{"stable_id":"0bbf6e47dd114b1f","slug":"johns-hopkins-access","company":"Johns Hopkins Children's Center / Johns Hopkins Medicine","workflow_name":"Autonomous AI diabetic eye exam at point of care (ACCESS RCT; LumineticsCore/IDx-DR)","function_code":"healthcare_screening","pattern_codes":["point_of_care_relocation","queue_elimination","autonomous_with_backstop"],"changed_assumption":"Closing the diabetic eye-screening care gap requires persuading youth to attend a separate specialist appointment. Instead, the diagnostic exam itself is brought into the routine diabetes visit and read autonomously by AI in under a minute.","evidence_strength":"verified","publication_tier":"showcase","freshness":"current","reviewed_at":"2026-08-23","updated_at":"2026-08-22","source_quality_summary":"3 peer reviewed, 2 primary; publication outcomes are verified.","caveat_summary":"Peer-reviewed RCT under trial conditions. Outcome is process completion, not vision preserved. Control-arm denominator appears as n=83 in the abstract and 18/82 (Table 3, n=163) in the body; both round to 22%. Denominators differ by design (intervention: DED-positive participants completing ECP follow-up; control: participants completing the ECP exam), so the comparison is not like-for-like; the paper defines the secondary outcome this way explicitly. Pre/post observational, single center, reported in a peer-reviewed implementation review by the operating team.","freshness_caveat":null,"workflow_summary":{"before":[{"actor":"Study coordinator / diabetes clinic staff","action":"Refers youth to an external eye-care provider with scripted education and a paper guide.","actor_type":"human"},{"actor":"Patient and family","action":"Schedule, travel to, and attend a separate eye appointment within 6 months","actor_type":"human"},{"actor":"Eye care provider","action":"Perform dilated diabetic eye exam and return results","actor_type":"control"}],"hinge":"Diabetic eye screening can happen during the existing clinic visit.","after":[{"actor":"Trained clinic operator (no ML expertise required)","action":"Captures two fundus images during the routine endocrinology visit without pharmacologic dilation.","actor_type":"human"},{"actor":"Autonomous AI diagnostic system (LumineticsCore)","action":"Returns one of three results within 60 seconds: DED present, absent, or insufficient image quality.","actor_type":"ai"},{"actor":"Clinic staff","action":"Communicates the result and, if DED is present, gives scripted referral education.","actor_type":"human"}],"decision_mode":"bounded_autonomy","decision_marker":"AI diagnoses; specialists retain treatment and the safety backstop"},"before":[{"order":1,"actor":"Study coordinator / diabetes clinic staff","action":"Refers youth to an external eye-care provider with scripted education and a paper guide.","handoff_to":"Eye care provider","control":"Referral documentation; completion tracked via EHR and phone follow-up"},{"order":2,"actor":"Patient and family","action":"Schedule, travel to, and attend a separate eye appointment within 6 months","handoff_to":"Eye care provider","control":"Patient initiative; 78% of the control arm never completed the exam in the window"},{"order":3,"actor":"Eye care provider","action":"Perform dilated diabetic eye exam and return results","handoff_to":"Diabetes clinic","control":"ECP clinical judgment"}],"after":[{"order":1,"actor":"Trained clinic operator (no ML expertise required)","action":"Captures two fundus images during the routine endocrinology visit without pharmacologic dilation.","handoff_to":"Autonomous AI system","control":"AI image-quality algorithm guides acquisition and forces retakes (max 3 attempts)"},{"order":2,"actor":"Autonomous AI diagnostic system (LumineticsCore)","action":"Returns one of three results within 60 seconds: DED present, absent, or insufficient image quality.","handoff_to":"Study coordinator / clinic staff","control":"Locked deterministic medical device under FDA De Novo regulation; in this off-label youth deployment, all images were additionally overread by a board-certified retina specialist (estimated sensitivity 100%, specificity 78.9% vs level-4 reference standard)"},{"order":3,"actor":"Clinic staff","action":"Communicates the result and, if DED is present, gives scripted referral education.","handoff_to":"Eye care provider (for DED-positive patients)","control":"Scripted educational intervention"}],"decision_rights":"The AI makes the screening diagnosis autonomously (its FDA De Novo authorization basis for adults is diagnosis without human oversight). In the youth trial, the AI output was the result communicated to the patient and drove the referral decision, with a retina-specialist overread of every image as a safety layer because the device is not cleared for under-22s. Treatment decisions remain with eye care providers.","exception_path":"Insufficient image quality after 3 attempts triggers referral for eye care; a 'DED present' output triggers scripted referral to an eye care provider for a dilated exam.","removed_work":["Separate screening appointment scheduling and travel","Pharmacologic dilation (unnecessary in youth)","Referral coordination and chase-up for the screening itself","Specialist time spent on screening-negative patients"],"outcomes":[{"metric":"Diabetic eye exam completion within 6 months (care-gap closure)","baseline":"22% (18/82) in the control arm (scripted ECP referral plus education)","result":"100% (81/81) in the intervention arm; difference 78 percentage points (95% CI 69-87), p<0.001; no significant differences by race, ethnicity, SES, or education","period":"6-month post-randomization window; enrolled 2021-11-24 to 2022-06-06, follow-up completed 2022-12-06","scale":"164 randomized youth aged 8-21 with type 1 or type 2 diabetes at a single academic pediatric diabetes center (two sites); 41% minority groups; 47% Medicaid","attribution_caveat":"Peer-reviewed RCT under trial conditions. Outcome is process completion, not vision preserved. Control-arm denominator appears as n=83 in the abstract and 18/82 (Table 3, n=163) in the body; both round to 22%.","evidence_label":"verified"},{"metric":"Follow-through with an eye care provider when indicated","baseline":"22% in the control arm","result":"64% (16/25) in the intervention arm (difference 42 points, 95% CI 21-63), p<0.001","period":"6-month window","scale":"25 intervention participants with a 'DED present' result","attribution_caveat":"Denominators differ by design (intervention: DED-positive participants completing ECP follow-up; control: participants completing the ECP exam), so the comparison is not like-for-like; the paper defines the secondary outcome this way explicitly.","evidence_label":"verified"},{"metric":"DED screening adherence in the routine pediatric deployment (pre-trial)","baseline":"49% baseline adherence","result":"95% after autonomous AI implementation; 310 exams in the first year; 85.7% sensitivity and 79.3% specificity vs level-2 reference","period":"First year of pediatric deployment (2018 onward)","scale":"310 youth with diabetes at the JHM multidisciplinary pediatric diabetes center","attribution_caveat":"Pre/post observational, single center, reported in a peer-reviewed implementation review by the operating team.","evidence_label":"verified"}],"executive_lesson":"The binding constraint on screening was never diagnostic capacity; it was the separate appointment. Moving an autonomous, minute-long diagnostic into a visit the patient already attends closed a care gap that referral-plus-education could not, and it did so without introducing racial, ethnic, or socioeconomic disparities.","anti_pattern":"Citing the 100% vs 22% trial result as proof of operational performance; the trial proves efficacy under RCT conditions, and the separate routine-care deployment evidence is what proves durability.","questions_for_leaders":["Which of our screening or compliance gaps are actually attendance problems that point-of-care automation could eliminate?","Where would we accept an off-label AI deployment with a specialist overread backstop, and who signs that protocol?","Screening completion is a process metric; what is our path to the outcome metric (here, vision preserved)?"],"collections":["queue-eliminated","regulated-autonomy","embodied-work"],"bundle_version":"1.0.0","bundle_fingerprint":"sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93","canonical_url":"https://brianletort.ai/transformations/johns-hopkins-access"}