Accuracy of Ex Vivo Semiautomatic Segmentation of Urinary Stone Size in Computed Tomography Compared With Manual Size Estimation in Radiographic Correlation

To evaluate the accuracy of semiautomated segmentation of urinary stone size in computed tomography (CT) compared with manual measurement. A total of 103 patients (32f, 71m ; mean age 52 years±18 that were diagnosed with urolithiasis and collected stones received standardized ex vivo CT-scans and ra...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Urology (Ridgewood, N.J.) N.J.), 2019-01, Vol.123, p.70-75
Hauptverfasser: Niemann, Tilo, Jerjen, I., Hefermehl, L., Wang, Z., Stampanoni, M., Lehmann, K., Kubik-Huch, R.A.
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To evaluate the accuracy of semiautomated segmentation of urinary stone size in computed tomography (CT) compared with manual measurement. A total of 103 patients (32f, 71m ; mean age 52 years±18 that were diagnosed with urolithiasis and collected stones received standardized ex vivo CT-scans and radiography of the stones. Stone size was segmented semiautomatically using commercial software (syngo.via, Siemens, Germany) and compared with manual caliper measurement on digital radiography. Mean size was 4.4 mm in CT and 4.6 mm in radiography. Depending on number of stones analyzed per patient, estimation of stone size showed moderate to excellent correlation for both methods. There was no significant difference in overall size measurement. Semiautomatic segmentation of urinary stone size in CT is possible and reduces measurement errors, allowing more precise estimation especially for smaller concrements. Neighboring stones may hamper segmentation of stone size.
ISSN:0090-4295
1527-9995
DOI:10.1016/j.urology.2018.06.044