AI and Digital Health for Rare Diseases: Promising but Thin Evidence
A scoping review of 245 studies on digital health in lysosomal storage disorders finds real momentum — and a field still running almost entirely on small, single-center observational data, zero completed RCTs, and a heavy bias toward two diseases out of dozens.
Explanation
Lysosomal storage disorders (LSDs) are a group of roughly 70 rare inherited metabolic diseases where the body can't break down certain cellular waste products. They're a useful proxy for rare diseases broadly: hard to diagnose, complex to manage, and chronically under-resourced. Only about 5% of all rare diseases have approved treatments.
This scoping review — covering 1,751 records from 2015–2024, with 245 included — is the first systematic map of where AI and connected care (telemedicine, remote monitoring, patient-engagement tools) actually sit in LSD care today.
The headline finding: there's genuine activity, but the evidence base is shallow. AI is mostly being used for diagnostic decision support and patient phenotyping (grouping patients by disease characteristics). Connected care tools are concentrated in telemedicine and remote monitoring. The bulk of the literature — nearly half — clusters around screening and diagnosis. Treatment optimization, rehabilitation, and end-of-life care are barely touched.
Two diseases, Gaucher and Fabry, dominate the literature. The other 68-odd LSDs are largely invisible in the digital health research record.
The methodological picture is blunt: no completed randomized controlled trials, no LSD-specific systematic reviews, and outcomes are mostly healthcare delivery metrics rather than what patients actually experience. Algorithms are rarely described transparently enough to replicate or audit.
Why does this matter now? The EU's rare disease policy agenda is accelerating, and digital health infrastructure decisions — registries, interoperability standards, AI procurement — are being made today. A field that can't yet demonstrate effectiveness risks either being ignored in those decisions or, worse, having unvalidated tools quietly embedded in clinical workflows. The review's practical output is a gap map that can anchor expert consensus and direct the next wave of multicenter, prospective studies toward the questions that actually need answering.
This scoping review (PICO-informed data-charting, no risk-of-bias assessment by design) maps 245 records across AI, connected care (CC), and enabling digital health technologies (DHTs) in LSDs — a field where the diagnostic odyssey alone averages years and where enzyme replacement or substrate reduction therapies exist for only a fraction of subtypes.
The intervention breakdown: 40 records on AI-driven DHTs, 89 on CC, 144 on other enabling DHTs (registries, interoperable data systems, digital biomarkers), with multilabeling inflating counts. AI applications cluster around diagnostic decision support, phenotyping, and risk stratification — largely retrospective pattern recognition on imaging or biomarker data. CC is dominated by telemedicine and remote monitoring, unsurprisingly accelerated post-2020. No completed RCTs. No LSD-specific systematic reviews. The evidentiary floor is observational, single-center, and small-n.
The disease concentration problem is significant: Gaucher and Fabry together account for a disproportionate share of records, reflecting their relatively larger patient registries and longer therapeutic histories. Ultra-rare LSDs — where digital tools could arguably have the highest marginal value for diagnosis — are nearly absent from the literature.
Outcome domains skew toward healthcare delivery performance (process measures, diagnostic yield, workflow efficiency) rather than patient-reported outcomes or societal metrics like employment or caregiver burden. This is a known limitation in rare disease research generally, but it's particularly acute here because regulatory and payer decisions increasingly require patient-relevant endpoints.
The methodological gap that most limits translation: algorithm transparency. Few records describe model architecture, training data provenance, or validation cohorts in enough detail to assess generalizability or bias risk — a direct collision with EU AI Act requirements for high-risk medical AI systems now entering enforcement.
Key open questions the review surfaces but cannot answer: Do AI-assisted diagnostic tools meaningfully shorten the diagnostic odyssey in prospective settings? Can CC platforms sustain engagement in patient populations with high disease burden and complex comorbidities? What interoperability standards are actually being adopted across European reference networks for LSDs?
The review's value is as a structured gap map, not an effectiveness synthesis — and it's honest about that. The next falsifiable step is prospective multicenter trials with pre-registered algorithms and patient-centered primary endpoints. Watch whether EU rare disease research consortia (ERNs) pick this up as a prioritization framework.
Reality meter
Why this score?
Trust Layer Score basis
A detailed evidence breakdown is being added. For now, the score basis is the source list below and the reality meter above.
- 48 sources on file
- Avg trust 42/100
- Trust 40–95/100
Time horizon
Community read
Glossary
- Scoping review
- A systematic mapping of literature across a broad topic to identify gaps, types of evidence, and research trends, without formally assessing the quality or risk of bias in individual studies.
- Diagnostic odyssey
- The prolonged period of clinical investigation and specialist consultations that patients with rare or complex diseases experience before receiving a correct diagnosis.
- Enzyme replacement therapy (ERT)
- A treatment approach for genetic disorders where a functional enzyme is supplied to patients who lack or have deficient production of that enzyme.
- Substrate reduction therapy
- A treatment that reduces the accumulation of harmful metabolic byproducts in cells by inhibiting the enzymes that produce them, used in certain genetic storage disorders.
- Digital biomarkers
- Quantifiable physiological or behavioral indicators measured through digital devices or platforms that reflect disease status or progression.
- Algorithm transparency
- The degree to which the internal workings, training data, and decision-making logic of an artificial intelligence model can be understood and explained to users and regulators.
- EU AI Act
- European Union legislation that establishes risk-based requirements for artificial intelligence systems, including strict oversight for high-risk applications like medical diagnostics.
What's your read?
Your read shapes future topic weighting.
Your vote feeds topic weights, community direction and future prioritisation. Open community direction
Sources
- Tier 3 Artificial Intelligence, Connected Care, and Enabling Digital Health Technologies in Rare Diseases With a Focus on Lysosomal Storage Disorders: Scoping Review
- Tier 3 Latest AI News, Developments, and Breakthroughs | 2026 | News
- Tier 3 The 2025 AI Index Report | Stanford HAI
- Tier 3 Artificial Intelligence News -- ScienceDaily
- Tier 3 AI Developments That Changed Vibrational Spectroscopy in 2025 | Spectroscopy Online
- Tier 3 AI breakthrough cuts energy use by 100x while boosting accuracy | ScienceDaily
- Tier 3 Reuters AI News | Latest Headlines and Developments | Reuters
- Tier 3 Inside the AI Index: 12 Takeaways from the 2026 Report
- Tier 1 Human scientists trounce the best AI agents on complex tasks
- Tier 3 Sony AI Announces Breakthrough Research in Real-World Artificial Intelligence and Robotics - Sony AI
- Tier 3 This new brain-like chip could slash AI energy use by 70% | ScienceDaily
- Tier 3 State AI Laws – Where Are They Now? // Cooley // Global Law Firm
- Tier 3 AI Regulation: The New Compliance Frontier | Insights | Holland & Knight
- Tier 3 The White House’s National Policy Framework for Artificial Intelligence: what it means and what comes next | Consumer Finance Monitor
- Tier 3 Trump Administration Releases National AI Policy Framework | Morrison Foerster
- Tier 3 What President Trump’s AI Executive Order 14365 Means For Employers | Law and the Workplace
- Tier 3 Manatt Health: Health AI Policy Tracker - Manatt, Phelps & Phillips, LLP
- Tier 3 Battle for AI Governance: White House’s Plan to Centralize AI Regulation and States’ Continuous Opposition
- Tier 3 AI Omnibus: Trilogue Underway…What to Expect as Negotiations Progress | Insights | Ropes & Gray LLP
- Tier 3 AI Regulation News Today 2025: Latest Updates on EU AI Act, US Rules & Global Impact - Prime News Mag
- Tier 3 AI regulation set to become US midterm battleground | Biometric Update
- Tier 3 Top Large Language Models of 2025 | Best LLMs Compared
- Tier 3 Large language model - Wikipedia
- Tier 1 [2604.27454] Exploring Applications of Transfer-State Large Language Models: Cognitive Profiling and Socratic AI Tutoring
- Tier 3 Top 50+ Large Language Models (LLMs) in 2026
- Tier 3 The Best Open-Source LLMs in 2026
- Tier 3 10 Best LLMs of April 2026: Performance, Pricing & Use Cases
- Tier 3 Emerging applications of large language models in ecology and conservation science
- Tier 3 From Elicitation to Evolution: A Literature-Grounded, AI-Assisted Framework for Requirements Quality, Traceability, and Non-Functional Requirement Management | IJCSE
- Tier 3 Labor market impacts of AI: A new measure and early ...
- Tier 3 Tracking the Impact of AI on the Labor Market - Yale Budget Lab
- Tier 3 AI and Jobs: Labor Market Impact Echoes Past Tech Transitions | Morgan Stanley
- Tier 3 The Jobs AI Is Likely to Boost—and Those It May Disrupt | Goldman Sachs
- Tier 3 How will Artificial Intelligence Affect Jobs 2026-2030 | Nexford University
- Tier 3 Young People Are Falling Behind, but Not Because of AI - The Atlantic
- Tier 3 AI is getting better at your job, but you have time to adjust, according to MIT | ZDNET
- Tier 3 New Data Challenges AI Job Loss Narrative | Robert H. Smith School of Business
- Tier 3 The impact of AI on the labour market | Management & Marketing | Springer Nature Link
- Tier 3 AI's impact on the job market is starting to show up in the data
- Tier 3 AI speeds up prior auth, coding while driving higher costs for health systems: PHTI report
- Tier 3 AI-enabled Medical Devices Market Size, Share | Forecast [2034]
- Tier 3 Generative AI analyzes medical data faster than human research teams | ScienceDaily
- Tier 3 Rede Mater Dei de Saúde: Monitoring AI agents in the revenue cycle with Amazon Bedrock AgentCore | Artificial Intelligence
- Tier 3 Artificial Intelligence (AI) in Healthcare & Medical Field
- Tier 3 AI in Healthcare Market Rises 37.66% Healthy CAGR by 2035
- Tier 3 Here's how the data fed into medical AI can help — or hurt — health care | GBH
- Tier 3 Future of AI in Healthcare: Trends and Predictions for 2027 and Beyond
- Tier 3 2026 Conference
Optional Submit a prediction Optional: add your prediction on the core question if you like.
Prediction
Will a completed randomized controlled trial evaluating an AI or connected care intervention specifically in lysosomal storage disorders be published by the end of 2027?