ARTIFICIAL INTELLIGENCE IN PEDIATRICS AND NEONATOLOGY: PROMPT ENGINEERING AS A NEW INTERDISCIPLINARY COMPETENCY OF THE PHYSICIAN

Abstract

Introduction. Modern medicine operates under conditions of exponential growth in medical data, creating a
fundamentally new type of cognitive burden for physicians and significantly complicating clinical decision-making processes.
The traditional model of professional activity, focused on knowledge accumulation, is becoming insufficient in the context of
continuous information updates. In this setting, artificial intelligence is increasingly viewed not only as a technological tool but as
a new cognitive environment capable of analyzing large datasets, identifying hidden patterns, and supporting clinical reasoning.
At the same time, the effectiveness of its application largely depends on the quality of interaction between the physician and
the algorithm, where prompt engineering plays a central role.
Objective. To substantiate prompt engineering as a new cognitive and professional competency of physicians, to determine
its place within the structure of clinical reasoning, and to assess its role in modern pediatrics and neonatology.
Materials and methods. This study has a narrative review and analytical design. It is based on a systematic analysis of
contemporary scientific publications addressing the application of artificial intelligence in medicine, as well as a synthesis of
the authors’ educational materials (presentation and accompanying commentary) covering historical, clinical, and educational
aspects of digital transformation. In addition, regulatory and legal documents from Ukraine and the European Union governing
the use of artificial intelligence in healthcare were analyzed.
Results and discussion. Artificial intelligence has been shown to be effectively applied in clinical practice for patient stratification,
early detection of neurodevelopmental disorders, and prediction of critical conditions, including neonatal sepsis. Machine
learning algorithms enable the identification of complex multifactorial patterns and allow prediction at the preclinical stage,
thereby opening opportunities for preventive medicine. At the same time, it has been demonstrated that the quality of results
directly depends on the structure, completeness, and precision of formulated prompts. Prompt engineering is conceptualized
as an integrative cognitive process that includes clinical situation analysis, information structuring, task formulation, and
metacognitive control of outputs. Key competencies required for effective interaction with AI systems include clinical literacy,
analytical and critical thinking, digital competence, communicative precision, and the ability to interpret probabilistic models.
Particular attention is given to the interdisciplinary nature of AI application and the need to integrate medical, technical, and
legal perspectives. Ethical and regulatory challenges are also analyzed, including algorithmic bias, limited explainability, data
privacy concerns, and legal responsibility.
Conclusions. Prompt engineering is emerging as a fundamental competency of the modern physician, enabling effective
interaction with artificial intelligence systems and improving the quality of clinical decision-making. Its development facilitates
the transformation of clinical reasoning into a structured digital format, optimizes information management, and supports the
implementation of personalized and preventive medicine. Integrating this competency into medical education is a necessary
condition for adapting physicians to the digital transformation of healthcare.