Skip to content

Data Collection

Develop and implement a protocol that will generate a diverse and ethically-sourced dataset for the study of type 2 diabetes.

Alt text

Multimodal Data from Diverse Participants

The AI-READI project aims to collect data across diverse participants. The data will be balanced across sexes (equal number of male and female), ethnicities/races (equal number of Asian, Black, Hispanic, and White), and health states (equal number of non-diabetic, diet-controlled diabetic, oral medication-controlled diabetic, and insulin-controlled diabetic).

To achieve that, data is being collected at three study sites: University of Alabama at Birmingham (UAB), University of California San Diego (UCSD), and University of Washington (UW). The same study protocol is followed at each site where a wide range of data types are collected from each participant including survey responses, vision assessment results, retinal imaging, and activity monitoring data. Extraction of electronic health records (EHRs) is also planned.

Management and Processing of Data

We are developing a novel platform called FAIRhub that is aimed at facilitating the management and sharing of data being collected in the project. The study management component of FAIRhub ( enables the data collecting sites to upload data on the go as it is being collected and to track data collection through dashboards. The platform also includes automation tools to process data and extract metadata according to our selected standards for making the data AI-ready (see Data Sharing page).