Ioannis Nikolakakis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Ioannis Nikolakakis has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Spyros N. Michaleas
Department of History of Medicine and Medical Ethics, Medical School, National and Kapodistrian University of Athens, Athens, Greece
2 shared publications - Camelia Cristina Diaconu
Faculty of Medicine, University of Medicine and Pharmacy Carol Davila Bucharest, 050474 Bucharest, Romania
1 shared publication E. Artemi
1 shared publicationI. Dimitriadis
1 shared publication- Konstantinos Karamouzis
Department of Clinical Therapeutics, School of Medicine, National and Kapodistrian University of Athens, 11528 Athens, Greece
1 shared publication - Konstantinos Laios
National and Kapodistrian University of Athens
1 shared publication
Latest Publications
- Open Access
- 20 November 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (22) , 9091
Large Language Models Are Not Interchangeable for Empiric Antibiotic Recommendations Under a Stewardship Benchmark: A Paired Retrospective Multi-Model Study
Background/Objectives: Large language models (LLMs) are increasingly evaluated for empiric antibiotic decision support, but it remains unclear whether contemporary systems behave similarly under the same clinical and stewardship constraints. We compared four LLM configurations on real internal medicine admissions using a common, auditable stewardship benchmark. Methods: This single-center, retrospective, paired, non-interventional shadow-mode study included 486 admissions from a previously published 493-admission internal medicine cohort. For each admission, the clinician-prescribed empiric regimen was compared with recommendations from OpenAI GPT-5.5, Anthropic Claude Opus 4.7, Google Gemini 2.5 Pro, and Mistral Large 2512; OpenAI GPT-5.2 was analyzed separately as a supplementary legacy bridge. All LLMs received identical clinical content, local epidemiology context, antibiotic ontology, and three-field output schema. The primary endpoint was any contextually unsupported broad-spectrum escalation, defined by the study’s prespecified guardrail rules; the key secondary endpoint was a weighted contextual escalation penalty. Results: Clinician regimens met the primary endpoint in 16.7% of admissions. Rates were 13.0% for GPT-5.5 (paired risk difference [RD] −3.7 percentage points; Holm-adjusted p = 0.216), 15.0% for Gemini 2.5 Pro (RD −1.6; Holm-adjusted p = 0.509), 29.8% for Claude Opus 4.7 (RD +13.2; Holm-adjusted p < 0.001), and 31.9% for Mistral Large 2512 (RD +15.2; Holm-adjusted p < 0.001). Mean weighted contextual escalation penalty was 0.288 for clinician prescribing, 0.173 for GPT-5.5, 0.165 for Gemini, 0.510 for Claude, and 0.348 for Mistral. Global heterogeneity across contemporary LLMs was significant for both endpoints (p < 0.001). Conclusions: Contemporary LLM configurations were not interchangeable under the same stewardship benchmark. The benchmark measures stewardship-process performance (contextually unsupported escalation), not microbiological adequacy or missed coverage; fewer violations should therefore not be read as better overall therapy. These findings support configuration-specific, version-locked local evaluation before prospective shadow-mode and human-in-the-loop assessment.
- 20 November 2026
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 15 (22) , 9091
Open AccessLarge Language Models Are Not Interchangeable for Empiric Antibiotic Recommendations Under a Stewardship Benchmark: A Paired Retrospective Multi-Model Study
Background/Objectives: Large language models (LLMs) are increasingly evaluated for empiric antibiotic decision support, but it remains unclear whether contemporary systems behave similarly under the same clinical and stewardship constraints. We compared four LLM configurations on real internal medicine admissions using a common, auditable stewardship benchmark. Methods: This single-center, retrospective, paired, non-interventional shadow-mode study included 486 admissions from a previously published 493-admission internal medicine cohort. For each admission, the clinician-prescribed empiric regimen was compared with recommendations from OpenAI GPT-5.5, Anthropic Claude Opus 4.7, Google Gemini 2.5 Pro, and Mistral Large 2512; OpenAI GPT-5.2 was analyzed separately as a supplementary legacy bridge. All LLMs received identical clinical content, local epidemiology context, antibiotic ontology, and three-field output schema. The primary endpoint was any contextually unsupported broad-spectrum escalation, defined by the study’s prespecified guardrail rules; the key secondary endpoint was a weighted contextual escalation penalty. Results: Clinician regimens met the primary endpoint in 16.7% of admissions. Rates were 13.0% for GPT-5.5 (paired risk difference [RD] −3.7 percentage points; Holm-adjusted p = 0.216), 15.0% for Gemini 2.5 Pro (RD −1.6; Holm-adjusted p = 0.509), 29.8% for Claude Opus 4.7 (RD +13.2; Holm-adjusted p < 0.001), and 31.9% for Mistral Large 2512 (RD +15.2; Holm-adjusted p < 0.001). Mean weighted contextual escalation penalty was 0.288 for clinician prescribing, 0.173 for GPT-5.5, 0.165 for Gemini, 0.510 for Claude, and 0.348 for Mistral. Global heterogeneity across contemporary LLMs was significant for both endpoints (p < 0.001). Conclusions: Contemporary LLM configurations were not interchangeable under the same stewardship benchmark. The benchmark measures stewardship-process performance (contextually unsupported escalation), not microbiological adequacy or missed coverage; fewer violations should therefore not be read as better overall therapy. These findings support configuration-specific, version-locked local evaluation before prospective shadow-mode and human-in-the-loop assessment.
- 30 October 2025
- Published by ALUNA in Wiadomości Lekarskie
- Vol. 78 (10) , 2160-2166
Rabies: Known and Unknown Historical Aspects
Aim: to provide a comprehensive historical overview of rabies, with a focus on its early recognition, cultural significance, and advancements in treatment from antiquity to the modern era. Special emphasis is placed on the discovery of the rabies virus and the development of the first vaccine in 1885 by Louis Pasteur and Émile Roux. Materials and Methods: A literature review was conducted using electronic databases such as Scopus and PubMed to gather data on rabies incidence, transmission patterns, and public health interventions across various Latin American countries. In addition, supplementary searches were carried out in university and public libraries to obtain historical sources and regional data not readily available in digital repositories. Conclusions: Although significant progress has been made in reducing rabies cases in Latin America through widespread vaccination campaigns and animal control strategies, the disease continues to pose a public health concern in certain areas. This persistence is particularly attributed to challenges in rural and remote regions, bat-mediated transmission, and deficiencies in disease surveillance. Continued investment in preventive measures, public education, and targeted interventions—especially among vulnerable populations—is crucial for achieving the complete elimination of rabies in the region.
- 30 October 2025
- Published by ALUNA in Wiadomości Lekarskie
- Vol. 78 (10) , 2160-2166
Rabies: Known and Unknown Historical Aspects
Aim: to provide a comprehensive historical overview of rabies, with a focus on its early recognition, cultural significance, and advancements in treatment from antiquity to the modern era. Special emphasis is placed on the discovery of the rabies virus and the development of the first vaccine in 1885 by Louis Pasteur and Émile Roux. Materials and Methods: A literature review was conducted using electronic databases such as Scopus and PubMed to gather data on rabies incidence, transmission patterns, and public health interventions across various Latin American countries. In addition, supplementary searches were carried out in university and public libraries to obtain historical sources and regional data not readily available in digital repositories. Conclusions: Although significant progress has been made in reducing rabies cases in Latin America through widespread vaccination campaigns and animal control strategies, the disease continues to pose a public health concern in certain areas. This persistence is particularly attributed to challenges in rural and remote regions, bat-mediated transmission, and deficiencies in disease surveillance. Continued investment in preventive measures, public education, and targeted interventions—especially among vulnerable populations—is crucial for achieving the complete elimination of rabies in the region.
- 27 March 2025
- Published by Elsevier in Ethics, Medicine and Public Health
- Vol. 33
Christian ethos and ethics as presuppositions for organ transplantation
- 27 March 2025
- Published by Elsevier in Ethics, Medicine and Public Health
- Vol. 33
Christian ethos and ethics as presuppositions for organ transplantation