Dataset Information

PiNET-An Automated Proliferation Index Calculator Framework for Ki67 Breast Cancer Images.

ABSTRACT: In this work, a novel proliferation index (PI) calculator for Ki67 images called piNET is proposed. It is successfully tested on four datasets, from three scanners comprised of patches, tissue microarrays (TMAs) and whole slide images (WSI), representing a diverse multi-centre dataset for evaluating Ki67 quantification. Compared to state-of-the-art methods, piNET consistently performs the best over all datasets with an average PI difference of 5.603%, PI accuracy rate of 86% and correlation coefficient R = 0.927. The success of the system can be attributed to several innovations. Firstly, this tool is built based on deep learning, which can adapt to wide variability of medical images-and it was posed as a detection problem to mimic pathologists' workflow which improves accuracy and efficiency. Secondly, the system is trained purely on tumor cells, which reduces false positives from non-tumor cells without needing the usual pre-requisite tumor segmentation step for Ki67 quantification. Thirdly, the concept of learning background regions through weak supervision is introduced, by providing the system with ideal and non-ideal (artifact) patches that further reduces false positives. Lastly, a novel hotspot analysis is proposed to allow automated methods to score patches from WSI that contain "significant" activity.

SUBMITTER: Geread RS

PROVIDER: S-EPMC7792768 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

piNET-An Automated Proliferation Index Calculator Framework for Ki67 Breast Cancer Images.

Geread Rokshana Stephny RS Sivanandarajah Abishika A Brouwer Emily Rita ER Wood Geoffrey A GA Androutsos Dimitrios D Faragalla Hala H Khademi April A

Cancers 20201222 1

In this work, a novel proliferation index (PI) calculator for Ki67 images called piNET is proposed. It is successfully tested on four datasets, from three scanners comprised of patches, tissue microarrays (TMAs) and whole slide images (WSI), representing a diverse multi-centre dataset for evaluating Ki67 quantification. Compared to state-of-the-art methods, piNET consistently performs the best over all datasets with an average PI difference of 5.603%, PI accuracy rate of 86% and correlation coef ...[more]

PMID: 33375043

Similar Datasets

Project description:BackgroundKi67 labeling index (LI) is critical for treatment options and prognosis evaluation in breast cancer. Visual assessment (VA) is widely used to assess Ki67 LI, but has some limitations. In this study, we compared the consistency between VA and automated digital image analysis (DIA) of Ki67 LI in breast cancer, and to evaluate the application value of DIA in Ki67 LI assessment.MethodsKi67 immunostained slides of 155 cases of primary invasive breast cancer were eyeballing assessed by five breast pathologists and automated digital image analyzed by one breast pathologist respectively. Two score methods, hot-spot score and average score, were used to choose score areas. The intra-class correlation coefficient (ICC) was used to analyze the consistency between VA and DIA, and Wilcoxon signed-rank test was used to compare the median of paired-difference between VA and DIA values.Results(1) A perfect agreement was demonstrated between VA and DIA of Ki67 LI by ICC analysis (P<0.0001) in the whole cohort. A perfect agreement between VA and DIA of Ki67 LI was also showed in G2-G3, ER positive/HER2 negative cases. Average score and hot-spot score methods both demonstrated a perfect concordance between VA and DIA of Ki67 LI. (2) All cases were classified into three groups by VA values (?10%, 11%-30% and >30% Ki67 LI). The concordance was relatively lower in intermediate Ki67 LI group (11%-30%) compared with high (>30%) Ki67 LI groups according to both methods. (3) All cases were classified into three groups by paired-difference (d) between VA values of hot-spot score and average score (d<5, 5?d<10, d?10) to evaluate the correlation between Ki67 staining distribution (heterogeneous or homogenous) and reproducibility of assessment. A perfect agreement was all demonstrated in three groups, and a slightly better Ki67 LI agreement between VA and DIA was indicated in homogenous staining slides than in heterogeneous staining ones. (4) VA values were relatively smaller than DIA values (average score: median of paired-difference -3.72; hot-spot score: median of paired-difference -9.12).ConclusionsAn excellent agreement between VA and DIA of Ki67 LI in breast cancer was demonstrated in the whole mixed cohort, suggesting that VA and DIA both could be used to assess Ki67 LI in clinical practice. Average score and hot-spot score methods both demonstrated a perfect concordance between VA and DIA of Ki67 LI. The almost perfect agreement between VA and DIA was observed in high Ki67 LI cases, displaying a homogenous staining pattern. The consistency between VA and DIA was relatively low in intermediate Ki67 LI group. The heterogeneity of tumors may slightly affect the concordance between VA and DIA of Ki67 LI. Assessment of VA provides lower Ki67 values than DIA, the biological importance of these values are not known at the moment.

Project description:Automated methods are needed to facilitate high-throughput and reproducible scoring of Ki67 and other markers in breast cancer tissue microarrays (TMAs) in large-scale studies. To address this need, we developed an automated protocol for Ki67 scoring and evaluated its performance in studies from the Breast Cancer Association Consortium. We utilized 166 TMAs containing 16,953 tumour cores representing 9,059 breast cancer cases, from 13 studies, with information on other clinical and pathological characteristics. TMAs were stained for Ki67 using standard immunohistochemical procedures, and scanned and digitized using the Ariol system. An automated algorithm was developed for the scoring of Ki67, and scores were compared to computer assisted visual (CAV) scores in a subset of 15 TMAs in a training set. We also assessed the correlation between automated Ki67 scores and other clinical and pathological characteristics. Overall, we observed good discriminatory accuracy (AUC = 85%) and good agreement (kappa = 0.64) between the automated and CAV scoring methods in the training set. The performance of the automated method varied by TMA (kappa range= 0.37-0.87) and study (kappa range = 0.39-0.69). The automated method performed better in satisfactory cores (kappa = 0.68) than suboptimal (kappa = 0.51) cores (p-value for comparison = 0.005); and among cores with higher total nuclei counted by the machine (4,000-4,500 cells: kappa = 0.78) than those with lower counts (50-500 cells: kappa = 0.41; p-value = 0.010). Among the 9,059 cases in this study, the correlations between automated Ki67 and clinical and pathological characteristics were found to be in the expected directions. Our findings indicate that automated scoring of Ki67 can be an efficient method to obtain good quality data across large numbers of TMAs from multicentre studies. However, robust algorithm development and rigorous pre- and post-analytical quality control procedures are necessary in order to ensure satisfactory performance.

Dataset Information

PiNET-An Automated Proliferation Index Calculator Framework for Ki67 Breast Cancer Images.

Publications

piNET-An Automated Proliferation Index Calculator Framework for Ki67 Breast Cancer Images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets