HookNet: Multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images

Rijthoven, Mart van; Balkenhol, Maschenka; Siliņa, Karīna; Laak, Jeroen van der; Ciompi, Francesco

doi:10.1016/j.media.2020.101890

Cited by 137 publications

(105 citation statements)

References 21 publications

Supporting

Mentioning

104

Contrasting

Order By: Relevance

“…Generally, TLSs represent sites of lymphoid neogenesis characterized by mature dendritic cells in a T cell zone adjacent to a B cell follicle including a germinal center. Recently, investigators developed a semantic segmentation model for whole-slide histopathological images, named HookNet, which can segment TLSs and germinal centers in lung cancer [ 26 ]. Inflammatory BCa (IBC) is an aggressive form of this disease, and it shows higher expression of TLS signatures with higher sensitivity to immune checkpoint inhibitors [ 27 ].…”

Section: Discussionmentioning

confidence: 99%

Prognostic impact of tertiary lymphoid structures in breast cancer prognosis: a systematic review and meta-analysis

Zhang

Líu

et al. 2021

Cancer Cell Int

View full text Add to dashboard Cite

Background Tertiary lymphoid structures (TLSs), organizationally resemble lymph nodes, are frequently present in breast cancer (BCa). It is usually, but not always, associated with a positive prognosis or immunotherapy response in cancer patients. This meta-analysis was performed to assess the prognostic and clinical impact of TLSs in BCa. Methods We conducted a systematic search in PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Infrastructure, and WanFang Database to obtain eligible research data up to May 30, 2021. This meta-analysis is focusing on the studies evaluated the prognostic value of TLSs and the associated clinicopathologic indicators, related gene expression and survival. STATA software 16.0 software was used to assess the prognostic significance and clinical impact of TLSs. Results Nine studies involved with 2281 cases were incorporated in this meta-analysis, in which four of them evaluated the prognostic value of TLSs. There are 6 studies assessed the relationship of TLSs and 4 studies investigated the clinicopathologic parameters as well as the key gene expression, respectively. The results showed the presence of TLSs were predicting a better OS (HR = 0.61, 95% CI: 0.51–0.73, p < 0.001) and DFS (HR = 0.40, 95% CI: 0.17–0.93, p < 0.001) of BCa patients. It also revealed that the presence of TLSs was significantly correlated with tumor differentiation (p < 0.001), pTNM stage (p < 0.001), lymph node metastasis (p < 0.001), and TILs density (p < 0.001) of BCa, and the expression of Her2 (p < 0.001), ER (p < 0.001), PR (p < 0.001) and Ki67 (p = 0.009) of the tumor cell. Conclusion Our results indicated that high levels of TLSs could predict a favorable prognosis for BCa. Moreover, the TLSs were significantly correlated with the clinicopathological indicators and the critical gene expression of BCa, indicating its potential clinical impact on BCa patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Prognostic impact of tertiary lymphoid structures in breast cancer prognosis: a systematic review and meta-analysis

Zhang

Líu

et al. 2021

Cancer Cell Int

View full text Add to dashboard Cite

show abstract

“…Thanks to the identification of regions of interest, the automated detection of cells of interest within the tumor stroma (in addition to the detection of stained tumor cells and/or recognition of less dense/cohesive tissue), and the classification of immune cell clusters according to their degree of organization, the computer system generates a grading of TLS for each of these diseases ( 118 ). It has also been possible to establish multi-class segmentation of tissues in breast cancer and grading of TLS in lung cancers ( 119 ).…”

Section: Artificial Intelligence a Valuable Tool For Fundamental And Clinical Research On Tlsmentioning

confidence: 99%

Tumor-Associated Tertiary Lymphoid Structures: From Basic and Clinical Knowledge to Therapeutic Manipulation

et al. 2021

View full text Add to dashboard Cite

The tumor microenvironment is a complex ecosystem almost unique to each patient. Most of available therapies target tumor cells according to their molecular characteristics, angiogenesis or immune cells involved in tumor immune-surveillance. Unfortunately, only a limited number of patients benefit in the long-term of these treatments that are often associated with relapses, in spite of the remarkable progress obtained with the advent of immune checkpoint inhibitors (ICP). The presence of “hot” tumors is a determining parameter for selecting therapies targeting the patient immunity, even though some of them still do not respond to treatment. In human studies, an in-depth analysis of the organization and interactions of tumor-infiltrating immune cells has revealed the presence of an ectopic lymphoid organization termed tertiary lymphoid structures (TLS) in a large number of tumors. Their marked similarity to secondary lymphoid organs has suggested that TLS are an “anti-tumor school” and an “antibody factory” to fight malignant cells. They are effectively associated with long-term survival in most solid tumors, and their presence has been recently shown to predict response to ICP inhibitors. This review discusses the relationship between TLS and the molecular characteristics of tumors and the presence of oncogenic viruses, as well as their role when targeted therapies are used. Also, we present some aspects of TLS biology in non-tumor inflammatory diseases and discuss the putative common characteristics that they share with tumor-associated TLS. A detailed overview of the different pre-clinical models available to investigate TLS function and neogenesis is also presented. Finally, new approaches aimed at a better understanding of the role and function of TLS such as the use of spheroids and organoids and of artificial intelligence algorithms, are also discussed. In conclusion, increasing our knowledge on TLS will undoubtedly improve prognostic prediction and treatment selection in cancer patients with key consequences for the next generation immunotherapy.

show abstract

“…To incorporate more contextual information from large histopathological images into the networks, Graham et al [23] used multi-scale network to perform instance-based nuclei segmentation. Another idea to incorporate more contextual information was proposed by Rijthoven et al [24], where they proposed multi-resolution networks to segment and classify various tissue regions in nonsmall cell lung cancer and breast cancer samples. While these methods provide accurate results, training of these networks requires, however, the use of large datasets with pixel-precise labels.…”

Section: Introductionmentioning

confidence: 99%

Domain Adaptation-Based Deep Learning for Automated Tumor Cell (TC) Scoring and Survival Analysis on PD-L1 Stained Tissue Images

Kapil

Meier

Steele

et al. 2021

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

We report the ability of two deep learningbased decision systems to stratify non-small cell lung cancer (NSCLC) patients treated with checkpoint inhibitor therapy into two distinct survival groups. Both systems analyze functional and morphological properties of epithelial regions in digital histopathology whole slide images stained with the SP263 PD-L1 antibody. The first system learns to replicate the pathologist assessment of the Tumor Cell (TC) score with a cut-point for positivity at 25% for patient stratification. The second system is free from assumptions related to TC scoring and directly learns patient stratification from the overall survival time and event information. Both systems are built on a novel unpaired domain adaptation deep learning solution for epithelial region segmentation. This approach significantly reduces the need for large pixel-precise manually annotated datasets while superseding serial sectioning or re-staining of slides to obtain ground truth by cytokeratin staining. The capacity of the first system to replicate the TC scoring by pathologists is evaluated on 703 unseen cases, with an addition of 97 cases from an independent cohort. Our results show Lin's concordance values of 0.93 and 0.96 against pathologist scoring, respectively. The ability of the first and second system to stratify anti-PD-L1 treated patients is evaluated on 151 clinical samples. Both systems show similar stratification powers (first system: HR =0.539, p=0.004 and second system: HR=0.525, p=0.003) compared to TC scoring by pathologists (HR=0.574, p=0.01).

show abstract

HookNet: Multi-resolution convolutional neural networks for semantic segmentation in histopathology whole-slide images

Cited by 137 publications

References 21 publications

Prognostic impact of tertiary lymphoid structures in breast cancer prognosis: a systematic review and meta-analysis

Prognostic impact of tertiary lymphoid structures in breast cancer prognosis: a systematic review and meta-analysis

Tumor-Associated Tertiary Lymphoid Structures: From Basic and Clinical Knowledge to Therapeutic Manipulation

Domain Adaptation-Based Deep Learning for Automated Tumor Cell (TC) Scoring and Survival Analysis on PD-L1 Stained Tissue Images

Contact Info

Product

Resources

About