Swift-Native MacOS DICOM Viewer Shows Drastically Lower Memory Footprint Than Horos

OpenDicomViewer, a new macOS DICOM viewer built in Swift, used 32% of the memory of Horos (9.6 vs 30.0 GB mean) while opening a 9.3 GB dataset, with a codebase >23 times smaller, potentially easing AI-assisted development after future external validation.

• Technical benchmark study opening the 19,022-file (9.3 GB) TCIA Pancreas-CT dataset on a Mac Studio; single-machine, no patient data.
• Mean peak memory: OpenDicomViewer 9.6±0.0 GB vs Horos 30.0±0.5 GB vs OsiriX MD 23.6±1.8 GB over three trials each; no system memory pressure reported for ODV.
• The full public Horos source tree contained 69.5 times as many source lines (900,763 lines) compared to ODV's 12,952 lines of Swift; the new viewer is not yet externally validated or FDA-cleared for clinical use.

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