ML-assisted Phantom-Less QCT at 100 kVp Improves Osteoporosis Screening on Ultra-Low-Dose Chest CT

For 100 kVp ultra-low-dose chest CT, a machine learning phantom-less QCT model reduced bone mineral density error vs standard 120 kVp model (2.39 vs 16.68 mg/cm³, p<0.0001) and agreed with DXA in 88.8%.

• External validation on an independent CT system yielded mean BMD error of -0.11±4.39 mg/cm³ for the 100 kVp model versus 13.78±5.10 mg/cm³ for the standard 120 kVp model.
• Mean effective dose was 0.82±0.19 mSv.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.