Personalized Data Science and Personalized (N-of-1) Trials: Promising Paradigms for Individualized Health Care

Area of Science:

• Data Science
• Personalized Medicine
• Self-Experimentation

Background:

• Traditional data science focuses on big data for population-level insights.
• Personalized data science (Per-DS) shifts focus to using personal data for individual decisions.
• Per-DS empowers individuals as investigator, participant, and beneficiary.

Purpose of the Study:

• To describe the features of personalized data science (Per-DS).
• To review the current state and future outlook of Per-DS.
• To highlight the shift from data extraction to 'data gardening' for personal knowledge.

Main Methods:

• Individuals design and implement studies using their own data.
• Methods include systematic monitoring, case-crossover, pre-post trials, and N-of-1 trials.
• Focus on 'data gardening': goal setting, study design, data collection, and analysis.

Main Results:

• Per-DS generates personal knowledge, benefiting the individual's future self.
• Contrasts with traditional data science, which benefits external populations.
• Requires a proactive approach to cultivating and analyzing personal data.

Conclusions:

• Per-DS offers a novel framework for self-informed decision-making.
• Facilitation requires support from various entities for resource development.
• Future outlook includes community forums, mobile apps, and guidance for Per-DS investigators.