A comparative study of diagnostic accuracy in 3026 pleural biopsies and matched pleural effusion cytology with clinical correlation: A methodological issue

Aim After reading the article by Ivan K. Poon et al. Certain issues regarding the methodology must be addressed. Result First, it was not sufficient using only sensitivity and specificity. Second, we think the receiver operator characteristic curve (ROC) as well as the index of area under curve (AUC...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Cancer medicine (Malden, MA) MA), 2023-02, Vol.12 (4), p.4363-4364
Hauptverfasser: Zhang, Wen‐Li, Fang, Xiang, Qiu, Jian‐Qing, Cui, Tao
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Aim After reading the article by Ivan K. Poon et al. Certain issues regarding the methodology must be addressed. Result First, it was not sufficient using only sensitivity and specificity. Second, we think the receiver operator characteristic curve (ROC) as well as the index of area under curve (AUC) could better be demonstrated in this, it was not sufficient using only sensitivity and specificity. Second, we think the receiver operator characteristic curve (ROC) as well as the index of area under curve (AUC) could better be demonstrated in this article for a more comprehensive understanding of this diagnostic test accuracy when comparing different tests. Conclusion With using more comprehensive indexes including Youden's index, LR and AUC, the conclusions of Ivan K. Poon and his colleagues would be sodified. The comparison of SP and SE in various methods using a fixed cut‐point was affected by point selection. So we proposed the utilization of AUC for displaying the overall capabilities of those methods regardless of cut‐points.
ISSN:2045-7634
2045-7634
DOI:10.1002/cam4.5525