Background and Objective: Spina Bifida (SB) is a congenital spinal malformation requiring timely diagnosis. Digitalization of healthcare and rapid development of artificial intelligence technologies (AI) facilitate AI implementation in various fields of medicine, including SB diagnosis. The aim of this study is to analyze the potential of AI for detecting signs of Spina Bifida in fetuses, children, and adults, evaluate existing approaches, and identify key areas for further research and development.

Methods: this review includes original and review publications, conference materials describing the application of AI algorithms to any type of data at any stage of the diagnostic process of SB in fetuses, children, and adults. A search of available literature was conducted in the following databases: PubMed, Google Scholar, and the Russian Science Citation Index (eLibrary.ru).

Results: seventeen publications on the use of AI in the diagnosis of SB were selected, which met the search criteria. Using AI algorithms, the following types of data were identified and analyzed: medical imaging data (magnetic resonance imaging, ultrasound, X-ray, video urodynamics), laboratory diagnostics data, and genetic information.

Discussion: AI algorithms have shown high efficiency in detecting SB and its complications at various stages of the diagnostic process. The prospect of using computer vision to detect SB in images of various modalities as well as machine learning algorithms in laboratory diagnostics and genetic research was demonstrated.

Background. The active implementation of artificial intelligence (AI) in healthcare significantly improves the effectiveness of early diagnosis of various diseases. Promising areas for AI implementation include diagnostic methods based on voice analysis, remote blood flow monitoring using photoplethysmography, eye movement tracking, and the use of smart devices for continuous health monitoring. A key objective is the development of integrated systems for comprehensive medical screening.

Aim. To evaluate the literature and analyze the results of artificial intelligence implementation for the early detection of diseases.

Materials and methods. A search was conducted in the databases PubMed (Medline), Google Scholar, eLibrary, Web of Science, Scopus, CyberLeninka for English- and Russian-language publications. We used the keywords "screening," "diagnostics," "artificial intelligence," "machine learning," "disease," "screening," "diagnostics," "artificial intelligence," "disease," "machine learning," and "deep learning." Studies from 2015 to 2025 were included based on independent review by three researchers who reached a consensus.

Results. Thirty-one papers from 1,141 publications that met the search criteria were included for this review.

Conclusions. AI-based disease screening and diagnosis provides valuable information about a patient's condition, reducing the risk of human error and missing early signs of disease.

Significance. The introduction of digital technologies into the healthcare system has a profound and multidimensional impact on its most valuable and vulnerable component - human resources. Digital transformation is changing the landscape of professional activity, leading to the emergence of fundamentally new specialties and requirements for competencies and skills of medical professionals.

Aim: to systematize current data on the impact of digitalization on the demand for medical personnel, the changing professional landscape, and the requirements for the competencies of medical professionals.

Materials and methods. We systematically analyzed international publications retrieved from Scopus, PubMed, and Google Scholar using the search terms "digital technologies," "artificial intelligence," or "telemedicine," "workforce," and "healthcare.". All types of studies assessing the impact of digital technologies (artificial intelligence, telemedicine, robotics, the Internet of Medical Things, and big data analysis) on the workload of medical staff were included.

Results. The review includes 61 international publications, which indicate that, as of the end of 2025, digital technologies do not offer the global healthcare system a solution to the workforce shortage. Instead, they provide a set of tools enabling it to function more efficiently, sustainably, and efficiently, even in the face of an objective global human resource shortage. Therefore, digitalization is considered a key tool in mitigating the consequences of the global workforce shortage in the majority of publications included in the review.

Conclusion. The review highlights the potential for a profound transformation of the medical profession landscape. physicians and nurses are evolving toward data management, critical interpretation, and enhanced patient engagement. There is a strong demand for fundamentally new hybrid professions at the intersection of medicine, information technology, and data science (bioinformaticians, medical software developers, cybersecurity specialists). Realizing the potential of digitalization requires overcoming regulatory barriers and significantly transforming the medical education system.

Early diagnosis of skin cancer remains a pressing issue in modern medicine. Most clinical approaches aimed at screening for pigmented and amelanotic basal cell carcinoma are considered ineffective. Computer vision and machine learning offer new opportunities for developing effective methods for detecting suspicious skin lesions. However, most algorithms developed in this area have limited evidence due to the lack of external prospective validation.

The aim of this study was to externally validate an artificial intelligence algorithm for assessing the malignancy risk of pigmented and amelanotic skin lesions based on photographs. The app's target audience is patients. It can also be used by clinicians, such as dermatologists, during initial consultations. The study involved 132 patients complaining of localized skin lesions. Histological and cytological results were used as the reference test, while the "ProRodinki" app served as the index test. The algorithm demonstrated high sensitivity of 86% in diagnosing basal cell carcinoma, but its specificity was 41%.

To improve the algorithm's specificity, retraining using sampling, hyperparameter optimization, and data dimensionality reduction is advisable. Automatic image quality assessment could also be used to exclude images of unsatisfactory quality. This study highlights the importance of validating developed algorithms aimed at classifying tumors and providing recommendations for further action.

This approach opens up prospects for further improvement of diagnostic systems based on neural networks.

The aim of the study was to improve the accuracy of semantic search of medical information in Russian by finetuning the RuBERT language model on the specialized RuMedDaNet dataset using the Matryoshka Representation Learning method to create compact and efficient vector representations of text.

Materials and Methods. The study utilized the RuMedDaNet dataset, which contains Russian-language medical texts. Various embedding training techniques were applied to optimize performance, including the “matryoshka” approach, which enables reducing the dimensionality of vector representations without loss of quality.

Results. Experiments demonstrated a significant improvement in key search metrics (NDCG, MRR) compared to the baseline RuBERT model. The language model trained in the study has been uploaded to the Hugging Face platform, where it is now available for open use.

Conclusion. The proposed RuBERT fine-tuning method was effective for search tasks in medical RAG systems. The current limitations of the approach and directions for further research are discussed.

The aim of this study was to evaluate the effectiveness of the automated MosMedReg software for subtraction analysis of longitudinal brain MRI in patients with multiple sclerosis in a routine outpatient setting. The study included 30 paired MRI examinations performed on 1.5 T scanners from different manufacturers using T2, FLAIR, and contrast-enhanced T1 sequences with variable slice thicknesses. Image processing was performed using registration and subtraction algorithms based on the SimpleElastix library. Images were assessed manually by an expert and with the assistance of the software; results were evaluated using clinical and technical scoring systems.

The software provided successful registration and subtraction in all cases, including series different in slice thickness and projections. The average number of newly identified lesions using MosMedReg did not differ from expert assessment (p = 0.25); however, in several cases, subtraction enabled the detection of clinically significant changes that were not observed in standard analysis. False-positive findings associated with technical artifacts due to scan parameter mismatches were also noted.

The results confirm the reproducibility and practical applicability of subtraction analysis with MosMedReg for improving the objectivity and standardization of multiple sclerosis diagnosis in outpatient practice.

Aim: To evaluate the effectiveness of artificial intelligence (AI) technologies in interpreting mammographic images using a comparative analysis of reports from primary care radiologists and specialists from the reference center of the of the A.I. Kryzhanovsky Krasnoyarsk Territorial Clinical Oncology Dispensary (KKKOD) when interpreting mammographic studies conducted in the Krasnoyarsk Territory.

We conducted a retrospective analysis of 1012 mammographic examinations performed in March-May 2025, sent to the KKKOD reference center according to the established procedure. The reports from KKKOD reference center and primary care radiologists, as well as the results of two AI services using the BI-RADS scale, were evaluated. Statistical processing was performed using StatTech 4.0.6, and a discordance index was calculated for clinical cases that impact subsequent patient management.

Results: in the Krasnoyarsk Region, the introduction of a second artificial intelligence (AI) service for mammography interpretation by radiologists led to an increase in the diagnostically challenging BI-RADS 3.4 categories, increasing the workload of the KKKOD reference center by 29.5%. However, the discordance rate remained unchanged (27.5%) compared to 2024, when only one AI service was used in the region. A retrospective analysis revealed differences in the performance of the two AI services when interpreting mammography examinations.

Conclusion: the use of multifunctional AI-based digital platforms improves the quality of disease prevention and diagnosis, reducing the likelihood of medical errors. However, the simultaneous use of several services increases the workload of specialists due to the need to analyze multiple interpretation options.