Vertebral compression fractures at abdominal CT: underdiagnosis, undertreatment, and evaluation of an artificial intelligence (AI) algorithm
Authors' conclusion
The AI algorithm had high accuracy for the detection of VCFs and could be very useful in increasing the detection rate of VCFs, as there was a substantial underdiagnosis of VCFs.
Study context
- Population
- Patients over age 60 yr with an abdominal CT between January 18, 2019 and January 18, 2020
- Participants
- 1,112 participants
- Outcome
- AI algorithm performance for VCF detection; prevalence of reported and unreported VCFs
Main finding
As reported in the abstract
VCF prevalence, radiologist reporting rate and AI detection of moderate and severe (grade 2-3) VCF
| Group | Sensitivity | Specificity | PPV | NPV |
|---|---|---|---|---|
| AI algorithm | 85.2% | 92.3% | 57.8% | 98.1% |
AI algorithm
Sensitivity85.2%
Specificity92.3%
PPV57.8%
NPV98.1%
- Patients with a VCF
- 187 (16.8%)
- Incident VCF
- 62
- Previously unknown prevalent VCF
- 49
- Radiologist reporting rate of these vertebral compression fractures (VCFs)
- 30% (33/111)
Observational data: associations, not effects.
Further findings
Osteoporosis treatment started within a year in patients with reported VCFs
- Started osteoporosis treatment
- 3 of 30
Original publication
Vertebral compression fractures at abdominal CT: underdiagnosis, undertreatment, and evaluation of an AI algorithm. J Bone Miner Res. Published 20 June 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →