Learning more from the inter-rater reliability of interstitial fibrosis assessment beyond just a statistic.

Interstitial fibrosis assessment by renal pathologists lacks good agreement, and we aimed to investigate its hidden properties and infer possible clinical impact. Fifty kidney biopsies were assessed by 9 renal pathologists and evaluated by intraclass correlation coefficients (ICCs) and kappa statistics. Probabilities of pathologists' assessments that would deviate far from true values were derived from quadratic regression and multilayer perceptron nonlinear regression. Likely causes of variation in interstitial fibrosis assessment were investigated. Possible misclassification rates were inferred on reported large cohorts. We found inter-rater reliabilities ranged from poor to good (ICCs 0.48 to 0.90), and pathologists' assessments had the worst agreements when the extent of interstitial fibrosis was moderate. 33.5% of pathologists' assessments were expected to deviate far from the true values. Variation in interstitial fibrosis assessment was found to be correlated with variation in interstitial inflammation assessment (r2 = 32.1%). Taking IgA nephropathy as an example, the Oxford T scores for interstitial fibrosis were expected to be misclassified in 21.9% of patients. This study demonstrated the complexity of the inter-rater reliability of interstitial fibrosis assessment, and our proposed approaches discovered previously unknown properties in pathologists' practice and inferred a possible clinical impact on patients.

Unsatisfactory reproducibility of interstitial inflammation scoring in allograft kidney biopsy.

Interstitial inflammation scoring is incorporated into the Banff Classification of Renal Allograft Pathology and is essential for the diagnosis of T-cell mediated rejection. However, its reproducibility, including inter-rater and intra-rater reliabilities, has not been carefully investigated. In this study, eight renal pathologists from different hospitals independently scored 45 kidney allograft biopsies with varying extents of interstitial inflammation. Inter-rater reliabilities and intra-rater reliabilities were investigated by kappa statistics and conditional agreement probabilities. Individual pathologists' scoring patterns were examined by chi-squared tests and proportions tests. The mean pairwise kappa values for inter-rater reliability were 0.27, 0.30, and 0.26 for the Banff i score, ti score, and i-IFTA, respectively. No rater pair performed consistently better or worse than others on all three scorings. After dichotomizing the scores into two groups (none/mild and moderate/severe inflammation), the averaged conditional agreements ranged from 47.1% to 50.0%. The distributions of the scores differed, but some pathologists persistently scored higher or lower than others. Given the important role of interstitial inflammation scoring in the diagnosis of T-cell mediated rejection, transplant practitioners should be aware of the possible clinical implications of the far-from-optimal reproducibility.

Deep Convolutional Neural Networks Detect Tumor Genotype from Pathological Tissue Images in Gastrointestinal Stromal Tumors.

Gastrointestinal stromal tumors (GIST) are common mesenchymal tumors, and their effective treatment depends upon the mutational subtype of the KIT/PDGFRA genes. We established deep convolutional neural network (DCNN) models to rapidly predict drug-sensitive mutation subtypes from images of pathological tissue. A total of 5153 pathological images of 365 different GISTs from three different laboratories were collected and divided into training and validation sets. A transfer learning mechanism based on DCNN was used with four different network architectures, to identify cases with drug-sensitive mutations. The accuracy ranged from 87% to 75%. Cross-institutional inconsistency, however, was observed. Using gray-scale images resulted in a 7% drop in accuracy (accuracy 80%, sensitivity 87%, specificity 73%). Using images containing only nuclei (accuracy 81%, sensitivity 87%, specificity 73%) or cytoplasm (accuracy 79%, sensitivity 88%, specificity 67%) produced 6% and 8% drops in accuracy rate, respectively, suggesting buffering effects across subcellular components in DCNN interpretation. The proposed DCNN model successfully inferred cases with drug-sensitive mutations with high accuracy. The contribution of image color and subcellular components was also revealed. These results will help to generate a cheaper and quicker screening method for tumor gene testing.

Improving Algorithm for the Alignment of Consecutive, Whole-Slide, Immunohistochemical Section Images.

BACKGROUND: Accurate and precise alignment of histopathology tissue sections is a key step for the interpretation of the proteome topology and cell level three-dimensional (3D) reconstruction of diseased tissues. However, the realization of an automated and robust method for aligning nonglobally stained immunohistochemical (IHC) sections is still challenging. In this study, we aim to assess the feasibility of multidimensional graph-based image registration on aligning serial-section and whole-slide IHC section images. MATERIALS AND METHODS: An automated, patch graph-based registration method was established and applied to align serial, whole-slide IHC sections at ×10 magnification (average 32,947 × 27,054 pixels). The alignment began with the initial alignment of high-resolution reference and translated images (object segmentation and rigid registration) and nonlinear registration of low-resolution reference and translated images, followed by the multidimensional graph-based image registration of the segmented patches, and finally, the fusion of deformed patches for inspection. The performance of the proposed method was formulated and evaluated by the Hausdorff distance between continuous image slices. RESULTS: Sets of average 315 patches from five serial whole slide, IHC section images were tested using 21 different IHC antibodies across five different tissue types (skin, breast, stomach, prostate, and soft tissue). The proposed method was successfully automated to align most of the images. The average Hausdorff distance was 48.93 µm with a standard deviation of 14.94 µm, showing a significant improvement from the previously published patch-based nonlinear image registration method (average Hausdorff distance of 93.89 µm with 50.85 µm standard deviation). CONCLUSIONS: Our method was effective in aligning whole-slide tissue sections at the cell-level resolution. Further advancements in the screening of the proteome topology and 3D tissue reconstruction could be expected.

A computer-aided diagnosis system for differentiation and delineation of malignant regions on whole-slide prostate histopathology image using spatial statistics and multidimensional DenseNet.

PURPOSE: Prostate cancer (PCa) is a major health concern in aging males, and proper management of the disease depends on accurately interpreting pathology specimens. However, reading prostatectomy histopathology slides, which is basically for staging, is usually time consuming and differs from reading small biopsy specimens, which is mainly used for diagnosis. Generally, each prostatectomy specimen generates tens of large tissue sections and for each section, the malignant region needs to be delineated to assess the amount of tumor and its burden. With the aim of reducing the workload of pathologists, in this study, we focus on developing a computer-aided diagnosis (CAD) system based on a densely connected convolutional neural network (DenseNet) for whole-slide histopathology images to outline the malignant regions. METHODS: We use an efficient color normalization process based on ranklet transformation to automatically correct the intensity of the images. Additionally, we use spatial probability to segment the tissue structure regions for different tissue recognition patterns. Based on the segmentation, we incorporate a multidimensional structure into DenseNet to determine if a particular prostatic region is benign or malignant. RESULTS: As demonstrated by the experimental results with a test set of 2,663 images from 32 whole-slide prostate histopathology images, our proposed system achieved 0.726, 0.6306, and 0.5209 in the average of the Dice coefficient, Jaccard similarity coefficient, and Boundary F1 score measures, respectively. Then, the accuracy, sensitivity, specificity, and the area under the ROC curve (AUC) of the proposed classification method were observed to be 95.0% (2544/2663), 96.7% (1210/1251), 93.9% (1334/1412), and 0.9831, respectively. DISCUSSIONS: We provide a detailed discussion on how our proposed system demonstrates considerable improvement compared with similar methods considered in previous researches as well as how it can be used for delineating malignant regions.