Artificial Intelligence for Ultrasound Training

Age: Any Age

Sex: Any

Trial Phase: Academic

Sponsor: Stanford University

Disqualifiers: Ultrasound elective, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: 1

150 Participants Needed

Trial Summary

Will I have to stop taking my current medications?

The trial protocol does not specify whether you need to stop taking your current medications.

What data supports the effectiveness of the treatment Ultrasound with Artificial Intelligence Enabled?

Research shows that using artificial intelligence (AI) with ultrasound can improve the accuracy of diagnosing conditions like breast cancer by reducing false positives (incorrectly identifying a condition when it's not present) and unnecessary biopsies. This suggests that AI can make ultrasound more reliable and efficient. ^{1 2 3 4 5}

Is ultrasound with artificial intelligence generally safe for humans?

Ultrasound contrast agents, which are sometimes used in ultrasound procedures, are generally safe and well-tolerated in humans. Serious adverse reactions are rare, and most side effects are mild and resolve on their own. ^{2 6 7 8 9}

How does the AI-enabled ultrasound treatment differ from other ultrasound treatments?

The AI-enabled ultrasound treatment is unique because it uses a deep-learning algorithm to guide novice operators in acquiring high-quality ultrasound images in real-time, which is not typically available in standard ultrasound procedures. This approach helps improve image acquisition and diagnostic accuracy, making it more accessible and efficient compared to traditional methods that rely heavily on operator expertise. ^{2 5 10 11 12}

What is the purpose of this trial?

Point-of care-ultrasonography has the potential to transform healthcare delivery through its diagnostic and therapeutic utility. Its use has become more widespread across a variety of clinical settings as more investigations have demonstrated its impact on patient care. This includes the use of point-of-care ultrasound by trainees, who are now utilizing this technology as part of their diagnostic assessments of patients. However, there are few studies that examine how efficiently trainees can learn point-of-care ultrasound and which training methods are more effective. The primary objective of this study is to assess whether artificial intelligence systems improve internal medicine interns' knowledge and image interpretation skills with point-of-care ultrasound. Participants shall be randomized to receive personal access to handheld ultrasound devices to be used for learning with artificial intelligence vs devices with no artificial intelligence. The primary outcome will assess their interpretive ability with ultrasound images/videos. Secondary outcomes will include rates of device usage and performance on quizzes.

Research Team

Andre D Kumar, MD

Principal Investigator

Stanford University

Eligibility Criteria

This trial is for internal medicine residents who haven't taken an ultrasound elective. They'll be using handheld ultrasound devices on general inpatient wards to see if artificial intelligence (AI) helps them learn better.

Inclusion Criteria

Internal medicine residents rotating on the general inpatient wards service

Exclusion Criteria

Residents who had taken an ultrasound elective offered by our residency program

Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Training

Participants are randomized to receive personal access to handheld ultrasound devices with or without artificial intelligence for learning purposes

4 weeks

Weekly training sessions

Assessment

Participants' interpretive ability with ultrasound images/videos is assessed

1 day

1 visit (in-person)

Follow-up

Participants are monitored for device usage rates and performance on quizzes

4 weeks

Treatment Details

Interventions

• Ultrasound with Artificial Intelligence Enabled
• Ultrasound without Artificial Intelligence Enabled

Trial Overview The study tests whether AI improves interns' ability to understand and interpret medical images from ultrasounds. Interns will use either AI-enabled or regular ultrasound devices, and their skills will be compared.

Participant Groups

Group I: Artificial Intelligence Group Experimental Treatment 1 Intervention

Group II: Non Artificial Intelligence Group Active Control 1 Intervention

Who Is Running the Clinical Trial?

Stanford University

Lead Sponsor

Trials

2,527

Recruited

17,430,000+

Findings from Research

In a study involving 23,188 abdominal examinations using the ultrasound contrast agent SonoVue, only 29 adverse events (AEs) were reported, indicating a very low incidence of complications.

The overall rate of serious adverse events was extremely low at 0.0086%, and SonoVue demonstrated a good safety profile, comparable to or better than other contrast agents used in radiology.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

The safety of Sonovue in abdominal applications: retrospective analysis of 23188 investigations. Piscaglia, F., Bolondi, L.[2022]

Microfocused Ultrasound with Visualization (MFU-V) has a well-established safety profile, showing only mild and transient side effects like tenderness and redness, with rare adverse events typically linked to improper technique.

When used correctly, MFU-V treatments are safe, with most side effects resolving quickly and without lasting effects, indicating its efficacy in cosmetic dermatology.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Review of the safety profile for microfocused ultrasound with visualization. Hitchcock, TM., Dobke, MK.[2022]

In a large retrospective analysis involving over 78,000 doses of ultrasound contrast agents (Definity and Optison), only 0.01% of patients experienced severe adverse reactions, indicating a strong safety profile for these agents.

The study found that severe reactions were primarily in outpatients, with no serious events reported in critically ill patients, suggesting that ultrasound contrast agents are safe for use in a variety of clinical settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

The safety of deFinity and Optison for ultrasound image enhancement: a retrospective analysis of 78,383 administered contrast doses. Wei, K., Mulvagh, SL., Carson, L., et al.[2022]

Ultrasound contrast agents are generally safe for clinical use, with serious adverse reactions being rare and most adverse events being minor and self-resolving, such as headache and nausea.

The Contrast Media Safety Committee found that while intolerance to some components can occur, generalized allergy-like reactions are infrequent, emphasizing the overall well-tolerated nature of these agents.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Safety of ultrasound contrast agents. Jakobsen, JA., Oyen, R., Thomsen, HS., et al.[2020]

References

1. England pubmed.ncbi.nlm.nih.gov

Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams. [2023]

2. England pubmed.ncbi.nlm.nih.gov

Deep Learning Pitfall: Impact of Novel Ultrasound Equipment Introduction on Algorithm Performance and the Realities of Domain Adaptation. [2022]

3. Switzerland pubmed.ncbi.nlm.nih.gov

Enhancing Performance of Breast Ultrasound in Opportunistic Screening Women by a Deep Learning-Based System: A Multicenter Prospective Study. [2022]

4. Germany pubmed.ncbi.nlm.nih.gov

Clinical value of artificial intelligence in thyroid ultrasound: a prospective study from the real world. [2023]

5. United States pubmed.ncbi.nlm.nih.gov

Utility of a Deep-Learning Algorithm to Guide Novices to Acquire Echocardiograms for Limited Diagnostic Use. [2022]

6. England pubmed.ncbi.nlm.nih.gov

The safety of Sonovue in abdominal applications: retrospective analysis of 23188 investigations. [2022]

7. England pubmed.ncbi.nlm.nih.gov

Review of the safety profile for microfocused ultrasound with visualization. [2022]

8. United States pubmed.ncbi.nlm.nih.gov

The safety of deFinity and Optison for ultrasound image enhancement: a retrospective analysis of 78,383 administered contrast doses. [2022]

9. Germany pubmed.ncbi.nlm.nih.gov

Safety of ultrasound contrast agents. [2020]

10. United States pubmed.ncbi.nlm.nih.gov

A Survey of Deep-Learning Applications in Ultrasound: Artificial Intelligence-Powered Ultrasound for Improving Clinical Workflow. [2020]

11. Ireland pubmed.ncbi.nlm.nih.gov

Artificial intelligence in ultrasound. [2021]

12. China pubmed.ncbi.nlm.nih.gov

Artificial intelligence-aided ultrasound in renal diseases: a systematic review. [2023]