- Design
- Single-centre prospective feasibility study, paired scans
- Population
- 48 adults with inflammatory myopathy-associated ILD
- Primary outcome
- Agreement of deep-learning ILD quantification, low vs standard dose
- Effect
- Mean differences <1% for all patterns; CTDIvol 0.77 vs 5.40 mGy
Patients with idiopathic inflammatory myopathy and interstitial lung disease are scanned repeatedly. This single-centre prospective study gave 48 such patients two consecutive photon-counting CT scans, one standard-dose and one tin-filtered low-dose, and quantified disease with commercial deep-learning software.
The low-dose protocol used a CTDIvol of 0.77 mGy against 5.40 mGy. Mean differences between protocols were under 1% for every pattern, with slightly more ground-glass and slightly less fibrosis measured at low dose. Reticular extent correlated with FVC and DLCO equally well on both protocols. Ultrathin 0.4 mm reconstructions added little.
This is a feasibility study in one centre, using one vendor's scanner and one software package. It does not show that low-dose scans detect progression as reliably over time.
Where photon-counting CT is available, a low-dose protocol for quantitative follow-up looks reasonable to evaluate locally.
- A low-dose photon-counting protocol cut CTDIvol from 5.40 to 0.77 mGy.
- Quantified ILD patterns differed by less than 1% from standard dose.
- Correlations with FVC and DLCO were preserved at low dose.
- Expect slightly more ground-glass and less fibrosis to be measured at low dose.
- Use the same protocol across follow-up scans so changes are comparable.
Why it matters
Repeated CT is the price of ILD monitoring, and a seventh of the dose changes that trade-off for young patients.
Don't overread it
This single-centre feasibility study has not shown low dose tracks progression reliably over time.
The statistics, in plain English
Bland-Altman mean differences under 1% show close average agreement, but individual scans may differ more. Correlations with lung function changed by no more than 0.02.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for radiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free