Formal representation of ambulatory assessment protocols in HTML5 for human readability and computer execution

Area of Science:

• Digital Health
• Behavioral Science
• Computer Science

Background:

• Ambulatory assessment (AA) collects longitudinal biopsychosocial data using mobile devices.
• Current AA protocol descriptions in natural language are difficult to transcribe and can be ambiguous, hindering reproducibility.
• Programming languages for AA protocols are not easily accessible to most researchers.

Purpose of the Study:

• To propose a novel method for describing ambulatory assessment (AA) protocols.
• To enhance the readability, formality, and executability of AA protocols.
• To bridge the gap between human-readable descriptions and computational implementation of AA studies.

Main Methods:

• Leveraging HyperText Markup Language (HTML) for describing AA materials.
• Developing custom HTML5 elements to represent AA procedures.
• Integrating these elements into a system for creating executable AA protocols.

Main Results:

• Demonstrated that custom HTML5 elements can effectively describe AA procedures.
• Showcased how HTML documents can be treated as executable programs for AA.
• Proposed a method to render AA materials and execute AA procedures on computational platforms.

Conclusions:

• Custom HTML5 elements offer a promising solution for creating executable AA protocols.
• This approach enhances scientific discourse in AA by improving protocol clarity and reproducibility.
• The proposed method facilitates the seamless integration of AA materials and procedures in digital health research.