Year 4 Call for Proposals

AIM-AHEAD Federated Network Program

Cohort 2

About AIM-AHEAD

The AIM-AHEAD Coordinating Center was established to foster broad participation in the field of artificial intelligence and machine learning (AI/ML), with an emphasis on promoting health for all Americans. To achieve this objective AIM-AHEAD is engaging in a fair and transparent process of building a consortium of AI/ML to advance health across all American communities and broaden participation in the AI/ML workforce. Many communities have untapped potential to contribute new expertise, data, recruitment strategies, and cutting-edge science to the AI/ML field. To promote health for all, AIM-AHEAD Coordinating Center (A-CC) seeks to increase participation and engagement in AI/ML through mutually beneficial partnerships, stakeholder engagement, and outreach.

AIM-AHEAD Federated-Network Information

Due to organizational policies, sites are often unable to share patient-level data outside of their establishments. Federated data networks allow for patient level-data to stay at the local site and enable sites to opt-in to participate. This allows sites to voluntarily join the network and participate in research projects based on their own data availability, research interests, and local policies. Because only aggregate data leaves the individual sites, this approach affords data privacy and safety while generating more comprehensive datasets and research. This model also allows for greater participation as sites that have not been able to engage in collaborations involving data sharing can participate in a federated model. Participating sites will receive assistance to set up local federated infrastructure and will be provided with centralized code to run on their data locally. Sites share ONLY aggregate data results as they feel comfortable. Aggregate data will be shared through use of a private GitHub repository so that data can be combined centrally and then shared with other network participants. The private Github repository will be password protected. ONLY AGGREGATE data will be shared; patient-level data will not be shared or stored outside of the local site. Easy-to-use dockerized environments can be set up at each participating site with data science tool kits to analyze electronic health record (EHR) data.

The AIM-AHEAD Federated Network allows sites to participate in collaborative research and share data across sites without the need for patient-level data to leave the local site. The program is open to organizations conducting research at the intersection of healthcare practice and delivery, AI/ML methods, and the advancement of health research. The AIM-AHEAD Consortium seeks projects that represent all fields of study, organizations, geography, and investigators.

This RFA calls for site participation in AIM-AHEAD federated research programs. The research team from each site is expected to (i) assemble EHR data locally from their institution to be used for AIM-AHEAD research projects; (ii) provide informatics expertise to participate in EHR data harmonization across sites; and (iii) facilitate the analysis of EHR data using AI/ML algorithms developed by the AIM-AHEAD researchers. While all patient-level EHR data remain behind the firewalls of the local site, the research team is expected to facilitate the sharing of aggregated data to support AIM-AHEAD research projects.

Expected outcomes for the participating site include protocols for data curation and data sharing fine tuned to their specific institution and incorporating lessons learned. The sites are expected to work with the core team members to assess capacity in running AI/ML models locally and sharing outputs of AI/ML models. During year 2 of the funding cycle, the participating sites are expected to generate and share aggregated data for relevant research questions to enable the research teams to perform federated analyses.

The participating sites and the AIM-AHEAD Infrastructure Core are expected to jointly work on publications detailing the processes of data curation, data harmonization and AI/ML analyses with local data, as well as federated analyses for selected research questions.

Specific Activities

Program Description

• AIM-AHEAD will select up to 5 sites to participate in Cohort 2 of the Federated Data Network. These new sites will work with up to 4 previously established sites from Cohort 1 of the AIM-AHEAD Federated Network Program.
• Selected sites will have patient populations and investigator backgrounds and expertise that ensure varied network and research questions that align with the AIM-AHEAD North Stars.
• Site investigators, in collaboration with AIM-AHEAD leadership, will determine the research questions for the network and may participate in or lead the analyses. A research question of interest should be submitted as part of the application.
• Research questions will be selected, in part, based on the types of data in the network, the resources available within the sites, and the feasibility of completing the research within the project period.
• IRB approval and associated agreements will be obtained to enable the request of EHR data for participating in research activities as part of the AIM-AHEAD federated network.
• No patient-level data will leave the participating sites. Analyses will be run locally, and only the aggregate results will be uploaded to a secure central location.
• A copy of the aggregate results will be posted to a private GitHub repository so that the results can be combined centrally and then shared with other network participants.
• Central access to the individual sites’ aggregate results will be limited to investigators participating in the study. However, the combined data from all sites will be made publicly available. Publications based on the aggregate data will not present data subdivided by site unless all sites approve.
• Expected outcomes for the participating site include protocols for data curation and data sharing tailored to their specific institution along with lessons learned. The sites are expected to work with the core team members to assess capacity in running AI/ML models locally and sharing outputs of AI/ML models. During year 2 of the funding cycle, the participating sites are expected to generate and share aggregated data for relevant research questions to enable the research teams to perform federated analyses.
• The participating sites and the core are expected to jointly work on publications detailing the process of data curation, data harmonization and AI/ML analyses with local data, as well as federated analyses for selected research questions.
• Each institution will compute the aggregate study data locally on their clinical data warehouse using SQL and R/python code that is provided to them by the AIM-AHEAD Infrastructure core.

Application Requirements

• Applicants must be able to extract EHR data, save the data to a secure location at the local site, run analyses on the data from software distributed to sites on Docker (e.g., R or Python), and share the aggregate results with others on the project team. Easy-to-use dockerized environments can be set up at each participating site with data science tool kits to analyze electronic health record (EHR) data. The site must have local infrastructure in place to securely store and work with extracted data. Only aggregate results will be shared outside of the local site.
• Applicants must provide counts of the number of patients whose data will be available to use in the network. The counts should be presented in a table, similar in format to the NIH Inclusion Enrollment Report showing breakdowns by patient characteristics. (See https://www.era.nih.gov/erahelp/ASSIST/Content/ASSIST_Help_Topics/3_Form_Screens/PHS_HS_CT/Incl_Enroll_Rprt.htm)
• Applicants must provide a brief description of the site’s patient population, including the primary geographic region of the patients and whether it contains adult and/or pediatric patients. Other characteristics may be highlighted as well, such as patients from health centers, ambulatory or dental clinics, geriatric care, etc.
• Applicants must provide a brief description of the available data, including the data types (demographics, diagnoses, medications, laboratory test results, etc.) and approximate start and end years of the data.
• Applicants must provide a brief description, to the extent possible, of the local environment, including how the investigators will access the EHR data, the secure location where data extracts will be stored, and what computational resources are available for analyses.
• Applicants must describe any prior experience working with EHR data, common data models or ontologies, or data from multi-site clinical studies.

Proposed Budget

Year 1: Site Engagement - Governance Phase $100K Total cost per application
Year 2: Site Engagement - Implementation Phase $300K Total cost per application

AIM-AHEAD Federated Network program - Cohort 2 - will begin on September 8, 2025

Year 1 of the AIM-AHEAD Federated Network Program will be focused on the collective establishment of the network’s governance policies, and Year 2 will be focused on developing the technical components to enable federated research.

Up to 5 sites will be awarded.

FAQs

Collections of answers to our most Frequently Asked Questions will be available here:

AIM-AHEAD Federated Network Program FAQs

Please visit this page regularly as we will continue to add additional questions and answers.

Helpdesk

Questions regarding the AIM-AHEAD Federated Research Network may be directed to: https://helpdesk.aim-ahead.net/ticket/create/federated_network

Directors: Paul Avillach, MD, PhD; Griffin Weber, MD, PhD; Usha Sambamoorthi, PhD; Gabriel Brat, MD, MPH; Tianxi Cai, PhD