The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
Help: Registration number
ACTRN12626000596303
Ethics application status
Help: Ethics application status
Approved
Date submitted
Help: Date submitted
2/04/2026
Date registered
Help: Date registered
12/05/2026
Date last updated
Help: Date last updated
12/05/2026
Date data sharing statement initially provided
Help: Data sharing statement date
12/05/2026
Type of registration
Help: Type of registration
Prospectively registered

Titles & IDs
Public title
Can a computer program help identify stroke patients faster in the Emergency Department? A study of automated triage note analysis.
Scientific title
Diagnostic Accuracy of a Pre-Trained Algorithm for Automated Code Stroke Pathway Identification in Emergency Department Triage Notes Compared to Neurologist Review and Clinical Outcome
Additional trial identifiers [1] 316946 0
Nil
Universal Trial Number (UTN)
Trial acronym
SCANSTROKE
Related trial records

Health condition
Health condition(s) studied:
Acute ischaemic stroke 341382 0
Haemorrhagic stroke 341383 0
Stroke mimic 341384 0
Condition category
Condition code
Stroke 337451 337451 0 0
Ischaemic
Stroke 337452 337452 0 0
Haemorrhagic
Emergency medicine 337453 337453 0 0
Other emergency care

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of the intervention or exposure
A pre-trained natural language algorithm will be applied retrospectively to de-identified Emergency Department triage notes from Monash Medical Centre. The algorithm analyses free-text triage notes for stroke-related indicators and produces a binary classification output (activate Code Stroke pathway: yes/no) with a generated justification. The algorithm will observe recorded patient symptoms and observations such as blood pressure and GCS then determine likelihood of fulfilling code stroke pathway. Stateless processesing is used - the model receives only the current patient's data and produces an output without any information being carried over from previous patients. Therefore each patients data will be processed by the model individually. The algorithm is locally deployed on secure departmental infrastructure, requires no internet access, and performs no model training on patient data. The duration of (retrospective) observation covers ED presentations over a 1 month period between 14/9/2023 and 2/10/2023.

Data collection will occur in April 2026 and be completed within one month.

Given the observational design results will not alter clinical outcomes and the analysis is retrospective.
Intervention code [1] 333671 0
Diagnosis / Prognosis
Description of the comparator or control
Two comparators are used:

(primary) Neurologist labeled diagnosis (reference comparator)
(secondary) Actual ED Code Stroke activations as documented in the medical record.
Control group
Active

Outcomes
Primary outcome [1] 344965 0
Sensitivity of a Natural language processing algorithm for Code Stroke pathway activation, compared against Neurologist diagnosis of confirmed stroke.
Timepoint [1] 344965 0
At completion of language processing algorithm analysis
Primary outcome [2] 345320 0
Sensitivity of a Natural language processing algorithm compared against historical presence of actived stroke code pathway.
Timepoint [2] 345320 0
At completion of language processing algorithm analysis
Secondary outcome [1] 460350 0
AUC-ROC of a Natural language processing algorithm for Code Stroke pathway activation, compared against Neurologist diagnosis of confirmed stroke.
Timepoint [1] 460350 0
At completion of language processing algorithm analysis
Secondary outcome [2] 461289 0
AUC-ROC of a Natural language processing algorithm compared against historical presence of actived stroke code pathway.
Timepoint [2] 461289 0
At completion of language processing algorithm analysis

Eligibility
Key inclusion criteria
-Adult patients (greater than or equal to 18 years) presenting to Emergency Department within the past 5 years
-Presenting complaint documented in ED triage notes
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Healthy volunteers
No
Key exclusion criteria
-Triage notes that are incomplete, missing, or contain insufficient free-text for algorithm processing
-Presentations where triage documentation is unavailable in the electronic system
-Paediatric presentations (under 18 years)

Study design
Purpose
Screening
Duration
Cross-sectional
Selection
Defined population
Timing
Retrospective
Statistical methods / analysis
Diagnostic accuracy of the algorithm will be assessed against each reference standard independently and reported as sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), with 95% confidence intervals calculated using the Wilson score method.

Discrimination will be assessed using the area under the receiver operating characteristic curve (AUC-ROC) with confidence intervals derived using the DeLong method.

Given the clinical priority of minimising missed stroke activations, sensitivity against the primary reference standard will be the principal outcome metric. Agreement between algorithm outputs and each reference standard will be quantified using Cohen's Kappa and Gwet's AC1 to account for anticipated prevalence imbalance in the cohort. Where the algorithm produces probabilistic confidence outputs alongside binary classifications, calibration will be assessed using calibration curves and the Brier Score. Subgroup analyses will be conducted by discharge diagnosis category, presenting symptom type, triage note length, and time of presentation where sample size permits. A systematic qualitative analysis of false negative and false positive cases relative to the primary reference standard will be performed to identify linguistic and documentation patterns associated with algorithm error. An exploratory analysis will characterise discordance between the two reference standards — specifically presentations with confirmed stroke or TIA in whom no Code Stroke was activated, and presentations in whom a Code Stroke was activated but discharge diagnosis was a mimic or unrelated condition — to contextualise the baseline performance of current triage practice at the study site.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
Actual
Date of last participant enrolment
Anticipated
Actual
Date of last data collection
Anticipated
Actual
Sample size
Target
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 29360 0
Monash Medical Centre - Clayton campus - Clayton
Recruitment postcode(s) [1] 45680 0
3168 - Clayton

Funding & Sponsors
Funding source category [1] 321540 0
Self funded/Unfunded
Name [1] 321540 0
Country [1] 321540 0
Primary sponsor type
Individual
Name
Dr Michael Valente (Monash Health)
Address
Country
Australia
Secondary sponsor category [1] 324311 0
None
Name [1] 324311 0
Address [1] 324311 0
Country [1] 324311 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 320075 0
Monash Health Human Research Ethics Committee A
Ethics committee address [1] 320075 0
Ethics committee country [1] 320075 0
Australia
Date submitted for ethics approval [1] 320075 0
16/03/2026
Approval date [1] 320075 0
24/03/2026
Ethics approval number [1] 320075 0
QA/127358/MonH-2026-524757

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 149350 0
Dr Michael Valente
Address 149350 0
Neurology Department, Monash Health, 246 Clayton Rd, Clayton, 3168, Victoria
Country 149350 0
Australia
Phone 149350 0
+61 407301010
Fax 149350 0
Email 149350 0
Contact person for public queries
Name 149351 0
Michael Valente
Address 149351 0
Neurology Department, Monash Health, 246 Clayton Rd, Clayton, 3168, Victoria
Country 149351 0
Australia
Phone 149351 0
+61 03 9594 6666
Fax 149351 0
Email 149351 0
Contact person for scientific queries
Name 149352 0
Michael Valente
Address 149352 0
Neurology Department, Monash Health, 246 Clayton Rd, Clayton, 3168, Victoria
Country 149352 0
Australia
Phone 149352 0
+61 03 9594 6666
Fax 149352 0
Email 149352 0

Data sharing statement
Will the study consider sharing individual participant data?
No


What supporting documents are/will be available?

No Supporting Document Provided


Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.