PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
October 15, 2024npj Digital MedicineOpen Access

Prospective clinical evaluation of deep learning for ultrasonographic screening of abdominal aortic aneurysms

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Abdominal aortic aneurysm often remains undetected until rupture due to limited access to diagnostic ultrasound.

Does deep learning-guided POCUS by novice nurses achieve comparable ultrasound scan quality to physician-performed scans for AAA screening in patients ≥65 years old?

Population

184 patients over 65

Comparison

DL-guided novice nurse scans vs physician-performed scans

Design

Prospective trial with blinded outcome assessment

Key result

Deep learning-guided point-of-care ultrasound by novice nurses achieved adequate scan quality in 87.5% of cases, comparable to the 91.3% achieved by experienced physicians (p=0.310).

Authors

ICI-Min ChiuTCTien‐Yu ChenYZYou-Cheng Zheng

Discussion

Loading...

Member takes

Overview

DL-guided novices matched physician AAA scan quality; supports broader screening access but remains hypothesis-generating pending RCTs.

Study Design

Type

Cross-Sectional (n=184)

Blinding

Single-blind (outcome assessors)

Multicenter

No

Structured PICO

Does deep learning-guided POCUS by novice nurses achieve comparable ultrasound scan quality to physician-performed scans for AAA screening in patients ≥65 years old?

P
Population
184 patients aged 65 and older visiting a cardiology outpatient clinic underwent abdominal aortic aneurysm screening by both novice nurses using deep learning guidance and experienced physicians.
I
Intervention
Deep learning (DL) guided point-of-care ultrasound (POCUS) screening for abdominal aortic aneurysm (AAA) performed by 10 novice nurses with no prior ultrasonography experience.
C
Comparator
Standard point-of-care ultrasound (POCUS) screening performed by experienced physicians without AI guidance.
O
Outcome
Ultrasound scan diagnostic quality, assessed by three blinded expert physicians.surrogate

Main Result

Absolute Event Rate: 87.5% vs 91.3%

p-value: p=0.310

Deep learning-guided POCUS enables novice nurses to acquire diagnostic-quality abdominal aortic ultrasound scans comparable to experienced physicians, potentially democratizing AAA screening.

Limitations

  • Conducted in a controlled clinical setting, which may not replicate remote or underserved areas
  • Relatively small number of patients and nurses limits generalizability
  • No control group of novice scanners without AI guidance
  • Conducted at a single tertiary academic hospital
  • Conducted in a controlled clinical setting in the outpatient clinic, which may not fully replicate POCUS use conditions in remote or underserved areas
  • Relatively small number of patients and nurses limits the assessment of generalizability
  • No control group for the nurse scanners (novices untrained with the algorithm)
  • Study location was one tertiary academic hospital

Cite This Study

Chiu et al. (2024) conducted a cross-sectional in Abdominal aortic aneurysm (n=184). Deep learning-guided point-of-care ultrasound by novice nurses vs. Physician-performed point-of-care ultrasound without AI guidance was evaluated on Ultrasound scan quality (rate of qualified videos) (p=0.310). Deep learning-guided point-of-care ultrasound by novice nurses achieved adequate scan quality in 87.5% of cases, comparable to the 91.3% achieved by experienced physicians (p=0.310).

synapsesocial.com/papers/6aa1401529de6bf265b56aa8https://doi.org/10.1038/s41746-024-01269-4
View Full Paper
Ask AI
Bookmark
Share