Antibody Supervised Training of a Deep Learning Based Algorithm for Leukocyte Segmentation in Papillary Thyroid Carcinoma

External validation of a deep learning-based algorithm for detection of tall cells in papillary thyroid carcinoma: A multicenter study

The tall cell subtype (TC-PTC) is an aggressive subtype of papillary thyroid carcinoma (PTC). The TC-PTC is defined as a PTC comprising at least 30% epithelial cells that are three times as tall as they are wide. In practice, this definition is difficult to adhere to, resulting in high inter-observer variability. In this multicenter study, we validated a previously trained deep learning (DL)-based algorithm for detection of tall cells on 160 externally collected hematoxylin and eosin (HE)-stained PTC whole-slide images. In a test set of 360 manual annotations of regions of interest from 18 separate tissue sections in the external dataset, the DL-based algorithm detected TCs with a sensitivity of 90.6% and a specificity of 88.5%. The DL algorithm detected non-TC areas with a sensitivity of 81.6% and a specificity of 92.9%. In the validation datasets, 20% and 30% TC thresholds correlated with a significantly shorter relapse-free survival. In conclusion, the DL algorithm detected TCs in unseen, external scanned HE tissue slides with high sensitivity and specificity without any retraining.

Auxiliary Diagnosis of Papillary Thyroid Carcinoma Based on Spectral Phenotype

Thyroid cancer, a common endocrine malignancy, is one of the leading death causes among endocrine tumors. The diagnosis of pathological section analysis suffers from diagnostic delay and cumbersome operating procedures. Therefore, we intend to construct the models based on spectral data that can be potentially used for rapid intraoperative papillary thyroid carcinoma (PTC) diagnosis and characterize PTC characteristics. To alleviate any concerns pathologists may have about using the model, we conducted an analysis of the used bands that can be interpreted pathologically. A spectra acquisition system was first built to acquire spectra of pathological section images from 91 patients. The obtained spectral dataset contains 217 spectra of normal thyroid tissue and 217 spectra of PTC tissue. Clinical data of the corresponding patients were collected for subsequent model interpretability analysis. The experiment has been approved by the Ethics Review Committee of the Wuhu Hospital of East China Normal University. The spectral preprocessing method was used to process the spectra, and the preprocessed signal respectively optimized by the first and secondary informative wavelengths selection was used to develop the PTC detection models. The PTC detection model using mean centering (MC) and multiple scattering correction (MSC) has optimal performance, and the reasons for the good performance were analyzed in combination with the spectral acquisition process and composition of the test slide. For model interpretable analysis, the near-ultraviolet band selected for modeling corresponds to the location of amino acid absorption peak, and this is consistent with the clinical phenomenon of significantly lower amino acid concentrations in PTC patients. Moreover, the absorption peak of hemoglobin selected for modeling is consistent with the low hemoglobin index in PTC patients. In addition, the correlation analysis was performed between the selected wavelengths and the clinical data, and the results show: the reflection intensity of selected wavelengths in normal cells has a moderate correlation with cell arrangement structure, nucleus size and free thyroxine (FT4), and has a strong correlation with triiodothyronine (T3); the reflection intensity of selected bands in PTC cells has a moderate correlation with free triiodothyronine (FT3).

Computer-Assisted Annotation of Digital H&E/SOX10 Dual Stains Generates High-Performing Convolutional Neural Network for Calculating Tumor Burden in H&E-Stained Cutaneous Melanoma

Deep learning for the analysis of H&E stains requires a large annotated training set. This may form a labor-intensive task involving highly skilled pathologists. We aimed to optimize and evaluate computer-assisted annotation based on digital dual stains of the same tissue section. H&E stains of primary and metastatic melanoma (N = 77) were digitized, re-stained with SOX10, and re-scanned. Because images were aligned, annotations of SOX10 image analysis were directly transferred to H&E stains of the training set. Based on 1,221,367 annotated nuclei, a convolutional neural network for calculating tumor burden (CNN_TB) was developed. For primary melanomas, precision of annotation was 100% (95%CI, 99% to 100%) for tumor cells and 99% (95%CI, 98% to 100%) for normal cells. Due to low or missing tumor-cell SOX10 positivity, precision for normal cells was markedly reduced in lymph-node and organ metastases compared with primary melanomas (p < 0.001). Compared with stereological counts within skin lesions, mean difference in tumor burden was 6% (95%CI, -1% to 13%, p = 0.10) for CNN_TB and 16% (95%CI, 4% to 28%, p = 0.02) for pathologists. Conclusively, the technique produced a large annotated H&E training set with high quality within a reasonable timeframe for primary melanomas and subcutaneous metastases. For these lesion types, the training set generated a high-performing CNN_TB, which was superior to the routine assessments of pathologists.

A deep learning–based algorithm for tall cell detection in papillary thyroid carcinoma

Introduction

According to the World Health Organization, the tall cell variant (TCV) is an aggressive subtype of papillary thyroid carcinoma (PTC) comprising at least 30% epithelial cells two to three times as tall as they are wide. In practice, applying this definition is difficult causing substantial interobserver variability. We aimed to train a deep learning algorithm to detect and quantify the proportion of tall cells (TCs) in PTC.

Methods

We trained the deep learning algorithm using supervised learning, testing it on an independent dataset, and further validating it on an independent set of 90 PTC samples from patients treated at the Hospital District of Helsinki and Uusimaa between 2003 and 2013. We compared the algorithm-based TC percentage to the independent scoring by a human investigator and how those scorings associated with disease outcomes. Additionally, we assessed the TC score in 71 local and distant tumor relapse samples from patients with aggressive disease.

Results

In the test set, the deep learning algorithm detected TCs with a sensitivity of 93.7% and a specificity of 94.5%, whereas the sensitivity fell to 90.9% and specificity to 94.1% for non-TC areas. In the validation set, the deep learning algorithm TC scores correlated with a diminished relapse-free survival using cutoff points of 10% (p = 0.044), 20% (p < 0.01), and 30% (p = 0.036). The visually assessed TC score did not statistically significantly predict survival at any of the analyzed cutoff points. We observed no statistically significant difference in the TC score between primary tumors and relapse tumors determined by the deep learning algorithm or visually.

Conclusions

We present a novel deep learning–based algorithm to detect tall cells, showing that a high deep learning–based TC score represents a statistically significant predictor of less favorable relapse-free survival in PTC.

Automatic CD30 scoring method for whole slide images of primary cutaneous CD30+ lymphoproliferative diseases

AIMS

Deep-learning methods for scoring biomarkers are an active research topic. However, the superior performance of many studies relies on large datasets collected from clinical samples. In addition, there are fewer studies on immunohistochemical marker assessment for dermatological diseases. Accordingly, we developed a method for scoring CD30 based on convolutional neural networks for a few primary cutaneous CD30+ lymphoproliferative disorders and used this method to evaluate other biomarkers.

METHODS

A multipatch spatial attention mechanism and conditional random field algorithm were used to fully fuse tumour tissue characteristics on immunohistochemical slides and alleviate the few sample feature deficits. We trained and tested 28 CD30+ immunohistochemical whole slide images (WSIs), evaluated them with a performance index, and compared them with the diagnoses of senior dermatologists. Finally, the model's performance was further demonstrated on the publicly available Yale HER2 cohort.

RESULTS

Compared with the diagnoses by senior dermatologists, this method can better locate the tumour area and reduce the misdiagnosis rate. The prediction of CD3 and Ki-67 validated the model's ability to identify other biomarkers.

CONCLUSIONS

In this study, using a few immunohistochemical WSIs, our model can accurately identify CD30, CD3 and Ki-67 markers. In addition, the model could be applied to additional tumour identification tasks to aid pathologists in diagnosis and benefit clinical evaluation.

Outcome and Biomarker Supervised Deep Learning for Survival Prediction in Two Multicenter Breast Cancer Series

BACKGROUND

Prediction of clinical outcomes for individual cancer patients is an important step in the disease diagnosis and subsequently guides the treatment and patient counseling. In this work, we develop and evaluate a joint outcome and biomarker supervised (estrogen receptor expression and ERBB2 expression and gene amplification) multitask deep learning model for prediction of outcome in breast cancer patients in two nation-wide multicenter studies in Finland (the FinProg and FinHer studies). Our approach combines deep learning with expert knowledge to provide more accurate, robust, and integrated prediction of breast cancer outcomes.

MATERIALS AND METHODS

Using deep learning, we trained convolutional neural networks (CNNs) with digitized tissue microarray (TMA) samples of primary hematoxylin-eosin-stained breast cancer specimens from 693 patients in the FinProg series as input and breast cancer-specific survival as the endpoint. The trained algorithms were tested on 354 TMA patient samples in the same series. An independent set of whole-slide (WS) tumor samples from 674 patients in another multicenter study (FinHer) was used to validate and verify the generalization of the outcome prediction based on CNN models by Cox survival regression and concordance index (c-index). Visual cancer tissue characterization, i.e., number of mitoses, tubules, nuclear pleomorphism, tumor-infiltrating lymphocytes, and necrosis was performed on TMA samples in the FinProg test set by a pathologist and combined with deep learning-based outcome prediction in a multitask algorithm.

RESULTS

The multitask algorithm achieved a hazard ratio (HR) of 2.0 (95% confidence interval [CI] 1.30-3.00), P < 0.001, c-index of 0.59 on the 354 test set of FinProg patients, and an HR of 1.7 (95% CI 1.2-2.6), P = 0.003, c-index 0.57 on the WS tumor samples from 674 patients in the independent FinHer series. The multitask CNN remained a statistically independent predictor of survival in both test sets when adjusted for histological grade, tumor size, and axillary lymph node status in a multivariate Cox analyses. An improved accuracy (c-index 0.66) was achieved when deep learning was combined with the tissue characteristics assessed visually by a pathologist.

CONCLUSIONS

A multitask deep learning algorithm supervised by both patient outcome and biomarker status learned features in basic tissue morphology predictive of survival in a nationwide, multicenter series of patients with breast cancer. The algorithms generalized to another independent multicenter patient series and whole-slide breast cancer samples and provide prognostic information complementary to that of a comprehensive series of established prognostic factors.

DeepHistoClass: A Novel Strategy for Confident Classification of Immunohistochemistry Images Using Deep Learning

A multitude of efforts worldwide aim to create a single-cell reference map of the human body, for fundamental understanding of human health, molecular medicine, and targeted treatment. Antibody-based proteomics using immunohistochemistry (IHC) has proven to be an excellent technology for integration with large-scale single-cell transcriptomics datasets. The golden standard for evaluation of IHC staining patterns is manual annotation, which is expensive and may lead to subjective errors. Artificial intelligence holds much promise for efficient and accurate pattern recognition, but confidence in prediction needs to be addressed. Here, the aim was to present a reliable and comprehensive framework for automated annotation of IHC images. We developed a multilabel classification of 7848 complex IHC images of human testis corresponding to 2794 unique proteins, generated as part of the Human Protein Atlas (HPA) project. Manual annotation data for eight different cell types was generated as a basis for training and testing a proposed Hybrid Bayesian Neural Network. By combining the deep learning model with a novel uncertainty metric, DeepHistoClass (DHC) Confidence Score, the average diagnostic performance improved from 86.9% to 96.3%. This metric not only reveals which images are reliably classified by the model, but can also be utilized for identification of manual annotation errors. The proposed streamlined workflow can be developed further for other tissue types in health and disease and has important implications for digital pathology initiatives or large-scale protein mapping efforts such as the HPA project.

•

A novel method for automated annotation of immunohistochemistry images.

•

Introduction of an uncertainty metric, the DeepHistoClass (DHC) confidence score.

•

Increased accuracy of automated image predictions.

•

Identification of manual annotation errors.

Automated Quantification of sTIL Density with H&E-Based Digital Image Analysis Has Prognostic Potential in Triple-Negative Breast Cancers

Triple-negative breast cancer (TNBC) is an aggressive and difficult-to-treat cancer type that represents approximately 15% of all breast cancers. Recently, stromal tumor-infiltrating lymphocytes (sTIL) resurfaced as a strong prognostic biomarker for overall survival (OS) for TNBC patients. Manual assessment has innate limitations that hinder clinical adoption, and the International Immuno-Oncology Biomarker Working Group (TIL-WG) has therefore envisioned that computational assessment of sTIL could overcome these limitations and recommended that any algorithm should follow the manual guidelines where appropriate. However, no existing studies capture all the concepts of the guideline or have shown the same prognostic evidence as manual assessment. In this study, we present a fully automated digital image analysis pipeline and demonstrate that our hematoxylin and eosin (H&E)-based pipeline can provide a quantitative and interpretable score that correlates with the manual pathologist-derived sTIL status, and importantly, can stratify a retrospective cohort into two significant distinct prognostic groups. We found our score to be prognostic for OS (HR: 0.81 CI: 0.72-0.92 p = 0.001) independent of age, tumor size, nodal status, and tumor type in statistical modeling. While prior studies have followed fragments of the TIL-WG guideline, our approach is the first to follow all complex aspects, where appropriate, supporting the TIL-WG vision of computational assessment of sTIL in the future clinical setting.