DDTNet: A dense dual-task network for tumor-infiltrating lymphocyte detection and segmentation in histopathological images of breast cancer

Zhang, Xiaoxuan; Zhu, Xiongfeng; Tang, Kai; Zhao, Yinghua; Lu, Zixiao; Feng, Qianjin

doi:10.1016/j.media.2022.102415

Cited by 34 publications

(22 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to clarifying the role of TILs, a comprehensive understanding of the tumor immune microenvironment can help to guide individualized immunotherapy, making it a hot spot in cancer immunotherapy research ( 20 ). Multiple studies have shown that TILs are often associated with better treatment response and prognosis ( 21 – 23 ). However, due to the heterogeneity of gastric cancer and the complexity of the immune microenvironment, PD-L1-based immune checkpoint inhibitors have limited benefits in the treatment of gastric cancer ( 24 ).…”

Section: Discussionmentioning

confidence: 99%

CD3D Is an Independent Prognostic Factor and Correlates With Immune Infiltration in Gastric Cancer

Shi

et al. 2022

Front. Oncol.

View full text Add to dashboard Cite

The protein encoded by CD3D is part of the T-cell receptor/CD3 complex (TCR/CD3 complex) and is involved in T-cell development and signal transduction. Previous studies have shown that CD3D is associated with prognosis and treatment response in breast, colorectal, and liver cancer. However, the expression and clinical significance of CD3D in gastric cancer are not clear. In this study, we collected 488 gastric cancer tissues and 430 paired adjacent tissues to perform tissue microarrays (TMAs). Then, immunohistochemical staining of CD3D, CD3, CD4, CD8 and PD-L1 was conducted to investigate the expression of CD3D in gastric cancer and the correlation between the expression of CD3D and tumor infiltrating lymphocytes (TILs) and PD-L1. The results showed that CD3D was highly expressed in gastric cancer tissues compared with paracancerous tissues (P<0.000). Univariate and multivariate analyses showed that CD3D was an independent good prognostic factor for gastric cancer (P=0.004, HR=0.677, 95%CI: 0.510-0.898 for univariate analyses; P=0.046, HR=0.687, 95%CI: 0.474-0.994 for multivariate analyses). In addition, CD3D was negatively correlated with the tumor location, Borrmann type and distant metastasis (P=0.012 for tumor location; P=0.007 for Borrmann type; P=0.027 for distant metastasis). In addition, the expression of CD3D was highly positively correlated with the expression of CD3, CD4, CD8, and PD-L1, and the combination of CD3D with CD3, CD4, CD8 and PD-L1 predicted the best prognosis (P=0.043). In summary, CD3D may play an important regulatory role in the tumor immune microenvironment of gastric cancer and may serve as a potential indicator of prognosis and immunotherapy response.

show abstract

Section: Discussionmentioning

confidence: 99%

CD3D Is an Independent Prognostic Factor and Correlates With Immune Infiltration in Gastric Cancer

Shi

et al. 2022

Front. Oncol.

View full text Add to dashboard Cite

show abstract

“…The reported Spearman correlation between computational sTIL and pathologist consensus was high (R = 0.73), greater than the reported interpathologist correlation (R = 0.66). DDTNet by Zhang et al 52 allowed end-to-end lymphocyte detection and segmentation in parallel and reached accuracies of up to 0.90 and 0.80, respectively. The quick, objective, and accurate nature of AI in TIL assessment suggested their viability for more prevalent use under the discretion of pathologists.…”

Section: T U M O U R -I N F I L T R a T I N G L Y M P H O C Y T E S (...mentioning

confidence: 99%

Artificial intelligence in breast cancer histopathology

Chan

Cheng

et al. 2022

Histopathology

View full text Add to dashboard Cite

This is a review on the use of artificial intelligence for digital breast pathology. A systematic search on PubMed was conducted, identifying 17,324 research papers related to breast cancer pathology. Following a semimanual screening, 664 papers were retrieved and pursued. The papers are grouped into six major tasks performed by pathologists—namely, molecular and hormonal analysis, grading, mitotic figure counting, ki‐67 indexing, tumour‐infiltrating lymphocyte assessment, and lymph node metastases identification. Under each task, open‐source datasets for research to build artificial intelligence (AI) tools are also listed. Many AI tools showed promise and demonstrated feasibility in the automation of routine pathology investigations. We expect continued growth of AI in this field as new algorithms mature.

show abstract

“…With the continuous development of CNN models in recognition tasks, especially the ResNet model (He et al, 2016) and the DenseNet model (Huang et al, 2017), it has been confirmed that the CNN model has a high accuracy and universality for feature extraction in classification tasks. In terms of improving the overall accuracy of the detection model, a deeper CNN can be used as the backbone network to extract the image features (Zhang et al, 2022). Attention mechanism such as SENet (Hu et al, 2018) is used to improve the sensitivity of the model to channel features.…”

Section: Related Workmentioning

confidence: 99%

“…In recent years, object detection technology has been widely used in pathology (Janowczyk and Madabhushi, 2016), especially in blood cells detection (Yang et al, 2017;Pan et al, 2018;Fujita and Han, 2020). Detecting blood cells can assist diagnosing many kinds of diseases, such as diagnosing breast cancer by detecting mitosis or lymphocytes (Cire ş an et al, 2013;Zhang et al, 2022). Object detection technology is constantly applied into application scenarios of medical image processing, and thus bringing more commercial value.…”

Section: Introductionmentioning

confidence: 99%

Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection

Liu²,

Chu³

et al. 2022

Front. Physiol.

View full text Add to dashboard Cite

Object detection technology has been widely used in medical field, such as detecting the images of blood cell to count the changes and distribution for assisting the diagnosis of diseases. However, detecting small objects is one of the most challenging and important problems especially in medical scenarios. Most of the objects in medical images are very small but influential. Improving the detection performance of small objects is a very meaningful topic for medical detection. Current researches mainly focus on the extraction of small object features and data augmentation for small object samples, all of these researches focus on extracting the feature space of small objects better. However, in the training process of a detection model, objects of different sizes are mixed together, which may interfere with each other and affect the performance of small object detection. In this paper, we propose a method called pixel level balancing (PLB), which takes into account the number of pixels contained in the detection box as an impact factor to characterize the size of the inspected objects, and uses this as an impact factor. The training loss of each object of different size is adjusted by a weight dynamically, so as to improve the accuracy of small object detection. Finally, through experiments, we demonstrate that the size of objects in object detection interfere with each other. So that we can improve the accuracy of small object detection through PLB operation. This method can perform well with blood cell detection in our experiments.

show abstract

DDTNet: A dense dual-task network for tumor-infiltrating lymphocyte detection and segmentation in histopathological images of breast cancer

Cited by 34 publications

References 23 publications

CD3D Is an Independent Prognostic Factor and Correlates With Immune Infiltration in Gastric Cancer

CD3D Is an Independent Prognostic Factor and Correlates With Immune Infiltration in Gastric Cancer

Artificial intelligence in breast cancer histopathology

Small Object Detection via Pixel Level Balancing With Applications to Blood Cell Detection

Contact Info

Product

Resources

About