Intelligent decision support systems for physicians in renal tumors

Introduction . Malignant neoplasms of the kidney remain one of the most significant problems in modern oncourology and pediatric oncology. According to the World Health Organization (2024), more than 430,000 new cases of malignant renal neoplasms are registered worldwide annually, of which approximately 5–7% occur in patients under 18 years of age. The purpose of the study – to demonstrate how intelligent software solutions improve diagnosis and treatment planning in children and adolescents with renal neoplasms, based on clinical cases from real-world practice. Materials and methods . A series of three cases examined and treated at the Stavropol Regional Children’s Clinical Hospital and the N.N. Blokhin National Medical Research Centre of Oncology, Ministry of Health of Russia is presented. Medvision AI (automated segmentation and 3D modeling), standard computer/magnetic resonance imaging protocols, and histological verification were used. Results . Intelligent tools provided more accurate delineation of tumor contours compared with analogs (Viewer “Kometa 3Di PACS”, Radiant), clarified relationships with vascular structures, and enabled quantitative assessment of lesion volume. Preoperative diagnosis fully matched morphological findings in 3 out of 3 cases, without specification of the histological tumor subtype. In one case, 3D planning changed the surgical strategy in favor of organ-sparing resection. No intraoperative complications were observed; visualization reduced the number of intraoperative errors and increased concordance between the surgical plan and the surgeon’s actual actions. Conclusions . Integration of Medvision AI into routine practice is feasible, safe, and clinically beneficial: it assists in determining the extent of surgical intervention and supports personalized decision-making. Prospective multicenter studies are required to quantitatively assess its impact on outcomes and resource efficiency.