegov veliki
potencijal za praktičnu primenu.
Uz prevazilaženje urođenih ograničenja, AI i ABM imaju potencijal da
revolucionišu kardiovaskularnu medicinu, vodeći ka personalizovanijim i
efikasnijim tretmanima. Budući pravci istraživanja uključuju unapređenje
integracije podataka, poboljšanje transparentnosti modela i sprovođenje studija
validizacije u stvarnom svetu radi pretvaranja računarskih uvida u značajne
kliničke rezultate. Nalazi ovog istraživanja doprinose rastućoj bazi dokaza
koji podržavaju ulogu ABM i AI surogat modeliranja u unapređenju našeg
razumevanja kardiovaskularnih bolesti. Potencijal ABM modeliranja, podržan
smanjenjem potrebnih računarskih resursa i ubrzanjem donošenja odluka
zahvaljujući surogat modeliranju, nudi obećavajuće puteve za bolju negu pacijenata
i upravljanje bolestima.
Authors Key words
atherosclerosis, plaque progression, multiscale modeling, agent-based
modeling, artificial intelligence
Authors Key words
ateroskleroza, progresija plaka, višeskalno modeliranje,
modeliranje zasnovano na agentima, veštačka inteligencija
Classification
616.13-004:004.8(043)
Type
Tekst
Abstract (en)
Cardiovascular diseases, particularly atherosclerosis, remain leading causes of morbidity
and mortality worldwide, necessitating innovative approaches for early detection, risk
stratification, and management. This research explores the application of advanced
computational techniques—artificial intelligence (AI) and agent-based modeling
(ABM)—to address the complexities of atherosclerosis progression. AI, leveraging
machine learning and deep learning algorithms, has demonstrated significant potential in
analyzing large-scale datasets, including electronic health records, medical imaging, and
genetic profiles, to predict disease onset and progression with greater accuracy than
traditional methods. Concurrently, ABM offers insights into the intricate biological
interactions within the cardiovascular system by simulating the behaviors of individual
agents, such as cells and tissues, in response to various stimuli. However, both
methodologies present limitations, including challenges related to data quality, model
interpretability, and the complexity of biological systems.
This research underscores the need for interdisciplinary collaboration between
computational scientists, clinicians, and engineers to refine these models and facilitate
their integration into clinical practice. Sensitivity analysis was conducted on the
developed ABM model and a virtual population was created from the data in order to
develop as surrogate model based on AI. The dataset captured a landscape of patientspecific
variability and provided significant variation for the model to learn. The
surrogate model for atherosclerotic plaque progression was based on artificial neural
networks and deep learning and performed with 95.4% accuracy and congruency with
the ABM indicating its strong potential to be used in practice.
By addressing their inherent limitations, AI and ABM hold the potential to revolutionize
cardiovascular medicine, leading to more personalized and effective treatments. Future
research directions include improving data integration, enhancing model transparency,
and conducting real-world validation studies to translate computational insights into
meaningful clinical outcomes. The findings of this study contribute to the growing body
of evidence supporting the role of ABM andAI surrogate modeling in advancing our
understanding of cardiovascular diseases. The potential of ABM modeling backed with
decreasing of computational resources necessary and enhanced speed of decission
making ensured by surrogate modeling offers promising pathways for better patient care
and disease management.
“Data exchange” service offers individual users metadata transfer in several different formats. Citation formats are offered for transfers in texts as for the transfer into internet pages. Citation formats include permanent links that guarantee access to cited sources. For use are commonly structured metadata schemes : Dublin Core xml and ETUB-MS xml, local adaptation of international ETD-MS scheme intended for use in academic documents.

