Pathologist Computer-Aided Diagnostic Scoring of Tumor Cell Fraction: A Swiss National Study.

Publication

Pathologist Computer-Aided Diagnostic Scoring of Tumor Cell Fraction: A Swiss National Study.

Journal Paper/Review - Sep 22, 2023

Frei Ana Leni, Oberson Raphaël, Baumann Elias, Perren Aurel, Grobholz Rainer, Lugli Alessandro, Dawson Heather, Abbet Christian, Lertxundi Ibai, Reinhard Stefan, Mookhoek Aart, Feichtinger Johann, Sarro Rossella, Gadient Gallus, Dommann-Scherrer Corina, Barizzi Jessica, Berezowska Sabina, Glatz Katharina, Dertinger Susanne, Banz Yara, Schönegg René, Rubbia-Brandt Laura, Fleischmann Achim, Saile Guenter, Mainil-Varlet Pierre, Biral Ruggero, Giudici Luca, Soltermann Alex, Chaubert Audrey Baur, Stadlmann Sylvia, Diebold Joachim, Egervari Kristof, Bénière Charles, Saro Francesca, Janowczyk Andrew, Zlobec Inti

Citation

Frei A, Oberson R, Baumann E, Perren A, Grobholz R, Lugli A, Dawson H, Abbet C, Lertxundi I, Reinhard S, Mookhoek A, Feichtinger J, Sarro R, Gadient G, Dommann-Scherrer C, Barizzi J, Berezowska S, Glatz K, Dertinger S, Banz Y, Schönegg R, Rubbia-Brandt L, Fleischmann A, Saile G, Mainil-Varlet P, Biral R, Giudici L, Soltermann A, Chaubert A, Stadlmann S, Diebold J, Egervari K, Bénière C, Saro F, Janowczyk A, Zlobec I. Pathologist Computer-Aided Diagnostic Scoring of Tumor Cell Fraction: A Swiss National Study. Mod Pathol 2023; 36:100335.

Type

Journal Paper/Review (English)

Journal

Mod Pathol 2023; 36

Publication Date

Sep 22, 2023

Issn Electronic

1530-0285

Pages

100335

Brief description/objective

Tumor cell fraction (TCF) estimation is a common clinical task with well-established large interobserver variability. It thus provides an ideal test bed to evaluate potential impacts of employing a tumor cell fraction computer-aided diagnostic (TCFCAD) tool to support pathologists' evaluation. During a National Slide Seminar event, pathologists (n = 69) were asked to visually estimate TCF in 10 regions of interest (ROIs) from hematoxylin and eosin colorectal cancer images intentionally curated for diverse tissue compositions, cellularity, and stain intensities. Next, they re-evaluated the same ROIs while being provided a TCFCAD-created overlay highlighting predicted tumor vs nontumor cells, together with the corresponding TCF percentage. Participants also reported confidence levels in their assessments using a 5-tier scale, indicating no confidence to high confidence, respectively. The TCF ground truth (GT) was defined by manual cell-counting by experts. When assisted, interobserver variability significantly decreased, showing estimates converging to the GT. This improvement remained even when TCFCAD predictions deviated slightly from the GT. The standard deviation (SD) of the estimated TCF to the GT across ROIs was 9.9% vs 5.8% with TCFCAD (P < .0001). The intraclass correlation coefficient increased from 0.8 to 0.93 (95% CI, 0.65-0.93 vs 0.86-0.98), and pathologists stated feeling more confident when aided (3.67 ± 0.81 vs 4.17 ± 0.82 with the computer-aided diagnostic [CAD] tool). TCFCAD estimation support demonstrated improved scoring accuracy, interpathologist agreement, and scoring confidence. Interestingly, pathologists also expressed more willingness to use such a CAD tool at the end of the survey, highlighting the importance of training/education to increase adoption of CAD systems.