Applications of Artificial Intelligence in the Use of ICD

Artificial Intelligence offers new possibilities for supporting the use of clinical information in morbidity and mortality processes based on international standards such as the International Classification of Diseases, 11th Revision (ICD-11).

During this session, Daniel Luna, Charly Otero, and Juan Carlos Díaz will present Artificial Intelligence applications that use available clinical information—such as a discharge or hospitalization summary—to identify relevant conditions and assist, as appropriate, in determining the main condition and developing a proposed cause-of-death sequence.

These tools can also suggest ICD-10 or ICD-11 codes, facilitate the review of their correspondence, and enable users to consult alternative terminology. The resulting proposals must be reviewed, modified when necessary, and validated by the responsible health professional.

The session will focus on using clinical information to support these processes and assist with cause-of-death certification. This process should be distinguished from the subsequent statistical coding and selection of the Underlying Cause of Death, which is conducted by applying the World Health Organization’s mortality coding and selection rules.

The speakers will also discuss the benefits and limitations of these applications, as well as considerations related to information quality, interoperability, semantic standards, ethics, data governance, transparency, and human oversight. Artificial Intelligence is presented as a support tool and does not replace clinical judgment or the responsibility of the certifying physician.

This event is being held within the framework of the Community of Practice on Civil Registration and Vital Statistics (CRVS) in Latin America and the Caribbean. It is part of the thematic area on the digital transformation of CRVS, which seeks to promote the active exchange of good practices and relevant information to strengthen these systems throughout the Region.

Overall objective of the series: To strengthen technical and institutional capacities to use ICD-11 as a central semantic standard in Civil Registration and Vital Statistics (CRVS) and Information Systems for Health (IS4H), improving the quality and integration of morbidity and mortality data and enabling the interoperability required for digital transformation in health and informed decision-making.

Specific objective of the session: To explore applications of Artificial Intelligence in the use of ICD-11 for morbidity and mortality, particularly for analyzing clinical information, identifying relevant conditions, supporting the determination of the main condition, and assisting in the development of a proposed cause-of-death sequence, always subject to review and validation by the responsible health professional.

Daniel LunaPhysician with master’s and doctoral degrees in Computer Science and extensive experience in health informatics. He led the development of the clinical information system at Hospital Italiano de Buenos Aires and currently heads its Department of Health Informatics.

Charly OteroPhysician with a master’s degree in Biomedical Informatics and more than 20 years of experience in digital health, interoperability, and information systems. He has led projects and advised health institutions and government agencies throughout Latin America.

Juan Carlos DíazData Engineering Advisor at the Pan American Health Organization (PAHO), specializing in transforming complex data into actionable insights to support decision-making. He promotes the creative and responsible use of data and Artificial Intelligence in public health.

Moderator: Myrna Martí, IS4H Advisor and Digital Literacy Program Coordinator at the Pan American Health Organization.