Stefan Agoston

Mr. Stefan Agoston


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
17 May 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
DADavid Andras
David Andras
RIRadu Alexandru Ilies
Radu Alexandru Ilies
VEVictor Esanu
Victor Esanu
SAStefan Agoston
Stefan Agoston
TJTudor Florin Marginean Jumate
Tudor Florin Marginean Jumate
GDGeorge Calin Dindelegan
George Calin Dindelegan
Background/Objectives: Breast cancer is the most common malignancy among women globally, with an increasing incidence, particularly in younger populations. Achieving complete surgical excision is essential to reduce recurrence. Artificial intelligence (AI), including large language models like ChatGPT, has potential for supporting diagnostic tasks, though its role in surgical oncology remains limited. Methods: This retrospective study evaluated ChatGPT’s performance (ChatGPT-4, OpenAI, March 2025) in predicting surgical margin status (R0 or R1) based on intraoperative mammograms of lumpectomy specimens. AI-generated responses were compared with histopathological findings. Performance was evaluated using sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), F1 score, and Cohen’s kappa coefficient. Results: Out of a total of 100 patients, ChatGPT achieved an accuracy of 84.0% in predicting surgical margin status. Sensitivity for identifying R1 cases (incomplete excision) was 60.0%, while specificity for R0 (complete excision) was 86.7%. The positive predictive value (PPV) was 33.3%, and the negative predictive value (NPV) was 95.1%. The F1 score for R1 classification was 0.43, and Cohen’s kappa coefficient was 0.34, indicating moderate agreement with histopathological findings. Conclusions: ChatGPT demonstrated moderate accuracy in confirming complete excision but showed limited reliability in identifying incomplete margins. While promising, these findings emphasize the need for domain-specific training and further validation before such models can be implemented in clinical breast cancer workflows.

Verified
Journal Article
Open access
PDF file
AOAlexia Onaciu
Alexia Onaciu
AGAlina Grama
Alina Grama
SAStefan Agoston
Stefan Agoston
AMAlexandra Mititelu
Alexandra Mititelu
BMBianca Raluca Maris
Bianca Raluca Maris
Tudor Lucian PopTudor Lucian Pop
Tudor Lucian Pop
Drug-induced liver injury (DILI) remains one of the most challenging adverse drug reactions in clinical practice, particularly in its idiosyncratic form, which is not dose-dependent and is largely driven by host-specific immune and genetic factors. Recent genomic studies have revealed strong associations between certain human leukocyte antigen (HLA) alleles and susceptibility to DILI, supporting an immunogenetic mechanism in which drug or metabolite–protein adducts act as neoantigens, triggering aberrant T-cell activation and hepatocellular injury. This review summarizes current evidence on the contribution of HLA polymorphisms to the pathogenesis of idiosyncratic DILI, highlighting allele-specific risk patterns, such as HLA-B*57:01 associated with flucloxacillin, HLA-DRB1*15:01–DQB1*06:02 in amoxicillin–clavulanate, and HLA-B*35:02 in minocycline-induced liver injury. Furthermore, ethnic variability and allele-haplotype interactions are discussed as potential modulators of susceptibility and clinical phenotype. By integrating genetic and immunological insights, the identification of HLA signatures offers promising tools for precision medicine, enabling earlier identification of at-risk individuals and improved prevention of severe hepatotoxic reactions.

2 May 2026 in Medical Sciences
Verified
Journal Article
Open access
PDF file
Victor EsanuVictor Esanu
Victor Esanu
Horațiu Alexandru ColosiHorațiu Alexandru Colosi
Horațiu Alexandru Colosi
SAStefan Agoston
Stefan Agoston
EMElisa Marziali
Elisa Marziali
RIRadu Alexandru Ilies
Radu Alexandru Ilies
VVVictor Volovici
Victor Volovici
Background/Objectives: Objective end-product assessment of microsurgical anastomoses is intensive and partly subjective. Vision-capable large language models (LLMs) may enable standardized image-based scoring, but their agreement with expert assessment remains uncertain. Methods: We studied 200 end-to-end femoral artery anastomoses, performed on chicken legs by novice, intermediate, and experienced microsurgeons. Images were scored independently by two blinded expert panels; disagreements were adjudicated by a third senior reviewer to establish expert consensus. Two LLMs, ChatGPT 5.2 Thinking Extended and Gemini 3.1 Pro, were evaluated using the exact same prompt and rubric. Each image was analyzed three times per model. Final scores were aggregated by median for numeric items and majority vote for categorical items. The primary endpoint was exact-match agreement with expert consensus. Agreement within ±1 was also assessed for numeric items. Agreement was measured using simple percentage agreement, Light’s kappa, and Krippendorff’s alpha; Bland–Altman analysis was used for numeric count items. Results: LLM 1 achieved a higher overall exact-match agreement than LLM 2 (0.659 vs. 0.539). Both models performed better on categorical than numeric items (0.713 vs. 0.610 and 0.651 vs. 0.445, respectively). LLM 1 showed the greatest advantages for gaps, knots, oblique stitches, and wide bites. Krippendorff’s alpha was positive for most endpoints with LLM 1, whereas LLM 2 showed negative values throughout. Allowing a ±1 tolerance for numeric items greatly improved agreement, suggesting only minor counting discrepancies, from 0.610 to 0.900 for LLM 1 and from 0.445 to 0.826 for LLM 2. Conclusions: Under a constrained scoring workflow, LLMs partially approximated intraluminal microsurgical end-product scoring. LLM 1 outperformed LLM 2, but agreement remained insufficient to replace the expert assessment entirely. These models can be assistive tools within a human-in-the-loop framework.

17 May 2026 in Cells
Verified
Journal Article
Open access
PDF file
SAStefan Agoston
Stefan Agoston
AGAlina Grama
Alina Grama
AOAlexia Onaciu
Alexia Onaciu
AMAlexandra Mititelu
Alexandra Mititelu
GBGabriel Benţa
Gabriel Benţa
Tudor Lucian PopTudor Lucian Pop
Tudor Lucian Pop
Autoimmune hepatitis (AIH) is a progressive inflammatory liver disease characterized by hypergammaglobulinemia, circulating antibodies, and distinctive histological features, with a higher prevalence in females. Immune responses targeting hepatic antigens are considered the main mechanism behind AIH. Many cytokines are involved in the inflammatory response typical of this disease. Interleukin 17 (IL-17) is a powerful pro-inflammatory protein that serves as a key link between the innate and adaptive immune systems. It plays an important role in regulating the inflammatory response in various tissues, including the liver. Several studies have shown that increased IL-17 levels are associated with the severity and progression of AIH. This review explores IL-17’s role in the AIH inflammatory pathway and summarizes existing evidence linking it to liver damage. We also highlight the potential of future therapies targeting this cytokine.