A Knowledge Graph (KG) is a semantic network that stores information about various entities and their interrelationships. In a KG, an entity or “node” can represent anything, such as a person, object, date, concept, or any other material object or abstract idea. Predicates or “edges” illustrate the connections and relationships between different entities within the KG [11]. The concept of KGs gained attention when Google incorporated it into their search algorithm in 2012 [12].

KGs are used to consolidate vast amounts of data from different sources into a single knowledge repository. They can be either comprehensive, covering a wide range of data types, or specialized, focusing on a specific subject area. For example, Wikidata [13] is a broad knowledge graph that includes a diverse range of information, while BioPortal [14] is a specialized graph containing over 140 billion facts related to biotechnology and medicine. These repositories are publicly accessible, and the number of open data sites on the internet continues to grow, forming a cloud of interconnected data known as the Linked Open Data Cloud [15].

Traditional databases excel at storing structured data but struggle to capture the links between different pieces of information. In contrast, KGs are designed to handle unstructured data and visualize complex interrelationships between data nodes [16]. Traditional databases rely on tables with rows and columns, suitable for structured data divided into predefined categories. In contrast, graph databases are built around nodes (entities) connected by edges (relationships), enabling them to efficiently manage complex interrelationships between objects [16].

Blockchain is a decentralized database where information is stored as a chain of blocks, each containing information about the previous block, ensuring data integrity and security [6]. In medicine, blockchain can store medical records, analysis results, research data, and facilitate the transparent and secure exchange of information among system participants [5]. A Decentralized Knowledge Graph (DKG) integrates blockchain technology into KG technology, providing reliable data protection from unauthorized access and tampering, crucial in the medical field where privacy is paramount [17].

The fundamental principles of constructing a DKG in medicine include several key aspects. First, it is necessary to identify the primary medical entities to be represented in the graph, such as diseases, symptoms, medications, procedures, and patients [18]. Next, the relationships between these entities should be defined to build a comprehensive understanding of medical knowledge and its interrelationships. Another important principle is structuring the data in the graph, identifying attributes for each entity that store additional information. For instance, for a disease, attributes could include symptoms, causes, diagnosis methods, and treatments. Structured data helps organize information effectively, making it more accessible for analysis [18].

Integrating data from various sources is another critical aspect of building a DKG. Medical information is often diverse and spread across different databases and sources. Integrating this data into the graph creates a single point of access, improving its quality and completeness [19]. Additionally, updating the KG is crucial. Medical information is constantly evolving, with new research and discoveries altering our understanding of diseases and treatments. Therefore, it is essential to develop mechanisms for automatically updating the data in the graph and monitoring its relevance [19]. DKG improves the security and confidentiality of medical data and addresses the issue of constant updates through the inherent economic incentives of blockchain technology [20] .

Various blockchain platforms can be used to create a DKG, such as Bitzenzor [21], Cortex [22], and Cyber [9, 10]. These platforms enable the decentralized improvement of automatic text processing methods, semantic analysis, machine learning, and more by automatically extracting, structuring, and analyzing information from various sources [23].

We used the Cyb.ai platform to create the DKG for the WSES-WJES bibliography. Cyb.ai’s key feature is the extremely low cost of adding information to the global knowledge graph while ensuring security. Additionally, Cyb.ai offers unmatched flexibility, allowing any structured data to be entered into the DKG. Therefore, we utilized the capabilities of Cyb.ai, a decentralized protocol in the Cosmos network [10], to begin creating the WSES Knowledge Graph based on WJES publications.

Our tasks included collecting and systematizing articles on the diagnosis and treatment of emergency surgical diseases, developing a semantic network schema based on Cyb.ai, and subsequently testing it. These steps enabled us to create a unified user environment for storing and presenting information about medical entities, their interrelationships, and the dynamics of changes in emergency surgery.

During this study, we identified significant advantages and prospects that DKG can offer for a new type of information storage, easy search, extraction, and utilization of medical Big Data. We believe that our “local” emergency surgical DKG based on WSES-WJES publications can be a first step toward creating a blockchain-based general medical knowledge base.