AI-guided DVT diagnosis in primary care: protocol for cohort with qualitative assessment
2024

AI-guided DVT Diagnosis in Primary Care

Sample size: 500 publication Evidence: high

Author Information

Author(s): Kerstin Nothnagel, Alastair Hay, Jessica Watson, Jonathan Banks

Primary Institution: Bristol Medical School (PHS), Bristol, United Kingdom

Hypothesis

Can AI-guided point-of-care ultrasound (POCUS) accurately diagnose deep vein thrombosis (DVT) when performed by non-specialists in primary care?

Conclusion

The study will evaluate the accuracy and acceptability of AI-guided POCUS for DVT diagnosis in primary care settings.

Supporting Evidence

  • The incidence of DVT is increasing as the global population ages.
  • AI-guided POCUS could allow any healthcare professional to perform DVT scans.
  • The study aims to assess the accuracy of AI-guided POCUS compared to standard ultrasound scans.

Takeaway

This study is looking at whether regular healthcare workers can use AI to help diagnose blood clots in people's legs, making it easier for patients to get checked without going to the hospital.

Methodology

The study will use a diagnostic test accuracy (DTA) approach and qualitative interviews to assess AI-guided POCUS in 500 individuals with suspected DVT.

Limitations

The study excludes housebound and frail patients and is conducted in a single geographical area, which may limit generalizability.

Participant Demographics

Participants will be recruited from primary care DVT clinics in the BNSSG region, with demographics such as age, sex, and postcode reported.

Statistical Information

Confidence Interval

95% confidence interval with a margin of error of ±5%

Digital Object Identifier (DOI)

10.3399/BJGPO.2024.0165

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication