Heart Diseases
Conditions
Brief summary
The purpose of this study is to assess the effect of artificial intelligence algorithms on image quality in echocardiography.
Detailed description
The study population will be recruited from appointed patients at the Clinic of Cardiology, St. Olavs Hospital. After being informed about the study, all patients giving informed consent and meeting the eligibility requirements will undergo their standard clinical echocardiographic exam performed by a sonographer at the clinic. Two additional examinations will then be performed by a different sonographer and an expert cardiologist, respectively. In one of the study arms the sonographer randomized to perform the second exam will use the AI algorithm (intervention).
Interventions
OTHERAI algorithm for apical foreshortening in echocardiography
The algorithm is based on artificial intelligence, giving the sonographer performing the echocardiographic exam real-time feedback on left ventricular apical foreshortening.The algorithm is developed using deep learning techniques by technologists at the Department of Circulation and Medical Imaging, NTNU.
Sponsors
St. Olavs Hospital
Helse Nord-Trøndelag HF
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
OTHER
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Capacity to consent
Exclusion criteria
* Indication for echocardiographic contrast agents
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Left ventricular apical foreshortening | 0 days | Apical foreshortening will be evaluated in a blinded manner by echocardiography experts post hoc. The two study arms will be compared. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Inter-observer variation of left ventricular size | 0 days | Throughout this study, the sonographers and expert cardiologist will perform measurements based on their individual echocardiographic images. Differences in measurements will be evaluated to determine inter-observer variation between sonographers and cardiologists. |
| Inter-observer variation of left ventricular function | 0 days | Throughout this study, the sonographers and expert cardiologist will perform measurements based on their individual echocardiographic images. Differences in measurements will be evaluated to determine inter-observer variation between sonographers and cardiologists. |
| Validation of AI algorithm to detect foreshortening | 0 days | Comparison of AI algorithm to detect foreshortening with manual reference |
| Image quality | 0 days | Image quality will be evaluated in a blinded manner by echocardiography experts post hoc. A standardized scoring system for image quality in echocardiography developed locally will be utilized for evaluation with score 1-6, where 1 is minimum and 6 is maximum score. The two study arms will be compared. |
| Inter-observer variation of left atrial function | 0 days | Throughout this study, the sonographers and expert cardiologist will perform measurements based on their individual echocardiographic images. Differences in measurements will be evaluated to determine inter-observer variation between sonographers and cardiologists. |
| Development of AI algorithm for real-time assessment of left atrium | 0 days | Cardiac structures as left atrial (LA) lumen, LA wall, LA appendix, mitral valve will be annotated and used as reference. |
| Variation by methodology | 0 days | Variation in measurement by the mode of measurement |
| Inter-observer variation of left atrial size | 0 days | Throughout this study, the sonographers and expert cardiologist will perform measurements based on their individual echocardiographic images. Differences in measurements will be evaluated to determine inter-observer variation between sonographers and cardiologists. |
Countries
Norway
Outcome results
None listed