Scan-AId: Artificial Intelligence Support in Novice's Decision-making for Assessing Ultrasound Fetal Weight Estimation - A Randomized Trial
| Registry | NCT06232187, acronym scan-AId |
|---|
| Status | Unknown |
|---|
| Phase | Not applicable |
|---|
| Study type | Interventional; randomized, parallel, masking: single, diagnostic |
|---|
| Enrollment | 75 (Estimated) |
|---|
| Sponsor | Copenhagen Academy for Medical Education and Simulation (Other); collaborators: Slagelse Hospital, Technical University of Denmark |
|---|
| Conditions | Fetal Weight, Ultrasound |
|---|
| Interventions | Artificial Intelligence feedback for ultrasound EFW standard plane images (Behavioral) |
|---|
| Dates | start 2024-02-14, primary completion 2024-06-30 (estimated), completion 2024-09-01 (estimated), first posted 2024-01-30, last updated 2024-05-10 |
|---|
| Locations | 1 site in Denmark |
|---|
Related Papers Related Patents Related Grants Related Experts
All trials for Fetal Weight All trials for Ultrasound All trials of Artificial Intelligence feedback for ultrasound EFW standard plane images
-Summary
The SCAN-AID study is a prospective, randomized, controlled, and unblinded study that compares the performance of novices in ultrasound fetal weight estimation. The study evaluates the impact of two levels of AI support: a straightforward black box AI and a more detailed explainable AI.
Expand all sections · Collapse all
+Detailed description
The goal of this randomized controlled clinical trial is to learn which type of artificial intelligence (AI) effects the diagnostic accuracy of ultrasound estimation of fetal weight (EFW), when performed by novices, in this study represented by medical students.
The study's objectives are:
* Which type of artificial intelligence support system works for novices in improving the ultrasound fetal weight diagnostic accuracy?
* Does the artificial intelligence improve image quality, evaluate the cognitive load placed on participants when utilizing AI support, and is the AI system usable for novices?
Participants will be tasked with conducting an ultrasound Estimated Fetal Weight (EFW) using either a simple black box AI or a detailed explainable AI feedback system. The AI systems will assist participants in determining if they have captured the appropriate image for EFW. The outcomes will then be compared to those of a control group.
Ultrasound procedures will be performed on pregnant women with fetuses at a gestational age of 28-42 weeks, who have previously undergone an EFW by an expert sonographer or doctor at the clinic within 5 days days leading up to the examinationday. One participant of each randomization arm, will perfrom an EFW on the same pregnant woman.
+Arms and interventions
| Arm | Type | Interventions | Description |
|---|
| Feedback Group 1 (FG1) | Experimental | Behavioral: Artificial Intelligence feedback for ultrasound EFW standard plane images | Participatns in FG1 will receive basic black box AI support, with simple explanation like "standard plane", "non standard plane" or "off plane". |
| Feedback Group 2 (FG2) | Experimental | Behavioral: Artificial Intelligence feedback for ultrasound EFW standard plane images | Participants in FG2 will receive explainable AI support, with more elaborate description of the anatomical structures and segmentation of the anatomy. |
| Control group (CG) | No intervention | | Participants in the CG will have a standard plane poster to help guide them to the EFW ultrasound standard plane images. |
- Artificial Intelligence feedback for ultrasound EFW standard plane images (Behavioral): AI feedback in two levels, in aid of the participants, to obtain the right standardplane images used in fetal ultrasound EFW calculation.
+Outcome measures
Primary outcomes
- Diagnostic accuracy [15 minutes]: The accuracy in each group was defined as the percentage difference between estimated fetal weight and the sonographer expert EFW
Secondary outcomes
- Image Quality [5 minutes pr. participant]: Salomon criteria score is used to rate the image quality. Points are given depending on the number of landmarks present, quality of the image optimization and caliper.placements.
Minimum: 1 Maximum: 18. A higher score indicates a better image quality.
Other outcomes
- The AI system usability [5 minutes]: The participants will be asked to answer a questionnaire: System Usability Scale (SUS), which is used to evaluate the AI feedback system.
Min 1 Maximum 100. A higher score indicating better system usability.
- Measurement of the reaction time [5 minutes]: Measurements of the participants reaction time will a measurement for the cognitive load.
The reaction time will be measured as a secondary task while the participants are performing the ultrasound scan.
+Eligibility
Sex: All, ages: 18 Years to any, accepts healthy volunteers
Ultrasound novice participants:
Inclusion Criteria:
* Medical students with no former fetal or abdominal ultrasound training.
* The participants will have to understand spoken and written Danish or English.
Exclusion Criteria:
• Medical students who received formal fetal or abdominal training prior to the inclusion in this study.
Pregnant women;
Inclusion Criteria:
* The participants will have to understand spoken and written Danish or English.
* BMI < 30
* Gestational age: 28-42
Exclusion Criteria:
* Age > 40 years
* Fefal anomaly
* Oligohydramnion
* Gestational Diabetes, Diabetes type 1 or 2.
+Locations (1)
| Facility | City | Country | Status |
|---|
| Rigshospitalet | Copenhagen | Denmark | |
+Publications
- Andreasen LA, Tabor A, Norgaard LN, Rode L, Gerds TA, Tolsgaard MG. Detection of growth-restricted fetuses during pregnancy is associated with fewer intrauterine deaths but increased adverse childhood outcomes: an observational study. BJOG. 2021 Jan;128(1):77-85. doi: 10.1111/1471-0528.16380. Epub 2020 Jul 27. PMID 32588532
- Andreasen LA, Tabor A, Norgaard LN, Taksoe-Vester CA, Krebs L, Jorgensen FS, Jepsen IE, Sharif H, Zingenberg H, Rosthoj S, Sorensen AL, Tolsgaard MG. Why we succeed and fail in detecting fetal growth restriction: A population-based study. Acta Obstet Gynecol Scand. 2021 May;100(5):893-899. doi: 10.1111/aogs.14048. Epub 2021 Jan 12. PMID 33220065
- Andreasen LA, Feragen A, Christensen AN, Thybo JK, Svendsen MBS, Zepf K, Lekadir K, Tolsgaard MG. Multi-centre deep learning for placenta segmentation in obstetric ultrasound with multi-observer and cross-country generalization. Sci Rep. 2023 Feb 8;13(1):2221. doi: 10.1038/s41598-023-29105-x. PMID 36755050
- Nicholls D, Sweet L, Hyett J. Psychomotor skills in medical ultrasound imaging: an analysis of the core skill set. J Ultrasound Med. 2014 Aug;33(8):1349-52. doi: 10.7863/ultra.33.8.1349. PMID 25063399
- Govaerts MJ, Schuwirth LW, Van der Vleuten CP, Muijtjens AM. Workplace-based assessment: effects of rater expertise. Adv Health Sci Educ Theory Pract. 2011 May;16(2):151-65. doi: 10.1007/s10459-010-9250-7. Epub 2010 Sep 30. PMID 20882335
- Tolsgaard MG, Pusic MV, Sebok-Syer SS, Gin B, Svendsen MB, Syer MD, Brydges R, Cuddy MM, Boscardin CK. The fundamentals of Artificial Intelligence in medical education research: AMEE Guide No. 156. Med Teach. 2023 Jun;45(6):565-573. doi: 10.1080/0142159X.2023.2180340. Epub 2023 Mar 2. PMID 36862064
- Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019 Jan;25(1):44-56. doi: 10.1038/s41591-018-0300-7. Epub 2019 Jan 7. PMID 30617339
- Tolsgaard MG, Boscardin CK, Park YS, Cuddy MM, Sebok-Syer SS. The role of data science and machine learning in Health Professions Education: practical applications, theoretical contributions, and epistemic beliefs. Adv Health Sci Educ Theory Pract. 2020 Dec;25(5):1057-1086. doi: 10.1007/s10459-020-10009-8. Epub 2020 Nov 3. PMID 33141345
- Degallier-Rochat S, Kurpicz-Briki M, Endrissat N, Yatsenko O. Human augmentation, not replacement: A research agenda for AI and robotics in the industry. Front Robot AI. 2022 Oct 4;9:997386. doi: 10.3389/frobt.2022.997386. eCollection 2022. No abstract available. PMID 36267424
- Vasey B, Novak A, Ather S, Ibrahim M, McCulloch P. DECIDE-AI: a new reporting guideline and its relevance to artificial intelligence studies in radiology. Clin Radiol. 2023 Feb;78(2):130-136. doi: 10.1016/j.crad.2022.09.131. PMID 36639172
- Cruz Rivera S, Liu X, Chan AW, Denniston AK, Calvert MJ; SPIRIT-AI and CONSORT-AI Working Group. Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension. Lancet Digit Health. 2020 Oct;2(10):e549-e560. doi: 10.1016/S2589-7500(20)30219-3. Epub 2020 Sep 9. PMID 33328049
- Liu X, Cruz Rivera S, Moher D, Calvert MJ, Denniston AK; SPIRIT-AI and CONSORT-AI Working Group. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension. Lancet Digit Health. 2020 Oct;2(10):e537-e548. doi: 10.1016/S2589-7500(20)30218-1. Epub 2020 Sep 9. PMID 33328048
- Salomon LJ, Alfirevic Z, Da Silva Costa F, Deter RL, Figueras F, Ghi T, Glanc P, Khalil A, Lee W, Napolitano R, Papageorghiou A, Sotiriadis A, Stirnemann J, Toi A, Yeo G. ISUOG Practice Guidelines: ultrasound assessment of fetal biometry and growth. Ultrasound Obstet Gynecol. 2019 Jun;53(6):715-723. doi: 10.1002/uog.20272. PMID 31169958
- Hadlock FP. Sonographic estimation of fetal age and weight. Radiol Clin North Am. 1990 Jan;28(1):39-50. PMID 2404304
- Borsci S, Federici S, Lauriola M. On the dimensionality of the System Usability Scale: a test of alternative measurement models. Cogn Process. 2009 Aug;10(3):193-7. doi: 10.1007/s10339-009-0268-9. Epub 2009 Jun 30. PMID 19565283
- Bloch R, Norman G. Generalizability theory for the perplexed: a practical introduction and guide: AMEE Guide No. 68. Med Teach. 2012;34(11):960-92. doi: 10.3109/0142159X.2012.703791. PMID 23140303