• Health Technologies
  • Intelligent Systems

CURATE.AI: Identifying and Optimising N-of-1 Learning Trajectory

PI: Christopher Lee ASPLUND, Agata BLASIAK, Dean HO, Theodore KEE, Thomas YEO

Opportunity

It is no secret that artificial intelligence (AI) works by combining large amounts of data with fast, iterative processing and intelligent algorithms in a way that allows the software to learn automatically from patterns or features in the data. However, while good data sets are fundamental to the success of AI applications, they are not always readily available. And when accessible, trends extracted from these data sets can prove to be too generic to be useful for specific profiles.

Notwithstanding these challenges, the huge potential AI promises as well as continuous research and innovation efforts by industries such as healthcare, finance and manufacturing are driving the strong growth of the global AI market. Valued at US$39.9 billion in 2019, it is expected to expand at a CAGR of 42.2% to reach US$733.7 billion by 2027.

Taking an alternative approach to overcome challenges such as inaccessible and unavailable good data sets, this novel technology explores the use of a small data analytics platform to identify and optimise N-of-1 learning trajectory.

Technology

This proposed technology-CURATE.AI-is an artificial intelligence (AI) platform that is mechanism-independent and indication-agnostic.

Via a digital interface such as Multi-Attribute Task Battery (MATB), CURATE.AI uses empirically recorded/derived input from participants to define their individual profiles rather than relying on synergy prediction between various inputs to globally optimise manipulation.

By modulating training intensity/level of the proposed manipulation, CURATE.AI develops N-of-1 learning trajectory profiles that may actionably mediate training optimisation on the single-subject level and dynamically identify training inputs that drive the best possible scoring outcome or output relating to cognitive ability and/or state.

Besides serving as a powerful optimisation platform for digital therapy, student learning and cognitive decline prevention, the integration of CURATE.AI’s dynamic optimisation with learning/training regimens could potentially enhance the current spectrum of digital therapeutics for depression, anxiety, Parkinson’s disease, and dementia.

Additionally, CURATE.AI has the potential to be applied to improving the administration of pedagogical techniques, optimising learning trajectories and outcomes, enhancing personnel training, and advancing other areas related to cognitive performance.

Document Status

Download

Technology Readiness Level (TRL)

3

Proof-of-concept with needs validated

Applications & Advantages

  • 01

    Mechanism-independent and indication-agnostic AI (data analytics) platform

  • 02

    Varies intensity and difficulty of training according to participants’ performance

  • 03

    Identifies learning trajectory profiles and optimises based on data input from a single individual; does not require large data sets

  • 04

    Applicable to areas related to cognitive performance—including digital therapeutics, online learning and personnel training