Nikola Ilić

Dr. Nikola Ilić

Clinical Genetics, Mother and Child Health Care Institute of Serbia, Belgrade, Beograd, Serbia
orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2025 in Journal of Personalized Medicine
Verified
Journal Article
Open access
PDF file
NINikola Ilić
Nikola Ilić
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
Background: Artificial intelligence (AI) is increasingly applied in the diagnosis of pediatric rare diseases, enhancing the speed, accuracy, and accessibility of genetic interpretation. These advances support the ongoing shift toward personalized medicine in clinical genetics. Objective: This review examines current applications of AI in pediatric rare disease diagnostics, with a particular focus on real-world data integration and implications for individualized care. Methods: A narrative review was conducted covering AI tools for variant prioritization, phenotype–genotype correlations, large language models (LLMs), and ethical considerations. The literature was identified through PubMed, Scopus, and Web of Science up to July 2025, with priority given to studies published in the last seven years. Results: AI platforms provide support for genomic interpretation, particularly within structured diagnostic workflows. Tools integrating Human Phenotype Ontology (HPO)-based inputs and LLMs facilitate phenotype matching and enable reverse phenotyping. The use of real-world data enhances the applicability of AI in complex and heterogeneous clinical scenarios. However, major challenges persist, including data standardization, model interpretability, workflow integration, and algorithmic bias. Conclusions: AI has the potential to advance earlier and more personalized diagnostics for children with rare diseases. Achieving this requires multidisciplinary collaboration and careful attention to clinical, technical, and ethical considerations.

13 November 2024 in Genes
Verified
Journal Article
Open access
PDF file
NINikola Ilić
Nikola Ilić
SKStasa Krasic
Stasa Krasic
NMNina Maric
Nina Maric
VGVladimir Gašić
Vladimir Gašić
JKJovana Krstic
Jovana Krstic
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
Background: Noonan syndrome (NS) is a congenital genetic disorder with a prevalence of 1 in 1000 to 2500 live births, and is characterized by distinctive facial features, short stature, chest deformities, and congenital heart disease. This study aims to evaluate the prevalence of specific genetic mutations and their impact on cardiovascular and other outcomes in NS. Methods: We conducted a retrospective clinical study of 25 pediatric patients diagnosed with NS at two institutions: The Mother and Child Health Care Institute of Serbia and the Clinic for Children Diseases, University Clinical Center of the Republic of Srpska. Patients underwent whole-exome sequencing (WES) to identify genetic mutations. Clinical data, including cardiovascular manifestations, psychomotor development, and stature, were analyzed in relation to mutation types. Results: The cohort comprised 60% male and 40% female patients, with a median age at diagnosis of 7.2 years. Cardiovascular abnormalities were present in 88% of patients. Mutations in PTPN11 were most commonly associated with pulmonary valve stenosis (PVS), while RAF1 mutations were prevalent in patients with hypertrophic cardiomyopathy (HCM). No significant association was found between cardiac disease and delayed psychomotor development (p = 0.755), even though the likelihood ratio showed significance in that regard (p = 0.018). Short stature was observed in 48% of patients but was not significantly correlated with genetic type of disease, presence of cardiac disease, or developmental delay. Conclusions: The study confirms the high prevalence of cardiovascular manifestations in NS and highlights genotype–phenotype correlations. While cardiac abnormalities are common, their impact on psychomotor development and stature is less clear. Further research is needed to explore genetic interactions influencing these outcomes and refine clinical management strategies.

1 March 2025 in Neuroglia
Verified
Journal Article
Open access
PDF file
NINikola Ilić
Nikola Ilić
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
Background/Objectives: Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition marked by challenges in social communication, restricted interests, and repetitive behaviors. Recent studies highlight the crucial roles of neuroglial cells—astrocytes, microglia, and oligodendrocytes—in synaptic function, neural connectivity, and neuroinflammation. These findings offer a fresh perspective on ASD pathophysiology. This review synthesizes current knowledge on neuroglial dysfunction in ASD, emphasizing its role in pathophysiological mechanisms, genetic influences, and potential therapeutic strategies. Methods: We conducted a comprehensive literature review, integrating insights from neuroscience, molecular biology, and clinical studies. Special focus was given to glial-mediated neuroinflammatory mechanisms, synaptic plasticity regulation, and the impact of genetic mutations on neuroglial signaling and homeostasis. Results: Neuroglial dysfunction in ASD is evident in abnormal synaptic pruning by microglia, impaired astrocytic glutamate regulation, and defective oligodendrocyte-driven myelination, which collectively disrupt neuronal architecture. Emerging therapies targeting these pathways, including anti-inflammatory drugs, microglial modulators, and cell-based approaches, show promise in alleviating key ASD symptoms. Additionally, advanced interventions such as gene editing and glial progenitor therapy present opportunities to correct underlying neuroglial dysfunction. Conclusions: This review establishes a comprehensive framework for understanding neuroglial contributions to ASD. By integrating insights from diverse disciplines, it enhances our understanding of ASD pathophysiology and paves the way for novel therapeutic strategies targeting neuroglial pathways.

28 June 2025 in Genes
Verified
Journal Article
Open access
PDF file
NINikola Ilić
Nikola Ilić
NMNina Maric
Nina Maric
DCDimitrije Cvetkovic
Dimitrije Cvetkovic
MBMarko Bogosavljević
Marko Bogosavljević
GBGordana Bukara-Radujković
Gordana Bukara-Radujković
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
Background/Objectives: Skeletal dysplasias are a heterogeneous group of rare genetic disorders with diverse and overlapping clinical presentations, posing diagnostic challenges even for experienced clinicians. With the increasing availability of artificial intelligence (AI) in healthcare, large language models (LLMs) offer a novel opportunity to assist in rare disease diagnostics. This study aimed to compare the diagnostic accuracy of two advanced LLMs, ChatGPT (version GPT-4) and DeepSeek, with that of a clinical expert panel in a cohort of pediatric patients with genetically confirmed skeletal dysplasias. Methods: We designed a prospective vignette-based diagnostic benchmarking study including 45 children with confirmed skeletal dysplasias from two tertiary centers. Both LLMs were prompted to provide primary and differential diagnoses based on standardized clinical case vignettes. Their outputs were compared with those of two human experts (a pediatric endocrinologist and a pediatric orthopedic surgeon), using molecular diagnosis as the gold standard. Results: ChatGPT and DeepSeek achieved a comparable diagnostic top-3 accuracy (62.2% and 64.4%, respectively), with a high intermodel agreement (Cohen’s κ = 0.95). The expert panel outperformed both models (82.2%). While LLMs performed well on more common disorders, they struggled with ultra-rare and multisystemic conditions. In one complex case missed by experts, the DeepSeek model successfully proposed the correct diagnosis. Conclusions: LLMs offer a complementary diagnostic value in skeletal dysplasias, especially in under-resourced medical settings. Their integration as a supportive tool in multidisciplinary diagnostic workflows may enhance early recognition and reduce diagnostic delays in rare disease care.

15 June 2024 in Genes
Verified
Journal Article
Open access
PDF file
NINikola Ilić
Nikola Ilić
NMNina Maric
Nina Maric
AMAles Maver
Ales Maver
LALluis Armengol
Lluis Armengol
RKRuzica Kravljanac
Ruzica Kravljanac
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
This study delves into the diagnostic yield of whole-exome sequencing (WES) in pediatric patients presenting with developmental delay/intellectual disability (DD/ID), while also exploring the utility of Reverse Phenotyping (RP) in refining diagnoses. A cohort of 100 pediatric patients underwent WES, yielding a diagnosis in 66% of cases. Notably, RP played a significant role in cases with negative prior genetic testing, underscoring its significance in complex diagnostic scenarios. The study revealed a spectrum of genetic conditions contributing to DD/ID, illustrating the heterogeneity of etiological factors. Despite challenges, WES demonstrated effectiveness, particularly in cases with metabolic abnormalities. Reverse phenotyping was indicated in half of the patients with positive WES findings. Neural network models exhibited moderate-to-exceptional predictive abilities for aiding in patient selection for WES and RP. These findings emphasize the importance of employing comprehensive genetic approaches and RP in unraveling the genetic underpinnings of DD/ID, thereby facilitating personalized management and genetic counseling for affected individuals and families. This research contributes insights into the genetic landscape of DD/ID, enhancing our understanding and guiding clinical practice in this particular field of clinical genetics.

17 July 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
NINikola Ilić
Nikola Ilić
Adrijan SarajlijaAdrijan Sarajlija
Adrijan Sarajlija
Background: Artificial intelligence (AI) is increasingly used in the diagnosis of pediatric rare diseases, helping to improve the speed, accuracy, and accessibility of genetic interpretations. This development supports the ongoing shift toward personalized medicine in clinical genetics. Objective: This review explores current applications of AI in pediatric rare disease diagnostics, with an emphasis on real-world data integration and implications for individualized care. Methods: A narrative review was conducted covering AI tools for variant prioritization, phenotype-genotype correlation, large language models (LLMs), and ethical considerations. Literature was identified through PubMed, Scopus, and Web of Science up to July 2025. Results: AI platforms offer promising support for genomic interpretation, especially in structured diagnostic workflows. Tools integrating HPO-based inputs and LLMs enable phenotype matching and support reverse phenotyping. Real-world data enhance AI’s applicability in complex, heterogeneous cases. However, challenges remain regarding data standardization, interpretability, workflow integration, and bias. Conclusion: AI has the potential to support earlier and more personalized diagnostics for children with rare diseases. To fully realize this, multidisciplinary collaboration and careful attention to clinical, technical, and ethical considerations are essential.