AI-Powered Segmentation of Invasive Carcinoma Regions in Breast Cancer Immunohistochemical Whole-Slide Images

Abstract

Aims: The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions.

Methods and results: In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists.

Conclusions: Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system.

Keywords: IHC quantification; Ki-67; artificial intelligence; breast cancer; invasive carcinoma.

Figure 1

Overview of the proposed approach for invasive carcinoma segmentation in breast cancer IHC-stained WSIs. The methodology includes the following steps: (1) Creation of an epithelial tissue segmentation dataset; (2) Training the segmentation model; (3) Thorough evaluation of the model’s segmentation performance on the test set; (4) Selection of the best-performing model for the application of new WSI inference.

Figure 2

Two-stage multi-scale segmentation model training framework. During Training Stage 1, we utilize semi-supervised learning to train the initial segmentation model. In Training Stage 2, the main focus is on training the multi-scale fusion modules.

Figure 3

Performance of the proposed method on the test set based on tumor type (DCIS or IC) and staining type (HER2, ER, PR, Ki-67).

Figure 4

Segmentation results of pure invasive carcinoma areas under various staining types. (a) Ki-67, 15%; (b) ER, negative; (c) HER2, 2+; (d) ER, positive; (e) HER2, 1+; (f) PR, positive. Rows 1–3 represent images, model predictions, and ground truth, respectively.

Figure 5

Segmentation results of pure DCIS areas (a,b) and areas with a mixture of DCIS and invasive carcinoma (c–f) under various staining types. Rows 1–3 represent images, model predictions, and ground truth, respectively.

Figure 6

Segmentation results from some special cases. (a) Pure lobular area; (b) area with a mixture of DCIS and invasive carcinoma with lymphocytic infiltration; (c) area with a mixture of DCIS and invasive carcinoma with lobular; (d) invasive carcinoma area with lymphocytic infiltration. Rows 1–3 represent images, model predictions, and ground truth, respectively.

Figure 7

Processing steps for WSI-level Ki-67 quantification.

Figure 8

Correlation plots of Ki-67 indices under different conditions.

Figure 9

Box plot of Ki-67 index errors for different tumor area masks.