Alexey A. Neznanov - Associate Professor, School of Data Analysis and Artificial Intelligence, Faculty of Computer Science, National Research University Higher School of Economics; Head of Information-Analytical Department, Federal State Budget Institute “Federal Scientific and Clinical Centre of Pediatric Hematology, Oncology and Immunology named after Dmitry Rogachev”, Ministry of Health of the Russian Federation.  
Address: 20, Myasnitskaya Street, Moscow, 101000, Russian Federation.
E-mail: aneznanov@hse.ru 

Julia V. Starichkova - Deputy Head of Information-Analytical Department, Federal State Budget Institute “Federal Scientific and Clinical Centre of Pediatric Hematology, Oncology and Immunology named after Dmitry Rogachev”, Ministry of Health of the Russian Federation. 
Address: 1, Samory Mashela Street., Moscow, 117997, Russian Federation.
E-mail: julia.starichkova@fnkc.ru

      Medical information systems constitute a separate class of corporate information systems, specifically designed to improve the efficiency of healthcare. The purpose of implementation of healthcare information systems in clinical centers is to provide a comprehensive solution of information support issues associated with delivery of health services, with emphasis on the formalization of healthcare business processes, collection and secure storage of patients’ personal data, optimal interface solutions for clinicians and nurses, special management of medications and expendable supplies. Key roles of medical information system users include heads of clinical departments, doctors and nurses. This paper addresses a range of challenges relating to clinical diagnoses in medical information systems, including formalization, input efficiency, validity and completeness checking of enhanced diagnoses, as well as ex-post analysis of clinical data focusing on specific signs of diagnoses.
      Traditionally a diagnosis constitutes an unstructured text in a natural language with further assignment of codes of the International Classification of Diseases or other universal classifications. There are standardized guidelines and local conventions to insert and to change this text, but basically these conventions are not formalized in medical information systems, and that leads to the above-listed problems. A comparative analysis of the International Classification of Diseases has been conducted involving a preliminary assessment of refinements suggested by experts and actually used relating to most common pediatric oncology and hematology diseases. This paper suggests an approach to formalize an additional classification of clinical diagnoses, both simple and effective, a prototype with description of templates and schemes to enhance diagnoses for certain diseases in the JSON format and to optimize interface of the “diagnosis” standard field in medical information systems. This approach has been successfully tested in pediatric oncology information system design.