Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Jan 13;14(1):1283.
doi: 10.1038/s41598-024-51723-2.

AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer

Amanda Dy  1 , Ngoc-Nhu Jennifer Nguyen  2 , Julien Meyer  3 , Melanie Dawe  4 , Wei Shi  4 , Dimitri Androutsos  5 , Anthony Fyles  4 , Fei-Fei Liu  4 , Susan Done  4 , April Khademi  5   6   7   8   9
Affiliations

AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer

Amanda Dy et al. Sci Rep. .

Abstract

The Ki-67 proliferation index (PI) guides treatment decisions in breast cancer but suffers from poor inter-rater reproducibility. Although AI tools have been designed for Ki-67 assessment, their impact on pathologists' work remains understudied. 90 international pathologists were recruited to assess the Ki-67 PI of ten breast cancer tissue microarrays with and without AI. Accuracy, agreement, and turnaround time with and without AI were compared. Pathologists' perspectives on AI were collected. Using AI led to a significant decrease in PI error (2.1% with AI vs. 5.9% without AI, p < 0.001), better inter-rater agreement (ICC: 0.70 vs. 0.92; Krippendorff's α: 0.63 vs. 0.89; Fleiss' Kappa: 0.40 vs. 0.86), and an 11.9% overall median reduction in turnaround time. Most pathologists (84%) found the AI reliable. For Ki-67 assessments, 76% of respondents believed AI enhances accuracy, 82% said it improves consistency, and 83% trust it will improve efficiency. This study highlights AI's potential to standardize Ki-67 scoring, especially between 5 and 30% PI-a range with low PI agreement. This could pave the way for a universally accepted PI score to guide treatment decisions, emphasizing the promising role of AI integration into pathologist workflows.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Examples of TMA with no AI aid (left) and TMA with AI tool overlay and calculated proliferation index (PI) (right). The TMA shown is case 7.
Figure 2
Figure 2
Graphs of absolute PI error. (A) Illustrates the absolute PI error vs. each case. (B) Displays the mean absolute PI error vs. years of experience. (C) Depicts the mean absolute PI error vs. career stages. Asterisk: statistical significance was found between pathologists and pathologists with AI using the paired Wilcoxon signed-rank test.
Figure 3
Figure 3
(A) Linear Regression of pathologists’ scores with AI assistance. (B) Bland–Altman of pathologists’ scores with AI assistance. (C) Linear regression of pathologists’ scores without AI assistance. (D) Bland–Altman of pathologists’ scores without AI assistance.
Figure 4
Figure 4
Graphs of TATs displayed in seconds. (A) Illustrates the absolute TAT vs. each case, the average of all cases (All), and the average of all cases excluding Question 1 (All-Q1). (B) Represents the TAT vs. the sequential question pairs in the study. (C) Displays the mean TAT vs. years of experience. (D) Depicts the mean TAT vs. career stages. Asterisk: Statistical significance was found between pathologists and pathologists with AI using the paired Wilcoxon signed-rank test.
Figure 5
Figure 5
(A) Pathologists’ opinions on AI for Ki-67 assessments. (B) Pathologists’ opinions on their personal implementation timeline of AI into Ki-67 assessments. (C) Pathologists’ opinions on the routine implementation timeline of AI into Ki-67 assessments.
See this image and copyright information in PMC

References

    1. Soliman NA, Yussif SM. Ki-67 as a prognostic marker according to breast cancer molecular subtype. Cancer Biol. Med. 2016;13:496–504. doi: 10.20892/j.issn.2095-3941.2016.0066. - DOI - PMC - PubMed
    1. Yerushalmi R, Woods R, Ravdin PM, Hayes MM, Gelmon KA. Ki67 in breast cancer: Prognostic and predictive potential. Lancet Oncol. 2010;11:174–183. doi: 10.1016/S1470-2045(09)70262-1. - DOI - PubMed
    1. Urruticoechea A, Smith IE, Dowsett M. Proliferation marker Ki-67 in early breast cancer. J. Clin. Oncol. 2005;23:7212–7220. doi: 10.1200/JCO.2005.07.501. - DOI - PubMed
    1. Petrelli F, et al. Prognostic value of different cut-off levels of Ki-67 in breast cancer: A systematic review and meta-analysis of 64,196 patients. Breast Cancer Res. Treat. 2015;153:477–491. doi: 10.1007/s10549-015-3559-0. - DOI - PubMed
    1. Johnston SRD, et al. Abemaciclib plus endocrine therapy for hormone receptor-positive, HER2-negative, node-positive, high-risk early breast cancer (monarchE): Results from a preplanned interim analysis of a randomised, open-label, phase 3 trial. Lancet Oncol. 2023;24:77–90. doi: 10.1016/S1470-2045(22)00694-5. - DOI - PMC - PubMed