An efficient CNN based algorithm for detecting melanoma cancer regions in H&E-stained images.

Histopathological images are widely used to diagnose diseases such as skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of abnormal cell nuclei and their distribution within multiple tissue sections would enable rapid comprehensive diagnostic assessment. In this paper, we propose a deep learning-based technique to segment the melanoma regions in Hematoxylin and Eosin-stained histopathological images. In this technique, the nuclei in an image are first segmented using a deep learning neural network. The segmented nuclei are then used to generate the melanoma region masks. Experimental results show that the proposed method can provide nuclei segmentation accuracy of around 90% and the melanoma region segmentation accuracy of around 98%. The proposed technique also has a low computational complexity.

Detection of malignant melanoma in H&E-stained images using deep learning techniques.

Histopathological images are widely used to diagnose diseases including skin cancer. As digital histopathological images are typically of very large size, in the order of several billion pixels, automated identification of all abnormal cell nuclei and their distribution within multiple tissue sections would assist rapid comprehensive diagnostic assessment. In this paper, we propose a deep learning-based technique to segment the melanoma regions in Hematoxylin and Eosin (H&E) stained histopathological images. In this technique, the nuclei in the image are first segmented using a Convolutional Neural Network (CNN). The segmented nuclei are then used to generate melanoma region masks. Experimental results with a small melanoma dataset show that the proposed method can potentially segment the nuclei with more than 94 % accuracy and segment the melanoma regions with a Dice coefficient of around 85 %. The proposed technique also has a small execution time making it suitable for clinical diagnosis with a fast turnaround time.

Automated proliferation index calculation for skin melanoma biopsy images using machine learning.

The Proliferation Index (PI) is an important diagnostic, predictive and prognostic parameter used for evaluating different types of cancer. This paper presents an automated technique to measure the PI values for skin melanoma images using machine learning algorithms. The proposed technique first analyzes a Mart-1 stained histology image and generates a region of interest (ROI) mask for the tumor. The ROI mask is then used to locate the tumor regions in the corresponding Ki-67 stained image. The nuclei in the Ki-67 ROI are then segmented and classified using a Convolutional Neural Network (CNN), and the PI value is calculated based on the number of the active and the passive nuclei. Experimental results show that the proposed technique can robustly segment (with 94 % accuracy) and classify the nuclei with a low computational complexity and the calculated PI values have less than 4 % average error.

Comparing docosahexaenoic acid (DHA) concomitant with neoadjuvant chemotherapy versus neoadjuvant chemotherapy alone in the treatment of breast cancer (DHA WIN): protocol of a double-blind, phase II, randomised controlled trial.

INTRODUCTION: Neoadjuvant chemotherapy for breast cancer treatment is prescribed to facilitate surgery and provide confirmation of drug-sensitive disease, and the achievement of pathological complete response (pCR) predicts improved long-term outcomes. Docosahexaenoic acid (DHA) has been shown to reduce tumour growth in preclinical models when combined with chemotherapy and is known to beneficially modulate systemic immune function. The purpose of this trial is to investigate the benefit of DHA supplementation in combination with neoadjuvant chemotherapy in patients with breast cancer. METHODS AND ANALYSIS: This is a double-blind, phase II, randomised controlled trial of 52 women prescribed neoadjuvant chemotherapy to test if DHA supplementation enhances chemotherapy efficacy. The DHA supplementation group will take 4.4 g/day DHA orally, and the placebo group will take an equal fat supplement of vegetable oil. The primary outcome will be change in Ki67 labelling index from prechemotherapy core needle biopsy to definitive surgical specimen. The secondary endpoints include assessment of (1) DHA plasma phospholipid content; (2) systemic immune cell types, plasma cytokines and inflammatory markers; (3) tumour markers for apoptosis and tumour infiltrating lymphocytes; (4) rate of pCR in breast and in axillary nodes; (5) frequency of grade 3 and 4 chemotherapy-associated toxicities; and (6) patient-perceived quality of life. The trial has 81% power to detect a significant between-group difference in Ki67 index with a two-sided t-test of less than 0.0497, and accounts for 10% dropout rate. ETHICS AND DISSEMINATION: This study has full approval from the Health Research Ethics Board of Alberta - Cancer Committee (Protocol #: HREBA.CC-18-0381). We expect to present the findings of this study to the scientific community in peer-reviewed journals and at conferences. The results of this study will provide evidence for supplementing with DHA during neoadjuvant chemotherapy treatment for breast cancer. TRIAL REGISTRATION NUMBER: NCT03831178.

Novel lymph node segmentation and proliferation index measurement for skin melanoma biopsy images.

The lymphatic system is the immune system of the human body, and includes networks of vessels spread over the body, lymph nodes, and lymph fluid. The lymph nodes are considered as purification units that collect the lymph fluid from the lymph vessels. Since the lymph nodes collect the cancer cells that escape from a malignant tumor and try to spread to the rest of the body, the lymph node analysis is important for staging many types skin and breast cancers. In this paper, we propose a Computer Aided Diagnosis (CAD) method that segments the lymph nodes and melanoma regions in a biopsy image and measure the proliferation index. The proposed method contains two stages. First, an automated technique is used to segment the lymph nodes in a biopsy image based on histogram and high frequency features. In the second stage, the proliferation index for the melanoma regions is calculated by comparing the number of active and passive nuclei. Experimental results on 76 different lymph node images show that the proposed segmentation technique can robustly segment the lymph nodes with more than 90% accuracy. The proposed proliferation index calculation has low complexity and has an average error rate of less than 1.5%.

Automated analysis and classification of melanocytic tumor on skin whole slide images.

This paper presents a computer-aided technique for automated analysis and classification of melanocytic tumor on skin whole slide biopsy images. The proposed technique consists of four main modules. First, skin epidermis and dermis regions are segmented by a multi-resolution framework. Next, epidermis analysis is performed, where a set of epidermis features reflecting nuclear morphologies and spatial distributions is computed. In parallel with epidermis analysis, dermis analysis is also performed, where dermal cell nuclei are segmented and a set of textural and cytological features are computed. Finally, the skin melanocytic image is classified into different categories such as melanoma, nevus or normal tissue by using a multi-class support vector machine (mSVM) with extracted epidermis and dermis features. Experimental results on 66 skin whole slide images indicate that the proposed technique achieves more than 95% classification accuracy, which suggests that the technique has the potential to be used for assisting pathologists on skin biopsy image analysis and classification.

Automatic measurement of melanoma depth of invasion in skin histopathological images.

Measurement of melanoma depth of invasion (DoI) in skin tissues is of great significance in grading the severity of skin disease and planning patient's treatment. However, accurate and automatic measurement of melanocytic tumor depth is a challenging problem mainly due to the difficulty of skin granular identification and melanoma detection. In this paper, we propose a technique for measuring melanoma DoI in microscopic images digitized from MART1 (i.e., meleanoma-associated antigen recognized by T cells) stained skin histopathological sections. The technique consists of four modules. First, skin melanoma areas are detected by combining color features with the Mahalanobis distance measure. Next, skin epidermis is segmented by a multi-thresholding method. The skin granular layer is then identified based on Bayesian classification of segmented skin epidermis pixels. Finally, the melanoma DoI is computed using a multi-resolution approach with Hausdorff distance measurement. Experimental results show that the proposed technique provides a superior performance in measuring the melanoma DoI than two closely related techniques.

Automatic Nuclear Segmentation Using Multiscale Radial Line Scanning With Dynamic Programming.

In the diagnosis of various cancers by analyzing histological images, automatic nuclear segmentation is an important step. However, nuclear segmentation is a difficult problem because of overlapping nuclei, inhomogeneous staining, and presence of noisy pixels and other tissue components. In this paper, we present an automatic technique for nuclear segmentation in skin histological images. The proposed technique first applies a bank of generalized Laplacian of Gaussian kernels to detect nuclear seeds. Based on the detected nuclear seeds, a multiscale radial line scanning method combined with dynamic programming is applied to extract a set of candidate nuclear boundaries. The gradient, intensity, and shape information are then integrated to determine the optimal boundary for each nucleus in the image. Nuclear overlap limitation is finally imposed based on a Dice coefficient measure such that the obtained nuclear contours do not severely intersect with each other. Experiments have been thoroughly performed on two datasets with H&E and Ki-67 stained images, which show that the proposed technique is superior to conventional schemes of nuclear segmentation.

Automatic Nuclei Detection Based on Generalized Laplacian of Gaussian Filters.

Efficient and accurate detection of cell nuclei is an important step toward automatic analysis in histopathology. In this work, we present an automatic technique based on generalized Laplacian of Gaussian (gLoG) filter for nuclei detection in digitized histological images. The proposed technique first generates a bank of gLoG kernels with different scales and orientations and then performs convolution between directional gLoG kernels and the candidate image to obtain a set of response maps. The local maxima of response maps are detected and clustered into different groups by mean-shift algorithm based on their geometrical closeness. The point which has the maximum response in each group is finally selected as the nucleus seed. Experimental results on two datasets show that the proposed technique provides a superior performance in nuclei detection compared to existing techniques.

New Robust and Reproducible Stereological IHC Ki67 Breast Cancer Proliferative Assessment to Replace Traditional Biased Labeling Index.

There is a pressing need for an objective decision tool to guide therapy for breast cancer patients that are estrogen receptor positive and HER2/neu negative. This subset of patients contains a mixture of luminal A and B tumors with good and bad outcomes, respectively. The 2 main current tools are on the basis of immunohistochemistry (IHC) or gene expression, both of which rely on the expression of distinct molecular groups that reflect hormone receptors, HER2/neu status, and most importantly, proliferation. Despite the success of a proprietary molecular test, definitive superiority of any method has not yet been demonstrated. Ki67 IHC scoring assessments have been shown to be poorly reproducible, whereas molecular testing is costly with a longer turnaround time. This work proposes an objective Ki67 index using image analysis that addresses the existing methodological issues of Ki67 quantitation using IHC on paraffin-embedded tissue. Intrinsic bias related to numerical assessment performed on IHC is discussed as well as the sampling issue related to the "peel effect" of tiny objects within a thin section. A new nonbiased stereological parameter (VV) based on the Cavalieri method is suggested for use on a double-stained Ki67/cytokeratin IHC slide. The assessment is performed with open-source ImageJ software with interobserver concordance between 3 pathologists being high at 93.5%. Furthermore, VV was found to be a superior method to predict an outcome in a small subset of breast cancer patients when compared with other image analysis methods being used to determine the Ki67 labeling index. Calibration methodology is also discussed to further this IHC approach.

Next generation sequencing profiling identifies miR-574-3p and miR-660-5p as potential novel prognostic markers for breast cancer.

BACKGROUND: Prognostication of Breast Cancer (BC) relies largely on traditional clinical factors and biomarkers such as hormone or growth factor receptors. Due to their suboptimal specificities, it is challenging to accurately identify the subset of patients who are likely to undergo recurrence and there remains a major need for markers of higher utility to guide therapeutic decisions. MicroRNAs (miRNAs) are small non-coding RNAs that function as post-transcriptional regulators of gene expression and have shown promise as potential prognostic markers in several cancer types including BC. RESULTS: In our study, we sequenced miRNAs from 104 BC samples and 11 apparently healthy normal (reduction mammoplasty) breast tissues. We used Case-control (CC) and Case-only (CO) statistical paradigm to identify prognostic markers. Cox-proportional hazards regression model was employed and risk score analysis was performed to identify miRNA signature independent of potential confounders. Representative miRNAs were validated using qRT-PCR. Gene targets for prognostic miRNAs were identified using in silico predictions and in-house BC transcriptome dataset. Gene ontology terms were identified using DAVID bioinformatics v6.7. A total of 1,423 miRNAs were captured. In the CC approach, 126 miRNAs were retained with predetermined criteria for good read counts, from which 80 miRNAs were differentially expressed. Of these, four and two miRNAs were significant for Overall Survival (OS) and Recurrence Free Survival (RFS), respectively. In the CO approach, from 147 miRNAs retained after filtering, 11 and 4 miRNAs were significant for OS and RFS, respectively. In both the approaches, the risk scores were significant after adjusting for potential confounders. The miRNAs associated with OS identified in our cohort were validated using an external dataset from The Cancer Genome Atlas (TCGA) project. Targets for the identified miRNAs were enriched for cell proliferation, invasion and migration. CONCLUSIONS: The study identified twelve non-redundant miRNAs associated with OS and/or RFS. These signatures include those that were reported by others in BC or other cancers. Importantly we report for the first time two new candidate miRNAs (miR-574-3p and miR-660-5p) as promising prognostic markers. Independent validation of signatures (for OS) using an external dataset from TCGA further strengthened the study findings.

Canadian Association of Pathologists-Association canadienne des pathologistes National Standards Committee for High Complexity Testing/Immunohistochemistry: guidelines for the preparation, release, and storage of unstained archived diagnostic tissue sections for immunohistochemistry.

OBJECTIVES: Formalin-fixed, paraffin-embedded unstained archived diagnostic tissue sections are frequently exchanged between clinical laboratories for immunohistochemical staining. The manner in which such sections are prepared represents a type of preanalytical variable that must be taken into account given the growing importance of immunohistochemical assays, especially predictive and prognostic tests, in personalized medicine. METHODS: Recommendations were derived from review of the literature and expert consensus of the Canadian Association of Pathologists-Association canadienne des pathologists National Standards Committee for High Complexity Testing/Immunohistochemistry. RESULTS: Relevant considerations include the type of glass slide on which to mount the unstained sections; the thickness of the tissue sections; the time from slide preparation to testing; the environment, particularly the temperature at which the unstained sections will be maintained prior to testing; the inclusion of on-slide positive control tissue where possible; and whether patient identifier(s) should be included on slide labels. CONCLUSIONS: Clear communication between requesting and releasing laboratories will facilitate the proper preparation of unstained sections and also ensure that applicable privacy considerations are addressed.

Outcome analysis of breast cancer patients who declined evidence-based treatment.

BACKGROUND: To analyze the characteristics and outcomes of women with breast cancer in the Northern Alberta Health Region (NAHR) who declined recommended primary standard treatments. METHODS: A chart review was performed of breast cancer patients who refused recommended treatments during the period 1980 to 2006. A matched pair analysis was performed to compare the survival data between those who refused or received standard treatments. RESULTS: A total of 185 (1.2%) patients refused standard treatment. Eighty-seven (47%) were below the age of 75 at diagnosis. The majority of those who refused standard treatments were married (50.6%), 50 years or older (60.9%), and from the urban area (65.5%). The 5-year overall survival rates were 43.2% (95% CI: 32.0 to 54.4%) for those who refused standard treatments and 81.9% (95% CI: 76.9 to 86.9%) for those who received them. The corresponding values for the disease-specific survival were 46.2% (95% CI: 34.9 to 57.6%) vs. 84.7% (95% CI: 80.0 to 89.4%). CONCLUSIONS: Women who declined primary standard treatment had significantly worse survival than those who received standard treatments. There is no evidence to support using Complementary and Alternative Medicine (CAM) as primary cancer treatment.

Inappropriate calibration and optimisation of pan-keratin (pan-CK) and low molecular weight keratin (LMWCK) immunohistochemistry tests: Canadian Immunohistochemistry Quality Control (CIQC) experience.

AIMS: Pan-cytokeratin (pan-CK) and low molecular weight cytokeratin (LMWCK) tests are the most common immunohistochemistry (IHC) tests used to support evidence of epithelial differentiation. Canadian Immunohistochemistry Quality Control (CIQC), a new provider of proficiency testing for Canadian clinical IHC laboratories, has evaluated the performance of Canadian IHC laboratories in two proficiency testing challenges for both pan-CK and LMWCK. METHODS: CIQC has designed a 70-sample tissue microarray (TMA) for challenge 1 and a 30-sample TMA for challenge 2. There were 13 participants in challenge 1, and 62 in challenge 2. All results were evaluated and scored by CIQC assessors and compared with reference laboratory results. RESULTS: Participating laboratories often produced false-negative results that ranged from 20% to 80%. False-positive results were also detected. About half of participating clinical laboratories have inappropriately calibrated IHC tests for pan-CK and LMWCK, which are the most commonly used markers for demonstration of epithelial differentiation. The great majority of laboratories were not aware of the problem with calibration of pan-CK and LMWCK tests because of inappropriate selection of external positive controls and samples for optimisation of these tests. Benign liver and kidney are the most important tissues to include as positive controls for both pan-CK and LMWCK. CONCLUSIONS: Participation in external quality assurance is important for peer comparison and proper calibration of IHC tests, which is also helpful for appropriate selection of positive control material and material for optimisation of the tests.

Canadian Association of Pathologists-Association canadienne des pathologistes National Standards Committee/Immunohistochemistry: best practice recommendations for standardization of immunohistochemistry tests.

Immunohistochemical and immunocytochemical assays are highly complex diagnostic analyses used to aid in the accurate identification and biologic characterization of tissue types in neoplastic and nonneoplastic diseases. Immunohistochemical tests are applied mainly to the diagnosis of neoplasms. Some immunohistochemical tests provide information of important prognostic and predictive value in selected human neoplasms and, as such, are often critical for the appropriate and effective treatment of patients. This document provides recommendations and opinions of the Canadian Association of Pathologists-Association canadienne des pathologistes National Standards Committee/Immunohistochemistry relevant to clinical immunohistochemical terminology, classification of immunohistochemical tests based on risk assessment, and quality control and quality assurance and summarizes matters to be considered for appropriate immunohistochemical/immunocytochemical test development, performance, and interpretation in diagnostic pathology and laboratory medicine.

Implementation of a Canadian external quality assurance program for breast cancer biomarkers: an initiative of Canadian Quality Control in immunohistochemistry (cIQc) and Canadian Association of Pathologists (CAP) National Standards Committee/Immunohistochemistry.

Immunohistochemistry results for estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 are used to guide breast carcinoma patient management and it is essential to monitor these tests in external quality assurance (EQA) programs. Canadian Immunohistochemistry Quality Control is a web-based program with novel approach to EQA. Canadian Immunohistochemistry Quality Control RUN2 included tissue microarray slides with 38 samples tested by 18 immunohistochemical laboratories. Deidentified results were posted for viewing at www.ciqc.ca including all used protocols matched with scanned slides for virtual microscopy and garrattograms. Sensitivity, specificity, Kendall W test (concordance between laboratories), and kappa statistics (agreement with designated reference values) were calculated. Kappa values were within the target range (>0.8, or "near perfect" agreement) for 85% results. Kendall coefficient was 0.942 for estrogen receptor, 0.930 for progesterone receptor, and 0.958 for human epidermal growth factor receptor 2. The anonymous participation, quick feedback, and unrestricted full access in EQA results provides rapid insight into technical or interpretive deficiencies, allowing appropriate corrective action to be taken whereas the use of tissue microarrays enables meaningful statistical analysis.

A young woman with distant recurrent metastatic primary vaginal carcinoma salvaged with radical radiotherapy.

The incidence of distant recurrent metastatic primary vaginal carcinoma is rare. The prognosis in such cases is poor, with cure being extremely rare. We report the case of a young woman, with distant recurrent metastatic primary vaginal carcinoma in which the patient remains disease-free 5 years after completing salvage radical radiotherapy. The clinical management of recurrent metastatic primary vaginal carcinoma must be tailored to the site of recurrence and the patient's performance status. Complete clinical remission and long-term survival without evidence of disease may be achieved in rare cases with radical radiotherapy.

Cytotoxic activities of nucleoside and nucleobase analog drugs in malignant mesothelioma: characterization of a novel nucleobase transport activity.

This study was designed to evaluate the cytotoxic activity of several nucleoside and nucleobase analog drugs as possible new agents for treatment of malignant mesothelioma and to identify factors responsible for the clinical variation of nucleoside analog drug response in chemotherapy of mesothelioma. Three human mesothelioma cell lines (MSTO-211H, H2452 and H2052) were tested for gemcitabine sensitivity and nucleoside transport activity. MSTO-211H, H2452 and H2052 exhibited differences in sensitivity to gemcitabine, nucleoside transport rates and hENT1 site densities. In H2052 cells, gemcitabine, 5-fluoro-2'-deoxyuridine, clofarabine and cladribine were most active with IC(50) values of 46, 43, 240 and 490 nM, respectively, whereas 5-fluorouracil was the least cytotoxic drug tested. In H2052 cells, the combination of gemcitabine and fludarabine or cladribine resulted in synergistic cytotoxic response. In nucleobase transport studies, hypoxanthine and 6-mercaptopurine but not 5-fluorouracil was transported into H2052 cells by a novel purine-specific, sodium-independent nucleobase transport activity. In summary differences in nucleoside analog drug transport activities are likely to contribute to the observed clinical variation in nucleoside analog response in patients and for the first time a correlation between nucleobase drug sensitivities and transport activities was shown. A novel combination of gemcitabine and fludarabine or cladribine had synergistic cytotoxic activity against the least sensitive mesothelioma cell line. These drug combinations merit further evaluation as effective therapeutic regimens in patients with aggressive mesothelioma.