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...
Gespeichert in:
Veröffentlicht in: | Cancer medicine (Malden, MA) MA), 2023-02, Vol.12 (4), p.4363-4364 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
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 |