SOURCE METADATA
Project: AI_READI
Source ID: bmj_protocol_publication
Source type: publication
Source URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11800295/pdf/bmjopen-2024-097449.pdf
Raw file: data/raw/AI_READI/bmjopen-2024-097449_row2.pdf
--------------------------------------------------------------------------------
Open access

Protocol

Cross- sectional design and protocol for
Artificial Intelligence Ready and
Equitable Atlas for Diabetes
Insights (AI- READI)

Cynthia Owsley      ,1 Dawn S Matthies,1 Gerald McGwin,1,2 Jeffrey C Edberg,3
Sally L Baxter      ,4 Linda M Zangwill,4 Julia P Owen,5 Cecilia S Lee,5 AI- READI
Consortium

To cite: Owsley C, Matthies DS,
McGwin G, et al.  Cross-
sectional design and protocol
for Artificial Intelligence Ready
and Equitable Atlas for Diabetes
Insights (AI- READI). BMJ Open
2025;15:e097449. doi:10.1136/
bmjopen-2024-097449

 ► Prepublication history for
this paper is available online.
To view these files, please visit
the journal online (https://doi.
org/10.1136/bmjopen-2024-
097449).

Received 02 December 2024
Accepted 22 January 2025

© Author(s) (or their
employer(s)) 2025. Re- use
permitted under CC BY- NC. No
commercial re- use. See rights
and permissions. Published by
BMJ Group.
1Ophthalmology and Visual
Sciences, The University of
Alabama at Birmingham,
Birmingham, Alabama, USA
2Epidemiology, The University of
Alabama at Birmingham School
of Public Health, Birmingham,
Alabama, USA
3Medicine, University of Alabama
at Birmingham, Birmingham,
Alabama, USA
4Ophthalmology, University of
California San Diego, La Jolla,
California, USA
5Ophthalmology, University
of Washington, Seattle,
Washington, USA

Correspondence to
Dr Cynthia Owsley;
 cynthiaowsley@ uabmc. edu

ABSTRACT
Introduction  Artificial Intelligence Ready and Equitable
for Diabetes Insights (AI- READI) is a data collection
project on type 2 diabetes mellitus (T2DM) to facilitate
the widespread use of artificial intelligence and machine
learning (AI/ML) approaches to study salutogenesis
(transitioning from T2DM to health resilience). The
fundamental rationale for promoting health resilience in
T2DM stems from its high prevalence of 10.5% of the
world’s adult population and its contribution to many
adverse health events.
Methods  AI- READI is a cross- sectional study whose
target enrollment is 4000 people aged 40 and older, triple-
balanced by self- reported race/ethnicity (Asian, black,
Hispanic, white), T2DM (no diabetes, pre- diabetes and
lifestyle- controlled diabetes, diabetes treated with oral
medications or non- insulin injections and insulin- controlled
diabetes) and biological sex (male, female) ( Clinicaltrials. org
approval number STUDY00016228). Data are collected in a
multivariable protocol containing over 10 domains, including
vitals, retinal imaging, electrocardiogram, cognitive function,
continuous glucose monitoring, physical activity, home air
quality, blood and urine collection for laboratory testing and
psychosocial variables including social determinants of
health. There are three study sites: Birmingham, Alabama;
San Diego, California; and Seattle, Washington.
Ethics and dissemination  AI- READI aims to establish
standards, best practices and guidelines for collection,
preparation and sharing of the data for the purposes of
AI/ML, including guidance from bioethicists. Following
Findable, Accessible, Interoperable, Reusable principles,
AI- READI can be viewed as a model for future efforts
to develop other medical/health data sets targeted
for AI/ML. AI- READI opens the door for novel insights
in understanding T2DM salutogenesis. The AI- READI
Consortium are disseminating the principles and processes
of designing and implementing the AI- READI data set
through publications. Those who download and use AI-
READI data are encouraged to publish their results in the
scientific literature.

INTRODUCTION
Artificial  Intelligence  Ready  and  Equitable
for  Diabetes  Insights  (AI- READI)1  is  one  of

STRENGTHS AND LIMITATIONS OF THIS STUDY
 ⇒ The  targeted  sample  size  of  4000  persons  is  the
largest publicly accessible data set currently avail-
able  containing  many  multidomain  variables  rele-
vant to type 2 diabetes mellitus (T2DM).

 ⇒ The  sample  is  designed  to  be  approximately  bal-
anced  with  respect  to  sex  and  race/ethnicities
(Asians,  blacks,  Hispanics  and  whites),  a  demo-
graphic improvement over many previous epidemi-
ological studies and clinical trials on T2DM.

 ⇒ The study population is designed to be balanced in
T2DM severity groups of no diabetes, pre- diabetes,
non- insulin controlled and insulin controlled, to fa-
cilitate artificial intelligence/machine learning model
developments.

 ⇒ The study design process enacted ethical and equi-
table data collection and management practices as
well as data sharing with adherence to the Findable,
Accessible, Interoperable, Reusable principles.
 ⇒ A limitation of our biorepository is that there are a
finite  number  of  samples  to  share  with  scientists
interested  in  using  them  in  research.  Procedures
for  reviewing  and  prioritising  written  requests  will
be developed before the biorepository is complete.

four  National  Institutes  of  Health- funded
Bridge2AI
(https://bridge2ai.org)  proj-
ects  that  aim  to  generate  flagship  biomed-
ical  and  behavioural  data  sets  that  are
ethically  sourced,  trustworthy,  well- defined
and  publicly  available.  AI- READI  is  a  data
generation project focused on type 2 diabetes
mellitus (T2DM) to facilitate the widespread
use  of  artificial  intelligence  and  machine
learning (AI/ML) approaches to study saluto-
genesis.2  Salutogenesis  in  this  context  refers
to the pathway from T2DM to health, which is
the opposite of studying how healthy people
proceed  through  pathogenesis  resulting  in
T2DM. The data set generated by AI- READI,
to  be  described  below,  is  multimodal,  given

1

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
Open access

the multifactorial and multisystemic nature of T2DM. The
fundamental  rationale  for  promoting  health  resilience
in T2DM stems from its high prevalence of 10.5% of the
world’s adult population.3 Persons with T2DM are at risk
for  many  types  of  adverse  health  consequences  such  as
stroke, kidney disease, heart disease, vision impairment,
cognitive  decline,  peripheral  neuropathy  and  physical
inactivity.4 Social determinants of health weigh heavily in
threatening medical status in T2DM,5 including reduced
access to healthcare, decline in treatment adherence and
challenges  in  seeking  follow- up  preventative  care.6  The
prevalence of T2DM is higher in certain racial and ethnic
populations,7  8  and  the  deleterious  consequences  are
exacerbated in these populations.9 10

There  are  certain  unique  features  of  AI- READI’s
design  which  make  it  ideally  suited  for  the  use  of  AI/
ML  analytic  approaches,  which  opens  the  door  for  crit-
ical novel insights into understanding salutogenesis. First,
the AI- READI team, a multidisciplinary group of investi-
gators,  including  clinicians,  data  scientists,  vision  scien-
tists,  computer  scientists,  organisational  scientists  and
ethicists,  designed  the  protocol  such  that  it  is  agnostic
to hypotheses. Rather, the team assembled a data collec-
tion protocol composed of assessments from many health
domains  that  are  likely  impacted  by  T2DM.  Second,  to
address inherent limitations of many existing large data
sets  for  AI/ML  training,  the  cohort  is  designed  to  be
balanced in race/ethnicity, T2DM severity and biological
sex. Third, the data set is large, with a target sample size
of 4000 persons from three geographical regions in the
USA. Fourth, AI- READI places special emphasis on estab-
lishing standards, best practices and guidelines for collec-
tion, preparation and sharing of the data that can be used
for future efforts to develop other data sets targeted for
AI/ML.  Fifth,  AI- READI  addresses  challenges  that  have
compromised  AI  implementation  in  clinical  research
by  including  ethical  and  equitable  data  collection  and
management,  and  adherence  to  Findable,  Accessible,
Interoperable,  Reusable  (FAIR)  principles.  These  issues
have been discussed previously1 but are incorporated in
administering the design and protocol described below.

METHODS
AI- READI was approved by the Institutional Review Board
(IRB) of the University of Washington (approval number
STUDY00016228),  including  reliance  agreements  with
the  IRBs  of  University  of  Alabama  at  Birmingham  and
University  of  California,  San  Diego.  Written  informed
consent is provided by all participants.

Patient and public involvement
A  Community  Advisory  Board  of  11  persons  from  the
three  sites  including  diversity  in  race  and  ethnicity  as
represented  in  the  AI- READI  sample  contributes  to  the
development of the protocol.

Design
AI- READI  is  a  cross- sectional  study  whose  target  enrol-
ment is 4000 people aged 40 and older, triple- balanced by

2

diabetes,

self- reported race/ethnicity, T2DM presence and severity,
and  biological  sex  (figure  1).  Building  balanced  data
sets is critical for the development of unbiased machine
learning  models,  so  rather  than  targeting  the  demo-
graphic  distribution  of  the  US  population,  the  study  is
triple- balanced by recruiting the following populations in
equal proportions: four race/ethnic groups (Asian, black,
white, Hispanic), four categories of T2DM (no diabetes,
pre- diabetes/lifestyle- controlled
diabetes
treated  with  oral  medications  or  non- insulin- injectable
medications, and insulin- controlled diabetes) and biolog-
ical males and females. Participants are recruited across
three data collection sites in different geographical loca-
tions:  Birmingham,  Alabama  (University  of  Alabama  at
Birmingham  (UAB)),  San  Diego,  California  (University
of  California,  San  Diego  (UCSD)),  and  Seattle,  Wash-
ington  (University  of  Washington  (UW)).  All  study
groups are recruited from each site to capture diversity,
but the proportion of each group will vary depending on
the  demographic  prevalence  of  each  group  in  the  site’s
geographical  area.  Participants  are  required  to  speak,
read  and  understand  English.  Pregnancy  and  type  1
diabetes are exclusionary for participation.

Source population
The study base is all patients aged ≥40 years of age who
had a medical encounter within each health system site
(UAB, UCSD, UW) between 2020 and 2025. Patients with
T2DM and pre- diabetes are identified by screening elec-
tronic health records for ICD- 10 diagnosis codes R73.09
and E11.X, respectively. Patients without diabetes will not
have encounters with these ICD codes.

Recruitment
Enrolment began on 18 July 2023 and will continue until
30  November  2026.  Participants  are  recruited  in  waves
to facilitate the efficient sampling of the study base. The
composition and size of each wave are influenced by the
observed  participation  characteristics  of  race/ethnicity,
gender,  severity  of  T2DM  and  site,  according  to  health
records.  As  recruitment  progresses,  the  composition  of
participants is monitored in terms of diversity and inclu-
sion  and  will  be  adjusted  by  under-  and  oversampling
groups as needed. Recruitment in waves also allows suffi-
cient  time  for  coordinators  to  respond  to  persons  who
are interested and follow- up on mailings with expediency.
For  each  recruitment  wave,  a  contact  pool  is  identified
by screening electronic health records at each site. Indi-
viduals  in  each  pool  are  mailed  a  hardcopy  invitation
letter and sent an invitation email, both personalised to
direct  them  into  our  online  REDCap  recruitment  inter-
face using links, access codes and QR codes. Once in the
REDCap  interface,  individuals  can  read  an  overview  of
the  research  programme,  expectations  for  participation
and answers to Frequently Asked Questions. They are also
given the opportunity to download the informed consent
document, request a call back from study staff, complete
a screening survey for qualification and enroll by signing

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
Open access

Figure 1  This project is generating an accessible, shared data set that includes a diverse set of health and behavioural
domains and is harmonised across all variables to be artificial intelligence and machine learning ready. Descriptions of variables
are available in tables 1–4. BMI, body mass index; OCTA, optical coherence tomography angiography.

our electronic consent. Once enrolled, they have access
to  study  questionnaires  (see  below).  Those  who  do  not
have access to the internet may call research staff for alter-
native methods of enrolment.

Protocol
The protocol is performed as a single visit lasting between
2.5  and  4 hours,  depending  on  the  participant.  Partic-
ipants  are  volunteers;  therefore,  there  is  selection  bias
known  as  volunteer  bias  which  may  limit  the  generalis-
ability  of  the  results  among  those  who  are  not  repre-
sented  in  the  study  population.  For  most  participants,
informed  consent  is  performed  remotely  by  computer,
tablet or smart phone through a link in the hardcopy invi-
tation letter or the invitation email which directed them
into  our  REDCap  database.  Once  enrolled,  participants
are presented with questionnaires for online completion.
If participants did not select the option to consent elec-
tronically,  it  is  performed  in- person  at  the  start  of  the
visit, followed by the completion of all questionnaires. All
research personnel at the three data sites who recruited,
enrolled  and  collected  data  underwent  training  by  the
data  manager  at  UAB  and  a  detailed  Manual  of  Proce-
dures  (MOP)  for  reference.  Successful  completion  of
a  certification  process  is  required  for  all  coordinators,

involved  mandatory

which
institutional  compliance
training  for  human  subject  research.  In  addition,  prior
to enrolling and testing participants, all coordinators are
required to successfully administer all protocol elements
to at least three volunteer practice subjects, meeting the
standards  outlined  in  the  MOP.  Data  managers  at  each
site  oversee  the  process.  Preliminary  pilot  enrolment
occurred between 18 July 2023 and 30 November 2023 in
order to secure sufficient familiarity for all coordinators.
The formal data collection process began on 1 December
2023. Preliminary data from the pilot enrollment period
were  included  in  the  first  released  data  set  released
in  May  2024.  A  subsequent  version  of  the  data  set  that
includes all data collected in the first year of the study (up
to 31 July 2024) is planned for release in November 2024.
Information on how data can be accessed is available at
https://fairhub.io/

Figure 1 presents information on all the data domains
collected in AI- READI. Questionnaires addressed several
content domains as listed in table 1. Height, weight and
waist  and  hip  circumference  are  measured,  followed  by
calculation  of  the  waist- hip  ratio  and  body  mass  index.
Systolic  blood  pressure,  diastolic  blood  pressure  and
heart  rate  were  measured  twice  separated  by  2 min  with

3

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
Open access

Table 1  Questionnaires

Questionnaire

Screening questionnaire

Domain addressed

Eligibility (no type 1 diabetes mellitus or pregnancy), diabetes health status,
diabetes treatments (lifestyle, medications), race and ethnicity, biological sex.

Demographic information
Center for Epidemiological Studies Depression Scale—1024 Screens for depressive symptoms.
Problem Areas in Diabetes525
Diabetes score (self- care)26

Date of birth, gender identification, marital status.

Diabetes- associated injuries, lifestyle changes and medical care.

Focuses on diabetes self- management querying dietary habits, exercise,
healthcare and foot care.

Dietary assessment27

Ophthalmic survey

Asks basic questions about food and drink habits over the past few months.

Asks about difficulties with vision and recent eye care.

Smoking, alcohol use, vaping and marijuana use*

Asks about the history of each of these behaviours.

General health

Social Determinants of Health (surveys28

Current Medications with RxNorm codes29

Asks about health history using this question: ‘Has a doctor or other healthcare
professional ever told you that you have/had?’ followed by a list of chronic health
conditions. Responses are yes/no; if yes, may be asked to specify a condition.

Asks about social determinants of health and access to healthcare. Surveys
on food and job insecurity, educational attainment, health insurance coverage,
access to healthcare, housing and neighbourhood environment, perceptions of
discrimination in medical settings and racial/ethnic discrimination (current and
lifetime).

Asks about all prescription and over- the- counter medications currently used.
Includes pills, injections, creams, salves, sprays, eye drops,and dermal patches.

*Marijuana use not assessed for participants living in Alabama because it is an illegal substance.

an  automatic  oscillometric,  medically  approved  device.
A  12- lead  ECG  is  performed  (Philips  Pagewriter  TC30
Cardiograph,  Amsterdam,  The  Netherlands)  while  the
participant sits in a reclining chair or lies supine; the posi-
tion is recorded (0°, 30°, 60° or 90°) relative to the supine
position.  Peripheral  neuropathy  is  performed  using  the
monofilament test11 assessing touch perception on both
feet  with  shoes  and  socks  removed.  During  testing,  the
participants’  eyes  are  closed  while  responding  yes/no
whether  they  feel  the  10 g  filament  in  three  locations
on  each  foot  (10  times  per  location).  A  general  cogni-
tive screener is carried out using the Montreal Cognitive
Assessment12 (MoCA), administered electronically on an
iPad using the MoCA Duo Application (MoCA Cognition,
Quebec,  Canada).  The  total  possible  score  is  30,  with
higher numbers representing better performance.

Visual acuity and contrast sensitivity are assessed under
both  photopic  (daylight)  conditions  and  mesopic  (dim
light) conditions. The right and left eyes are tested sepa-
rately. Photopic letter visual acuity is measured with the
Electronic Visual Acuity tester13 (M&S Technology, Niles,
Illinois) using the instrument’s routine protocol. Mesopic
acuity is measured by the participant viewing the display
through a 2.0 neutral density (ND) filter which reduces
the light level of the test to a mesopic level.14 Visual acuity
is measured while the participant views the display under
best- corrected conditions, expressed as the logarithm of
the  minimum  angle  of  resolution.  Lower  numbers  are
better  resolution.  Photopic  letter  contrast  sensitivity  is
measured  with  the  Mars  chart  (Mars  Perceptrix,  Chap-
paqua, New York)15 using the routine procedure. Mesopic
contrast sensitivity is assessed by viewing through the 2.0

ND filter. Contrast sensitivity is expressed as log sensitivity,
with higher numbers meaning better  sensitivity. Autore-
fraction  data  for  each  eye  are  obtained  (spherical  and
cylindrical  components  expressed  as  diopter  with  cylin-
drical  axis)  using  the  autorefractor’s  standard  protocol
(Topcon  KR  800,  Topcon  Healthcare,  Oakland,  New
Jersey).

Blood  (non- fasting)  and  urine  samples  are  collected
from  participants  during  the  visit.  Whole  blood,  blood
derivatives  and  urine  are  used  for  clinical  lab  testing,
summarised in table 2. Clinical Laboratory Improvement
Amendments- certified  laboratories  local  to  each  data
collection  site  perform  complete  blood  count  (CBC)
testing on fresh whole blood samples; all other lab tests
are performed by the University of Washington Nutrition
and  Obesity  Research  Center  (NORC)  testing  facility
using  stored,  frozen  samples.  Additionally,  blood  deriv-
atives  are  sent  to  the  biorepository  at  UAB’s  Center  for
Clinical  and  Translational  Science  (CCTS)  and  will  be
available  to  researchers  for  future  ancillary  studies  (see
table  3  for  a  summary  and  detailed  discussion  of  the
biorepository).

The  retinal  imaging  protocol  is  designed  to  capture
imaging  from  multiple  devices.  Both  eyes  are  imaged
on  each  participant.  Table  4  details  the  scans  and  data
formats  used  for  each  retinal  imaging  device.  The
imaging  devices  were  chosen  due  to  their  different,  yet
partially overlapping, modalities. The specific scan types
were chosen to provide widespread coverage of the retina
with the overarching goal of detecting pixel- level differ-
ences in pathology. All retinal images are collected under
dilated conditions, except for the Optomed (see table 4).

4

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
Open access

Table 2  Clinical laboratory tests and biorepository specimens

Test

EDTA plasma tests

Units

Reference range*

Rationale for inclusion

 N- terminal pro- B- type natriuretic peptide

pg/mL

Varies by age

Severity and outcome predictor of heart
failure

 Troponin- T

 C peptide

 Insulin

Serum tests

ng/L

ng/mL

ng/mL

Female<11; male<16 Marker of myocardial injury

1.1–4.4

0.0–24.9

Indicator of insulin production

Marker for diabetes

 C reactive protein, high sensitivity

mg/L

0.0–10.0

 Total cholesterol

 Triglycerides

 Hign- density lipoprotein cholesterol

 Low- density lipoprotein cholesterol
(calculated)

 Glucose

 Blood urea nitrogen

 Creatinine

Blood urea nitrogen/creatinine ratio

 Sodium

 Potassium

 Chloride

 Carbon dioxide, total

 Calcium

 Protein, total

 Albumin

 Globulin, total (calculated)

 A/G ratio (calculated)

 Bilirubin, total

 Alkaline phosphatase

 Aspartate aminotransferase

 Alanine aminotransferase

Whole blood tests

 HbA1c

 White blood cell

 Red blood cell

 Haemoglobin

 Haematocrit

 MCV

 MCH

 MCHC

 RDW

 Platelets
Urine tests

mg/dL

mg/dL

mg/dL

mg/dL

mg/dL

mg/dL

mg/dL

mEq/L

mEq/L

mEq/L

mEq/L

mg/dL

g/dL

g/dL

g/dL

<200

<150

>39

<130

62–125

8.0–21.0

Female: 0.38–1.02;
male: 0.51–1.18

135–145

3.6–5.2

98–108

22–32

8.9–10.2

6.0–8.2

3.5–5.2

Inflammation marker; risk factor for
T2DM

Cardiovascular disease risk factor

Cardiovascular disease risk factor

Cardiovascular disease risk factor

Cardiovascular disease risk factor

Marker for diabetes

Marker for kidney and liver function

Marker for kidney function

Indicator of kidney function

Marker of metabolic health

Marker of metabolic health

Marker of metabolic health

Marker of metabolic health

Marker of metabolic health

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

mg/dL

0.2–1.3

Marker of liver health

IU/L

IU/L

IU/L

%

×10E3/μL

×10E6/μL

g/dL

%

fL

pg

g/dL

%

Varies by age

Marker of liver health

9–38

Female age 7–33 and
male age 0–49: 10–64;
male age 50+: 10–48

Marker of liver health

Marker of liver health

4.0–6.0

3.4–10.8

3.77–5.80

11.1–17.7

34.0–51.0

79–97

26.6–33.0

31.5–35.7

11.6–15.4

Marker for diabetes

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

Useful tool for overall health status

×10E3/μL

150–450

Useful tool for overall health status

Continued

5

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449



































Open access

Table 2  Continued

Test

 Urine creatinine

 Urine albumin

Biorepository specimens

 Buffy coats (from EDTA anticoagulated
vacutainers)

 Genomic DNA

 Plasma (from EDTA anticoagulated
vacutainers)

 Serum

 PAXgene vacutainers
 Peripheral blood mononuclear cells

Units

mg/dL

mg/dL

–

–

–

–

–
–

Reference range*

Rationale for inclusion

N/A

N/A

–

–

–

–

–
–

Urinary biomarker for early diabetic
nephropathy

Urinary biomarker for early diabetic
nephropathy

DNA isolations for future studies such as
whole genome sequencing

Genomic applications

Future proteomics and metabolomics
studies

Future proteomics and metabolomics
studies

Future RNA isolations
Peripheral blood mononuclear cells for
future immunological studies; generation
of iPSCs

*Reference range is a set of values that represent low and high ends of results that are considered normal. The above
reference ranges were provided by the testing laboratories.
HbA1c, hemoglobin A1c; iPSCs, induced pluripotent stem cells; MCH, mean corpuscular hemoglobin; MCHC, mean
corpuscular hemobglobin concentration; MCV, mean corpuscular volume; RDW, red blood cell distribution width.

All participant images are exported from imaging devices
in  their  raw  format  and  some  images  required  conver-
sion to a DICOM standard format prior to upload to an
external AI- READI data storage site (discussed below).

At  the  end  of  their  in- person  visit,  participants  are
sent  home  with  three  data  monitoring  devices  as  listed
in  table  4:  (1)  home  environmental  sensor,  (2)  Garmin
fitness  tracker  and  (3)  Dexcom  Continuous  Glucose
Monitor  (CGM).  Coordinators  provide  instructions  on
how to use these monitoring devices continuously for 10
days. Participants are instructed to return the devices to
study staff by overnight mail (mailing materials and fees
provided by the site) or in- person after 10 days. The envi-
ronmental  sensor  and  the  Garmin  fitness  tracker  were
specifically  chosen  with  participant  privacy  as  a  primary
concern. These devices do not capture video or audio, are
not equipped with GPS location monitoring and are not
synced  with  participant- owned  devices.  Participants  are
masked to the results from the Dexcom CGM during the
monitoring  period.  Participants  return  the  monitoring
devices with a data sheet indicating which wrist the fitness
tracker was worn on, their dominant hand and where the
environmental sensor was placed in the home.

The  following  results  are  returned  to  participants
either as they leave the visit or later by HIPAA- compliant,
encrypted  email.  Results  from  heart  rate,  blood  pres-
sure  and  visual  acuity  assessments  are  provided  to  the
participant  on  an  exam  card  before  they  leave  the  visit,
along  with  instructions  in  lay  language  for  interpreting
the data and any follow- up recommendations. Following
the 10- day at- home monitoring period, a Dexcom report
providing average blood glucose levels (hourly, daily and

overall) and an overall Glucose Management Indicator is
provided  by  encrypted  email.  At  the  time  of  the  yearly
data  release  (see  below),  laboratory  test  results  are  sent
with  information  on  normative  values  by  Health  Insur-
ance Portabillity and Accountability- compliant, encrypted
email.

Coordinators  alert  participants  to  several  incidental
findings before they leave a visit. Participants are recom-
mended  to  go  for  emergency  care  for  the  following
reasons: individuals with systolic blood pressure readings
of  over  180 mm  Hg  or  less  than  100 mm  Hg  (with  any
symptoms  of  haemodynamic  instability),  diastolic  blood
pressure readings of greater than 120 mm Hg or less than
60 mm Hg (along with any symptoms of haemodynamic
instability) or heart rate readings of greater than 100 bpm
or less than 60 bpm (if not known to be usual for them
and  with  any  symptoms  of  haemodynamic  instability).
Retinal imaging technicians are trained to detect certain
conditions that are potentially life or vision threatening
(retinal detachment, tumour and optic disc oedema). If
one of these conditions is suspected, immediate follow- up
with  the  participant  is  performed  with  a  referral  to  the
emergency  department  (disc  oedema)  or  referral  to  an
ophthalmologist for immediate care (retinal detachment,
tumour).

Data management
REDCap is used for collecting patient- reported data from
questionnaires  and  the  following  clinical  data  from  the
visit: medications assessment, vitals, visual acuity, contrast
sensitivity, monofilament test results and CBC test results.
Data from the following devices are exported in their raw

6

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449












Open access

Table 3  Biospecimen collection including processing and purpose

Specimen collection

Sample type

Location

Processing details

Purpose

EDTA vacutainers

Whole blood

Whole blood

Plasma

Local clinical lab

UW NORC*

UW NORC*

Plasma

Biobanking at UAB CCTS

Buffy coats

Biobanking at UAB CCTS

None

None

Processed locally at data site
on the day of collection

Complete blood count analysis

Testing for HbA1c

Testing for NT- proBNP,
troponin- T, C- peptide and
insulin

Processed locally at data site
on the day of collection

Future proteomics and
metabolomics studies

Processed locally at data site
on the day of collection

DNA extractions for future
genomics

Serum separator
vacutainers

Serum

Serum

Biobanking at UAB CCTS

Processed locally at data site
on the day of collection

Future proteomics and
metabolomics studies

UW NORC*

Processed locally at data site
on the day of collection

Testing for CRP- HS, glucose,
BUN, creatinine, carbon
dioxide, total protein, albumin,
globulin, bilirubin, alkaline
phosphatase, AST, ALT,
electrolytes and lipids

Urine collection kit

Urine

UW NORC*

Processed locally at data site
on the day of collection

Testing for creatinine and
albumin

CPT mononuclear cell
preparation tubes

Peripheral blood
mononuclear cells

Biobanking at UAB CCTS

UAB—processed locally on
day of collection

Future immunological studies;
generation of iPSCs

PAXgene RNA vacutainers Stabilised,

Biobanking at UAB CCTS

unisolated RNA

UCSD/UW—tubes sent to
UAB for next- day processing

UAB—processed locally on
day of collection

UCSD/UW—tubes sent to
UAB for next- day processing

Future gene expression studies

*Samples analysed by the UW NORC laboratory (serum, plasma, whole blood and urine) are first stored locally at UAB and UCSD at −70°C, then
batch- shipped to UW on dry ice approximately once per quarter.
ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CRP- HS, C reactive protein, high sensitivity; Electrolytes,
sodium, potassium, chloride, calcium; Lipid analysis, total cholesterol, high- density lipoprotein cholesterol, low- density lipoprotein cholesterol,
triglycerides; NT- proBNP, N- terminal pro- B- type natriuretic peptide; UAB, University of Alabama at Birmingham; UAB CCTS, University of Alabama
at Birmingham Centre for Clinical and Translational Science; UCSD, University of California, San Diego; UW, University of Washington; UW NORC,
University of Washington Nutrition Obesity Research Center.

format to local storage: ECG (.xml), MoCA Duo applica-
tion (total score, section subscores, Memory Index Score
and task completion times; .csv), retinal imaging (table 4)
and at- home monitoring devices (environmental sensors,
CGMs, fitness trackers; Table 3). Blood and urine testing
by the NORC lab is provided in .csv format. All data are
mapped to applicable data standard formats, such as the
Observational  Medical  Outcomes  Partnership  Common
Data Model for clinical data and the DICOM format for
retinal  imaging.  All  data  are  uploaded  at  regular  inter-
vals to the AI- READI- specific data management platform
called FAIRhub (https://fairhub.io/), a Microsoft Azure
cloud- based platform developed for this project. For some
devices,  the  data  are  transformed  from  its  proprietary
state  into  a  standard  model  format  prior  to  upload  to
FAIRhub (tables 3 and 4). The Data Manager at each data
site is responsible for the quality control of all data before
uploading  to  FAIRhub.  All  data  are  stored  and  shared
using  FAIRhub  as  ‘AI- ready’  which  enables  immediate

AI/ML research without reformatting and preprocessing.
More  details  about  the  AI- readiness  of  the  data  set  are
provided  in  the  data  set  documentation  (https://docs.
aireadi.org/).

Data availability
A  more  detailed  explanation  of  FAIRhub  was  provided
previously1  and  at  https://aireadi.org.  Deidentified
data  are  released  approximately  yearly  in  two  data  sets:
a  controlled  access  set  and  publicly  accessible  set  with
fewer requirements for access. The controlled access data
set  contains  all  data  available  in  the  publicly  accessible
set,  plus  more  sensitive  data  that  are  only  available  for
use by scientists whose institutions have completed legal
and privacy use agreements with the AI- READI research
programme.  The  publicly  accessible  data  set  from  pilot
data collection was made available in May 2024. All data
collected  through  31  July  2024  were  made  available
in  November  2024.16  Information  regarding  data  set

7

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449


Open access

Table 4  Retinal imaging and at- home monitoring devices

Imaging device

Scans collected

Data format

Manufacturer

Aurora IQ

EIDON widefield truecolor
confocal fundus system

DICOM

DICOM

Fundus: macula and disc centred
(CFP) colour fundus photo (undilated)

Ultra- wide Field Central CFP
Ultra- wide Field Nasal CFP
Ultra- wide Field Temporal CFP
Ultra- wide Field Central IR
Ultra- ide Field Central FAF
Mosaic of Ultra- wide Field images

Number of images
per participant

4

Optomed
(Oulu, Finland)

iCare USA, Inc (Raleigh, North
Carolina)

12

Spectralis HRA optical
coherence tomography
(OCT) and optical coherence
tomography angiography
(OCTA)

Optic Nerve Head- Radial Circle- High
Resolution (OCT)
Posterior Pole Macula- HR- 61 lines
(OCT)
Macula- 20×20- HS- 512 lines (OCTA)

Maestro2 3D OCT- 1

3D Wide (H) 12×9–512×128 (OCT)
3D Macula 6×6–512×128 (OCT)
Macula 6×6–360×360 (OCTA)

Triton DRI OCT

Cirrus 5000

3D(H)+Rad 12×9–512×256 (OCT)
Macula 6×6–320×320 (OCTA)
Macula 12×12–512×512 (OCTA)

Disc Cube, −200×200 (OCT)
Macula Cube, −512×128 (OCT)
Macula, −6×6 (OCTA)
Disc, −6×6 (OCTA)

Fluorescence Lifetime Imaging
Ophthalmoscopy

Macula- centred

DICOM

Heidelberg Engineering GmbH
(Heidelberg, Germany)

6

Topcon Healthcare (Itabashi-
ku, Tokyo, Japan)

Topcon Healthcare (Itabashi-
ku, Tokyo, Japan)

6

6

Carl Zeiss Meditec AG (Jena,
Germany)

8

Heidelberg Engineering GmbH
(Heidelberg, Germany)

2

Exported in
.fda file format;
converted to
DICOM for the
data set

Exported in
.fda file format;
converted to
DICOM for the
data set

DICOM

Exported in
.sdt file format;
converted to
DICOM for the
data set

Monitoring device

Data variables collected

Data format

Manufacturer

Dexcom G6 Continuous
Glucose Monitor

Blood glucose measurements (mg/dL)
every 5 min

.csv

Dexcom, Inc. (San Diego, California)

Garmin VivoSmart 5 Physical
Activity Monitor

Environmental Sensor

Exported in
.FIT file format;
converted
to mHealth
standard for the
data set

.csv

Number of steps
Heart rate
Sleep duration (circadian and diurnal
rhythm)
Oxygen saturation

Ambient temperature
Relative humidity
Nitrogen oxides (NO and NO2)
Volatile organic compounds
Particulate matter (PM1.0, PM2.5,
PM4, PM10)
Multi- spectral light intensity
measurements (11)

Garmin Ltd (Schaffhausen, Switzerland)

Custom designed, Karalis Johnson Retina Center,
University of Washington (Seattle, Washington)30

availability  and  access  may  be  found  at  https://aireadi.
org/goals/data-sharing. The requirements for controlled
access data sets are being currently developed by the Data
Access Committee.

Biospecimen processing and repository
Blood  (53 mL)  is  collected  for  clinical  lab  assays  and
biobanking  purposes,  and  a  urine  specimen  is  also
collected  during  the  study  encounter  (table  2).  The
collection  materials,  sample  type,  specimen  handling

and  location  details  are  summarised  in  table  3.  Local
processing  for  plasma,  serum  and  buffy  coats
is
performed  using  standardised  operating  procedures,
ensuring consistent handling of these biospecimens for
subsequent  biobanking  and  clinical  lab  analyses.  One
EDTA vacutainer is dedicated for delivery to a local clin-
ical  testing  lab  for  CBC  analysis.  Whole  blood,  plasma,
serum and urine collected for central clinical lab assays
are  batch- shipped  to  the  UW  NORC  on  a  regular  basis

8

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
(table  3).  Likewise,  plasma,  serum  and  buffy  coats
collected at UW and UCSD are batch- shipped to the UAB
CCTS for integration into a central study- wide biobank.
Genomic  DNA  extractions  from  stored  buffy  coats  are
performed at UAB CCTS. Because of the complexity in
isolating and cryopreserving peripheral blood mononu-
clear  cells  (PBMCs),  centralised  processing  of  the  CPT
tubes occurs at the UAB CCTS. While this strategy does
introduce  an  additional  time  variable  for  isolation  and
cryopreservation of PBMC from participants recruited at
UW and UCSD, the ability to pre- centrifuge these tubes
locally  followed  by  overnight  delivery  to  UAB  allows
for  consistency  in  initial  vacutainer  handling.  Finally,
the  PAXgene  RNA  vacutainers  are  stable  for  up  to  72
hours  after  collection,  allowing  for  overnight  shipment
of  vacutainers  from  UW  and  UCSD  to  the  UAB  CCTS.
Biobanked samples will eventually be available to scien-
tists  according  to  procedures  and  policies  that  are  in
development.

AI- READI  has  several  strengths  which  make  it  highly
suitable  for  AI/ML  studies  on  T2DM.  The  targeted
sample size of 4000 persons is the largest publicly acces-
sible data set currently available containing many multi-
domain  variables  relevant  to  T2DM.  Furthermore,  after
1 year  of  enrolment,  the  sites  are  currently  on  schedule
for reaching a final sample of 4000 persons within 3 years
of  data  collection.  The  sample  will  be  approximately
balanced  with  respect  to  Asians,  blacks,  Hispanics  and
whites, a demographic improvement over many previous
epidemiological  and  clinical  trials  on  T2DM.  A  meta-
epidemiological review of T2DM studies published from
2000  to  2020  indicated  that  samples  lacked  racial  and
ethnic  diversity,17  thus  making  AI- READI  an  important
step forward in addressing these demographic inequities
in research. The study design process enacted ethical and
equitable  data  collection  and  management  practices,  as
well as data sharing with adherence to the FAIR principles.
Challenges  will  also  be  encountered.  It  is  a  well-
documented  observation
that
recruiting  black  men  into  studies  is  more  difficult  than
recruiting  other  populations,18  yet  this  population
represents  a  disproportionate  percentage  of  burden  in
T2DM.19  Factors  such  as  medical  mistrust  and  fear  of
safety  stem  from  historical  events  such  as  the  Tuskegee
syphilis study.20 Researchers have identified strategies to
encourage  interest  among  black  men  to  participate  in
research such as tailoring printed materials and using a
personalised,  participatory  approach  to  recruitment.21
These  trends  for  under- recruitment  and  for  disease
burden  have  also  been  reported  for  Hispanic22  and
Asian  persons.23  If  sample  balancing  on  race/ethnicity
emerges as a challenge in AI- READI, we will implement
new  recruitment  strategies  to  overcome  the  imbalance.
Requiring that eligible participants speak, read and under-
stand English may have created challenges in recruiting
Hispanic and/or Asian persons. A limitation of our biore-
pository  is  that  there  are  a  finite  number  of  samples  to
share with scientists interested in using them in research.

in  health  research

Open access

Procedures for reviewing and prioritising written requests
will be developed before the biorepository is complete.

In  summary,  AI- READI  is  an  NIH  Common  Fund
Bridge2AI- supported data collection effort to generate a
4000- person data set with T2DM that is suitable for AI/
ML.  It  is  triple- balanced  with  respect  to  race/ethnicity,
biological  sex  and  T2DM  severity  (including  those  who
do not have T2DM). It is multimodal with over 10 variable
domains that can be associated with adverse conditions in
T2DM, yet it is hypothesis agnostic. AI- READI emphasises
making  data  ‘AI- ready’  and  establishing  standards,  best
practices and guidelines for collection, preparation and
sharing  of  the  data.  This  approach  can  be  viewed  as  an
example  for  future  efforts  to  develop  other  health  data
sets  targeted  for  AI/ML.  AI- READI  opens  the  door  for
novel insights into understanding T2DM salutogenesis.

ETHICS AND DISSEMINATION
AI- READI aims to establish standards, best practices and
guidelines for collection, preparation and sharing of the
data for the purposes of AI/ML, including guidance from
bioethicists.  Following  FAIR  principles,  AI- READI  can
be viewed as a model for future efforts to develop other
medical/health data sets targeted for AI/ML. AI- READI
opens  the  door  for  novel  insights  into  understanding
T2DM  salutogenesis.  The  AI- READI  Consortium  are
disseminating  the  principles  and  processes  of  designing
and implementing the AI- READI data set through publi-
cations.  Those  who  download  and  use  AI- READI  data
are  encouraged  to  publish  their  results  in  the  scientific
literature.

X Cynthia Owsley @cynthiaowsley

Collaborators  AI- READI Consortium. Amir Bahmani, Sally Baxter, Edward Boyko,
Christopher Chute, Aaron Cohen, Jorge Contreras, Garrison Cottrell, Virginia de
Sa, Jeffrey Edberg, Nicole Ehrhardt, Nicholas Evans, Irl Hirsch, Michelle Hribar,
Samantha Hurst, Aaron Lee, Cecilia Lee, T.Y. Alvin Liu, Bonnie Maldenado, Gerald
McGwin Jr., Shannon McWeeney, Cynthia Owsley, Bhavesh Patel, Sara Singer,
Michael Snyder, Bradley Voytek, Joseph Yracheta, Linda Zangwill.

Contributors  CO is the guarantor. All authors are submitting authors. They meet
the four criteria of the IJMJE Recommendations 2019 as follows: substantial
contributions to the conception or design of the work; drafting the work or revising
it critically for important intellectual content; final approval of the version to be
published; and agreement to be accountable for all aspects of the work in ensuring
that questions related to the accuracy or integrity of any part of the work are
appropriately investigated and resolved. The AI- READI Consortium are collaborators
(group authorship). CO is the corresponding author.

Funding  This research is supported by National Institutes of Health grants
OT2OD032644, P30DK035816, and UL1TR003096 and Research to Prevent
Blindness.

Competing interests  CO: Johnson & Johnson Vision, Sanofi (consultant). SLB:
Topcon (consultant, travel). LZ: AbbVie, Topcon Medical Systems (consultant);
AISight Health Inc. (stock or stock options). DM, GMcG, JE, JO, CL: none.

Patient and public involvement  Patients and/or the public were involved in the
design, or conduct, or reporting, or dissemination plans of this research. Refer to
the Methods section for further details.

Patient consent for publication  Not applicable.

Provenance and peer review  Not commissioned; externally peer reviewed.

Open access  This is an open access article distributed in accordance with the
Creative Commons Attribution Non Commercial (CC BY- NC 4.0) license, which

9

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
Open access

permits others to distribute, remix, adapt, build upon this work non- commercially,
and license their derivative works on different terms, provided the original work is
properly cited, appropriate credit is given, any changes made indicated, and the use
is non- commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

ORCID iDs
Cynthia Owsley http://orcid.org/0000-0003-3424-011X
Sally L Baxter http://orcid.org/0000-0002-5271-7690

REFERENCES
  1  AI- READI Consortium. AI- READI: rethinking AI data collection,

preparation and sharing in diabetes research and beyond. Nat Metab
2024;6:2210–2.

  2  Kalra S, Baruah MP, Sahay R. Salutogenesis in Type 2 Diabetes
Care: A Biopsychosocial Perspective. Indian J Endocrinol Metab
2018;22:169–72.

  3  International Diabetes Federation. IDF diabetes atlas. 2021.
  4  Bonora E, DeFronzo RA. Diabetes complications, comorbidities and

related disorders. Springer, 2020.

  5  Hill- Briggs F, Fitzpatrick SL. Overview of Social Determinants
of Health in the Development of Diabetes. Diabetes Care
2023;46:1590–8.

  6  Khunti N, Khunti N, Khunti K. Adherence to type 2 diabetes

management. Br J Diabetes 2019;19:99–104.

  7  Cheng YJ, Kanaya AM, Araneta MRG, et al. Prevalence of Diabetes
by Race and Ethnicity in the United States, 2011- 2016. JAMA
2019;322:2389–98.

  8  Unnikrishnan R, Pradeepa R, Joshi SR, et al. Type 2 Diabetes:
Demystifying the Global Epidemic. Diabetes 2017;66:1432–42.

  9  Wilson C, Alam R, Latif S, et al. Patient access to healthcare

services and optimisation of self- management for ethnic minority
populations living with diabetes: a systematic review. Health Soc
Care Community 2012;20:1–19.

 10  Office of Minority Health. Diabetes and american Indians/Alaskan
natives (US department of health and human services). 2021.
Available: https://minorityhealth.hhs.gov/diabetes-and-american-
indiansalaska-natives

 11  Boulton AJM, Armstrong DG, Albert SF, et al. Comprehensive foot
examination and risk assessment: a report of the task force of the
foot care interest group of the American Diabetes Association,
with endorsement by the American Association of Clinical
Endocrinologists. Diabetes Care 2008;31:1679–85.

 12  Nasreddine ZS, Phillips NA, Bédirian V, et al. The Montreal Cognitive

Assessment, MoCA: a brief screening tool for mild cognitive
impairment. J Am Geriatr Soc 2005;53:695–9.

 13  Beck RW, Moke PS, Turpin AH, et al. A computerized method
of visual acuity testing: adaptation of the early treatment of

diabetic retinopathy study testing protocol. Am J Ophthalmol
2003;135:194–205.

 14  Owsley C, Swain TA, McGwin G Jr, et al. How Vision Is Impaired
From Aging to Early and Intermediate Age- Related Macular
Degeneration: Insights From ALSTAR2 Baseline. Transl Vis Sci
Technol 2022;11:17.

 15  Arditi A. Improving the design of the letter contrast sensitivity test.

Invest Ophthalmol Vis Sci 2005;46:2225–9.

 16  AI- READI Consortium. Flagship dataset of type 2 diabetes from the
AI- READI project (1.0.0) FAIRhub. Available: https://docs.aireadi.org/

 17  Ahmed R, de Souza RJ, Li V, et al. Twenty years of participation

of racialised groups in type 2 diabetes randomised clinical trials: a
meta- epidemiological review. Diabetologia 2024;67:443–58.

 18  Randolph S, Coakley T, Shears J. Recruiting and engaging African-

American men in health research. Nurse Res 2018;26:8–12.

 19  Liburd LC, Namageyo- Funa A, Jack L Jr. Understanding 'masculinity'
and the challenges of managing type- 2 diabetes among African-
American men. J Natl Med Assoc 2007;99:550–2.

 20  Scharff DP, Mathews KJ, Jackson P, et al. More than Tuskegee:

understanding mistrust about research participation. J Health Care
Poor Underserved 2010;21:879–97.

 21  Woods VD, Montgomery SB, Herring RP. Recruiting Black/African
American men for research on prostate cancer prevention. Cancer
2004;100:1017–25.

 22  O Rojo M, Jing J, Wells C, et al. Hispanics’ Perceptions of

Participation in Research Studies and Solutions for Improvement in
Participation. J Family Med Community Health 2024;11:1–9.
 23  Lim J- W, Paek M- S. Recruiting Chinese- and Korean- Americans in
Cancer Survivorship Research: Challenges and Lessons Learned. J
Cancer Educ 2016;31:108–14.

 24  Andresen EM, Malmgren JA, Carter WB, et al. Screening for

depression in well older adults: evaluation of a short form of the
CES- D (Center for Epidemiologic Studies Depression Scale). Am J
Prev Med 1994;10:77–84.

 25  McGuire BE, Morrison TG, Hermanns N, et al. Short- form measures

of diabetes- related emotional distress: the Problem Areas in Diabetes
Scale (PAID)- 5 and PAID- 1. Diabetologia 2010;53:66–9.

 26  Hashim MJ, Nurulain SM, Riaz M, et al. Diabetes Score questionnaire
for lifestyle change in patients with type 2 diabetes. Clin Diabetes
2020;9:379–86.

 27  Paxton AE, Strycker LA, Toobert DJ, et al. Starting the conversation
performance of a brief dietary assessment and intervention tool for
health professionals. Am J Prev Med 2011;40:67–71.

 28  Hamilton CM, Strader LC, Pratt JG, et al. The PhenX Toolkit: get the
most from your measures. Am J Epidemiol 2011;174:253–60.

 29  Unified Medical Language System (UMLS). RxNorm. Available:

https://www.nlm.nih.gov/research/umls/rxnorm/index.html [Accessed
7 Aug 2024].

 30  Shaffer J, Gim N, Wei R, et al. Portable environmental sensor

enabling studies of exposome on ocular health. Invest Ophthalmol
Vis Sci 2024;65:6370.

10

Owsley C, et al. BMJ Open 2025;15:e097449. doi:10.1136/bmjopen-2024-097449
