Automated diagnosis of 7 canine skin tumors using machine learning on H&E-stained whole slide images.

Microscopic evaluation of hematoxylin and eosin-stained slides is still the diagnostic gold standard for a variety of diseases, including neoplasms. Nevertheless, intra- and interrater variability are well documented among pathologists. So far, computer assistance via automated image analysis has shown potential to support pathologists in improving accuracy and reproducibility of quantitative tasks. In this proof of principle study, we describe a machine-learning-based algorithm for the automated diagnosis of 7 of the most common canine skin tumors: trichoblastoma, squamous cell carcinoma, peripheral nerve sheath tumor, melanoma, histiocytoma, mast cell tumor, and plasmacytoma. We selected, digitized, and annotated 350 hematoxylin and eosin-stained slides (50 per tumor type) to create a database divided into training, n = 245 whole-slide images (WSIs), validation (n = 35 WSIs), and test sets (n = 70 WSIs). Full annotations included the 7 tumor classes and 6 normal skin structures. The data set was used to train a convolutional neural network (CNN) for the automatic segmentation of tumor and nontumor classes. Subsequently, the detected tumor regions were classified patch-wise into 1 of the 7 tumor classes. A majority of patches-approach led to a tumor classification accuracy of the network on the slide-level of 95% (133/140 WSIs), with a patch-level precision of 85%. The same 140 WSIs were provided to 6 experienced pathologists for diagnosis, who achieved a similar slide-level accuracy of 98% (137/140 correct majority votes). Our results highlight the feasibility of artificial intelligence-based methods as a support tool in diagnostic oncologic pathology with future applications in other species and tumor types.

SFRP1 Expression is Inversely Associated With Metastasis Formation in Canine Mammary Tumours.

BACKGROUND: Canine mammary tumours (CMTs) are the most frequent tumours in intact female dogs and show strong similarities with human breast cancer. In contrast to the human disease there are no standardised diagnostic or prognostic biomarkers available to guide treatment. We recently identified a prognostic 18-gene RNA signature that could stratify human breast cancer patients into groups with significantly different risk of distant metastasis formation. Here, we assessed whether expression patterns of these RNAs were also associated with canine tumour progression. METHOD: A sequential forward feature selection process was performed on a previously published microarray dataset of 27 CMTs with and without lymph node (LN) metastases to identify RNAs with significantly differential expression to identify prognostic genes within the 18-gene signature. Using an independent set of 33 newly identified archival CMTs, we compared expression of the identified prognostic subset on RNA and protein basis using RT-qPCR and immunohistochemistry on FFPE-tissue sections. RESULTS: While the 18-gene signature as a whole did not have any prognostic power, a subset of three RNAs: Col13a1, Spock2, and Sfrp1, together completely separated CMTs with and without LN metastasis in the microarray set. However, in the new independent set assessed by RT-qPCR, only the Wnt-antagonist Sfrp1 showed significantly increased mRNA abundance in CMTs without LN metastases on its own (p = 0.013) in logistic regression analysis. This correlated with stronger SFRP1 protein staining intensity of the myoepithelium and/or stroma (p < 0.001). SFRP1 staining, as well as ß-catenin membrane staining, was significantly associated with negative LN status (p = 0.010 and 0.014 respectively). However, SFRP1 did not correlate with ß-catenin membrane staining (p = 0.14). CONCLUSION: The study identified SFRP1 as a potential biomarker for metastasis formation in CMTs, but lack of SFRP1 was not associated with reduced membrane-localisation of ß-catenin in CMTs.

Pan-tumor T-lymphocyte detection using deep neural networks: Recommendations for transfer learning in immunohistochemistry.

The success of immuno-oncology treatments promises long-term cancer remission for an increasing number of patients. The response to checkpoint inhibitor drugs has shown a correlation with the presence of immune cells in the tumor and tumor microenvironment. An in-depth understanding of the spatial localization of immune cells is therefore critical for understanding the tumor's immune landscape and predicting drug response. Computer-aided systems are well suited for efficiently quantifying immune cells in their spatial context. Conventional image analysis approaches are often based on color features and therefore require a high level of manual interaction. More robust image analysis methods based on deep learning are expected to decrease this reliance on human interaction and improve the reproducibility of immune cell scoring. However, these methods require sufficient training data and previous work has reported low robustness of these algorithms when they are tested on out-of-distribution data from different pathology labs or samples from different organs. In this work, we used a new image analysis pipeline to explicitly evaluate the robustness of marker-labeled lymphocyte quantification algorithms depending on the number of training samples before and after being transferred to a new tumor indication. For these experiments, we adapted the RetinaNet architecture for the task of T-lymphocyte detection and employed transfer learning to bridge the domain gap between tumor indications and reduce the annotation costs for unseen domains. On our test set, we achieved human-level performance for almost all tumor indications with an average precision of 0.74 in-domain and 0.72-0.74 cross-domain. From our results, we derive recommendations for model development regarding annotation extent, training sample selection, and label extraction for the development of robust algorithms for immune cell scoring. By extending the task of marker-labeled lymphocyte quantification to a multi-class detection task, the pre-requisite for subsequent analyses, e.g., distinguishing lymphocytes in the tumor stroma from tumor-infiltrating lymphocytes, is met.

Pan-tumor CAnine cuTaneous Cancer Histology (CATCH) dataset.

Due to morphological similarities, the differentiation of histologic sections of cutaneous tumors into individual subtypes can be challenging. Recently, deep learning-based approaches have proven their potential for supporting pathologists in this regard. However, many of these supervised algorithms require a large amount of annotated data for robust development. We present a publicly available dataset of 350 whole slide images of seven different canine cutaneous tumors complemented by 12,424 polygon annotations for 13 histologic classes, including seven cutaneous tumor subtypes. In inter-rater experiments, we show a high consistency of the provided labels, especially for tumor annotations. We further validate the dataset by training a deep neural network for the task of tissue segmentation and tumor subtype classification. We achieve a class-averaged Jaccard coefficient of 0.7047, and 0.9044 for tumor in particular. For classification, we achieve a slide-level accuracy of 0.9857. Since canine cutaneous tumors possess various histologic homologies to human tumors the added value of this dataset is not limited to veterinary pathology but extends to more general fields of application.