Real-world validation of CLAIRE (previous name: SiMed): Verification of the functionality and process suitability of an AI system for recognizing pills in wards

Real-world validation of CLAIRE (previous name: SiMed): Verification of the functionality and process suitability of an AI system for recognizing pills in wards - CLAIRE-Validation

Status

Active, not recruiting

Phases

Unknown

Study type

Interventional

Source

DRKS

Registry ID

DRKS00040381

Enrollment

250

Registered

2026-05-19

Start date

2026-07-01

Completion date

Unknown

Last updated

2026-06-01

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Process validation and system validation in clinical practice

Interventions

Group 1: •Process Analysis: AI-supported dispensing process in the clinical setting

medication recognition and error analysis by CLAIRE •Validation: Confirmation of the functionality of CLAIRE through follow-up checks Pretest for phase 3: Based on real data from the hospital informa

Eligibility

Sex/Gender

All

Age

18 Years to No maximum

Inclusion criteria

Inclusion criteria: •Inpatient in the wards at the University Hospital Graz •Age = 18 years •Signed and dated Informed Consent •Willingness and ability to participate

Exclusion criteria

Exclusion criteria: •Missing or invalid IC •Age < 18 years •Mental illness •Delirium/Dementia •Unresponsive •Language barriers in explaining the study •Active legal adult representative

Design outcomes

Primary

Measure	Time frame
Primary outcome measure CLAIRE consists of hardware and software components and, with the support of artificial intelligence (AI), compares the medicines listed in doctors’ prescriptions (data source: electronic medical records) – which are subsequently dispensed by a healthcare professional – with the contents of the dispenser. The primary objective of the Phase 3 clinical trial is to evaluate the accuracy of CLAIRE’s medicine recognition. The aim is to achieve a medication error rate of zero following AI-supported dispensing with CLAIRE. As part of the study, a total of 250 dispensers are to be monitored using CLAIRE. The relevant data will be exported from the technical system documentation and analysed. Supplementary observations and study-relevant information will be documented by the study team using Case Report Forms (CRF).	—

Measure

Time frame

Primary outcome measure CLAIRE consists of hardware and software components and, with the support of artificial intelligence (AI), compares the medicines listed in doctors’ prescriptions (data source: electronic medical records) – which are subsequently dispensed by a healthcare professional – with the contents of the dispenser. The primary objective of the Phase 3 clinical trial is to evaluate the accuracy of CLAIRE’s medicine recognition. The aim is to achieve a medication error rate of zero following AI-supported dispensing with CLAIRE. As part of the study, a total of 250 dispensers are to be monitored using CLAIRE. The relevant data will be exported from the technical system documentation and analysed. Supplementary observations and study-relevant information will be documented by the study team using Case Report Forms (CRF).

—

Secondary

Measure	Time frame
Secondary outcome measures - Overall medication recognition rate - Error detection and types of errors: Pills in the wrong compartment Additional pills in the dispenser (incorrect dose) Missing pills in the dispenser Incorrect pills identified - Correct dispenser label recognized per patient - Correct medication packaging assigned to the correct prescription - External validation matches with CLAIRE results - Frequency of technical system errors or scanning errors of SiMed in the practical use The parameters are recorded automatically in the background (technical report). All other observations are documented using a CRF (paper form).	—

Measure

Time frame

Secondary outcome measures - Overall medication recognition rate - Error detection and types of errors: Pills in the wrong compartment Additional pills in the dispenser (incorrect dose) Missing pills in the dispenser Incorrect pills identified - Correct dispenser label recognized per patient - Correct medication packaging assigned to the correct prescription - External validation matches with CLAIRE results - Frequency of technical system errors or scanning errors of SiMed in the practical use The parameters are recorded automatically in the background (technical report). All other observations are documented using a CRF (paper form).

—

Countries

Austria

Contacts

Public ContactLars-Peter Kamolz

Medizinische Universität Graz - Univ. Klinik für Chirurgie

lars.kamolz@medunigraz.at+43 316 385-14685

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Jun 11, 2026