Methods & About
Data Sources
28 public sources registered for the twelve projects, each with its access terms and the limit on what it can support.
| Project | Sources | Access |
|---|---|---|
| 5-1 · AI drug discovery | 3 | Open · Open paper · Data access unverified |
| 8-1 · Synthetic data | 3 | Open and controlled files · Open · Credentialed access |
| 18-2 · Medication safety | 2 | Public listing · API key required · Credentialed access |
| 19-1 · Rare disease diagnosis | 3 | Open · Public aggregate data |
| 16-2 · Adaptive cognitive training | 2 | Application required · Membership required for most data |
| 16-3 · Dementia diagnosis | 3 | Account and data agreement · Application required · Membership required for most data |
| 18-1 · Fall early warning | 3 | Public research data · Credentialed access |
| 18-3 · Surgical safety | 3 | Public · Noncommercial terms · Dataset-specific conditions |
| 12-1 · Acute crisis signals | 1 | Credentialed access |
| 12-2 · Safety-plan transitions | 1 | Instrument documentation |
| 12-3 · Social isolation | 3 | Open paper · Public research data · Public aggregate data |
| 12-4 · Student crisis handoffs | 1 | Application for eligible Korean users |
Sources by project
5-1 · Structure-Based AI Small-Molecule Discovery
The PDB archive distributes structure files under CC0. Comparable experimental binding free energies are unavailable for some structures.
Activity and compound data are licensed CC BY-SA 3.0. Distinct activity endpoints and assay conditions require separate treatment.
The authors describe public access. The data server returned an access error during review. Compound lists lack measured binding or toxicity evidence.
8-1 · Early Boost for Core Health R&D Technology
Files are divided into open and controlled tiers. For controlled access, dbGaP approval and eRA Commons authentication are required.
Platforms and variable definitions vary by study. Verify the endpoint and subject-level linkage for each series.
Deidentified records still require credentials, training and a data use agreement. Sharing conditions apply to derived models and data.
18-2 · Medication Administration and Dispensing Safety
Provides basic identification attributes by product. Images captured in clinical settings and confirmed real-world errors require separate reference data.
Covers hospital admissions and critical care. Linking Korean product codes to pill images and identifying dispensing errors require separate evidence.
19-1 · Rare Disease Diagnostic Research
Includes conflicting interpretations from submitters. Review status and public release dates should be tracked independently.
Population frequencies and quality metadata describe the population sample. Case-level diagnosis requires its own reference.
Provides phenotype terms and gene annotations. Coverage gaps and release changes can affect rankings.
16-2 · Non-Pharmacological Dementia Research
Provides longitudinal clinical and cognitive assessments. Responses to the proposed training intervention require separate reference data.
Speech and transcripts provide context for speaking characteristics. Access and tasks vary by corpus.
16-3 · Multimodal Dementia Diagnostic Research
The longitudinal imaging, clinical and cognition cohort requires a NITRC account. Linkage to DementiaBank participants remains unverified.
Includes imaging, diagnosis and cognitive assessments. Confirm coverage by study phase and visit-level variables.
Audio, transcripts and diagnostic coverage vary by corpus. Matching the same people to scans remains unverified.
18-1 · Fall Prediction and Prevention
Contains enacted falls and everyday activities. Performance for spontaneous falls in hospitals requires hospital data.
Healthy volunteers performed falls and everyday activities. Verify impact time and observation coverage before each event.
Multicentre critical care records. A reference linking individual gait sensors to real fall time and impact remains unverified.
18-3 · Surgical Environment Error Prevention
CC BY-NC-SA 4.0 applies. Tool presence in visible frames covers a different endpoint from instrument movement throughout the room or retained objects.
Supports tool presence and segmentation in surgical video. Coverage is limited to the observable field.
The repository specifies CC BY-NC-SA 4.0. It provides phase, instrument and action labels, with views outside the abdominal cavity blurred.
12-1 · Suicide Prevention Research for People with Mental Illness
Contains deidentified discharge summaries and radiology reports. Longitudinal psychiatric notes and subsequent crisis labels require separate data.
12-2 · Suicide Prevention Research for Bereaved People and Attempt Survivors
Documents a public research assessment instrument. Deidentified emergency-department follow-up and subsequent attempt labels require separate data.
12-3 · Social Isolation and Suicide Prevention Research
A university student sensing and survey study. Reference outcomes for sustained isolation transitions and suicide-related endpoints remain unverified.
Laboratory-collected wearable measures of stress and affect. Longitudinal isolation data require a separate source.
Aggregated household types provide regional context. Individual isolation or crisis status requires separate measures.
12-4 · Student Suicide Prevention Research
Contains counselling data for children aged 7–12. Coverage of adolescent terminology, multi-turn crisis assessment and handoff labels remains unverified.