Affordable Access

deepdyve-link
Publisher Website

A 'Real-Life' Experience on Automated Digital Image Analysis of FGFR2 Immunohistochemistry in Breast Cancer.

Authors
  • Braun, Marcin1
  • Piasecka, Dominika1, 2
  • Bobrowski, Mateusz3
  • Kordek, Radzislaw1
  • Sadej, Rafal2
  • Romanska, Hanna M1
  • 1 Department of Pathology, Chair of Oncology, Medical University of Lodz, 92-213 Lodz, Poland. , (Poland)
  • 2 Department of Molecular Enzymology and Oncology, Intercollegiate Faculty of Biotechnology, University of Gdansk and Medical University of Gdansk, 80-211 Gdansk, Poland. , (Poland)
  • 3 Sysmex Polska Sp. z o.o., 02-486 Warszawa, Poland. , (Poland)
Type
Published Article
Journal
Diagnostics
Publisher
MDPI AG
Publication Date
Dec 07, 2020
Volume
10
Issue
12
Identifiers
DOI: 10.3390/diagnostics10121060
PMID: 33297384
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

We present here an assessment of a 'real-life' value of automated machine learning algorithm (AI) for examination of immunohistochemistry for fibroblast growth factor receptor-2 (FGFR2) in breast cancer (BC). Expression of FGFR2 in BC (n = 315) measured using a certified 3DHistech CaseViewer/QuantCenter software 2.3.0. was compared to the manual pathologic assessment in digital slides (PA). Results revealed: (i) substantial interrater agreement between AI and PA for dichotomized evaluation (Cohen's kappa = 0.61); (ii) strong correlation between AI and PA H-scores (Spearman r = 0.85, p < 0.001); (iii) a small constant error and a significant proportional error (Passing-Bablok regression y = 0.51 × X + 29.9, p < 0.001); (iv) discrepancies in H-score in cases of extreme (strongest/weakest) or heterogeneous FGFR2 expression and poor tissue quality. The time of AI was significantly longer (568 h) than that of the pathologist (32 h). This study shows that the described commercial machine learning algorithm can reliably execute a routine pathologic assessment, however, in some instances, human expertise is essential.

Report this publication

Statistics

Seen <100 times