Dataset Viewer
Auto-converted to Parquet Duplicate
uid
stringclasses
10 values
dataset_id
stringclasses
1 value
care_setting
stringclasses
3 values
observation_abnormality_signal
stringclasses
2 values
escalation_signal
stringclasses
3 values
response_signal
stringclasses
4 values
response_timing_signal
stringclasses
5 values
clinical_summary
stringclasses
10 values
question
stringclasses
1 value
ground_truth_label
stringclasses
2 values
ground_truth_rationale
stringclasses
7 values
outcome_signal
stringclasses
3 values
source_citation
stringclasses
1 value
COER-001
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
abnormal
escalated
responded
on_time
Abnormal obs triggered escalation and timely response.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Obs matched by escalation and response.
stable
Synthetic
COER-002
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
abnormal
not_escalated
not_responded
none
Abnormal obs but no escalation or response.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
incoherent
Missed deterioration response.
harm_risk
Synthetic
COER-003
clinical-observation-escalation-response-coherence-risk-v0.1
ED
abnormal
escalated
responded
timely
ED response aligned.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Aligned.
managed
Synthetic
COER-004
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
abnormal
escalated
responded
delayed
Escalated but response delayed.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
incoherent
Timing mismatch.
harm_risk
Synthetic
COER-005
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
normal
not_required
not_required
none
Normal obs no escalation needed.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Appropriate.
stable
Synthetic
COER-006
clinical-observation-escalation-response-coherence-risk-v0.1
ICU
abnormal
escalated
responded
on_time
ICU response aligned.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Aligned.
stable
Synthetic
COER-007
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
abnormal
escalated
wrong_team
none
Abnormal obs escalated to wrong team.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
incoherent
Wrong routing.
harm_risk
Synthetic
COER-008
clinical-observation-escalation-response-coherence-risk-v0.1
Ward
abnormal
escalated
responded
late
Late response to abnormal obs.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
incoherent
Late response.
harm_risk
Synthetic
COER-009
clinical-observation-escalation-response-coherence-risk-v0.1
ED
normal
not_required
not_required
none
Normal obs.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Appropriate.
stable
Synthetic
COER-010
clinical-observation-escalation-response-coherence-risk-v0.1
ICU
abnormal
escalated
responded
on_time
Aligned ICU response.
Is abnormal observation vs escalation response coherent. Answer coherent or incoherent.
coherent
Aligned.
managed
Synthetic

What this repo is for

Detect when abnormal vital signs

do not trigger timely escalation and response

Examples you can use NEWS2 high score but no review hypoxia recorded but no oxygen escalation tachycardia trend ignored

You use it to flag deterioration risk before arrest or ICU transfer

Downloads last month
12