Tumour infiltrating lymphocytes and survival after adjuvant chemotherapy in patients with gastric cancer: post-hoc analysis of the CLASSIC trial.

BACKGROUND: Only a subset of gastric cancer (GC) patients with stage II-III benefits from chemotherapy after surgery. Tumour infiltrating lymphocytes per area (TIL density) has been suggested as a potential predictive biomarker of chemotherapy benefit. METHODS: We quantified TIL density in digital images of haematoxylin-eosin (HE) stained tissue using deep learning in 307 GC patients of the Yonsei Cancer Center (YCC) (193 surgery+adjuvant chemotherapy [S + C], 114 surgery alone [S]) and 629 CLASSIC trial GC patients (325 S + C and 304 S). The relationship between TIL density, disease-free survival (DFS) and clinicopathological variables was analysed. RESULTS: YCC S patients and CLASSIC S patients with high TIL density had longer DFS than S patients with low TIL density (P = 0.007 and P = 0.013, respectively). Furthermore, CLASSIC patients with low TIL density had longer DFS if treated with S + C compared to S (P = 0.003). No significant relationship of TIL density with other clinicopathological variables was found. CONCLUSION: This is the first study to suggest TIL density automatically quantified in routine HE stained tissue sections as a novel, clinically useful biomarker to identify stage II-III GC patients deriving benefit from adjuvant chemotherapy. Validation of our results in a prospective study is warranted.

The clinical importance of the host anti-tumour reaction patterns in regional tumour draining lymph nodes in patients with locally advanced resectable gastric cancer: a systematic review and meta-analysis.

BACKGROUND: The status of regional tumour draining lymph nodes (LN) is crucial for prognostic evaluation in gastric cancer (GaC) patients. Changes in lymph node microarchitecture, such as follicular hyperplasia (FH), sinus histiocytosis (SH), or paracortical hyperplasia (PH), may be triggered by the anti-tumour immune response. However, the prognostic value of these changes in GaC patients is unclear. METHODS: A systematic search in multiple databases was conducted to identify studies on the prognostic value of microarchitecture changes in regional tumour-negative and tumour-positive LNs measured on histopathological slides. Since the number of GaC publications was very limited, the search was subsequently expanded to include junctional and oesophageal cancer (OeC). RESULTS: A total of 28 articles (17 gastric cancer, 11 oesophageal cancer) met the inclusion criteria, analyzing 26,503 lymph nodes from 3711 GaC and 1912 OeC patients. The studies described eight different types of lymph node microarchitecture changes, categorized into three patterns: hyperplasia (SH, FH, PH), cell-specific infiltration (dendritic cells, T cells, neutrophils, macrophages), and differential gene expression. Meta-analysis of five GaC studies showed a positive association between SH in tumour-negative lymph nodes and better 5-year overall survival. Pooled risk ratios for all LNs showed increased 5-year overall survival for the presence of SH and PH. CONCLUSIONS: This systematic review suggests that sinus histiocytosis and paracortical hyperplasia in regional tumour-negative lymph nodes may provide additional prognostic information for gastric and oesophageal cancer patients. Further studies are needed to better understand the lymph node reaction patterns and explore their impact of chemotherapy treatment and immunotherapy efficacy.

Histological 3D reconstruction and in vivo lineage tracing of the human endometrium.

Regular menstrual shedding and repair of the endometrial functionalis is unique to humans and higher-order primates. The current consensus postulates endometrial glands to have a single-tubular architecture, where multi-potential stem cells reside in the blind-ending glandular-bases. Utilising fixed samples from patients, we have studied the three-dimensional (3D) micro-architecture of the human endometrium. We demonstrate that some non-branching, single, vertical functionalis glands originate from a complex horizontally interconnecting network of basalis glands. The existence of a multipotent endometrial epithelial stem cell capable of regenerating the entire complement of glandular lineages was demonstrated by in vivo lineage tracing, using naturally occurring somatic mitochondrial DNA mutations as clonal markers. Vertical tracking of mutated clones showed that at least one stem-cell population resides in the basalis glands. These novel findings provide insight into the efficient and scar-less regenerative potential of the human endometrium. © 2020 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

The role of the longitudinal muscle in the anal sphincter complex: Implications for the Intersphincteric Plane in Low Rectal Cancer Surgery?

Intersphincteric resection (ISR) enables radical sphincter-preserving surgery in a subset of low rectal tumors impinging on the anal sphincter complex (ASC). Excellent anatomical knowledge is essential for optimal ISR. This study describes the role of the longitudinal muscle (LM) in the ASC and implications for ISR and other low rectal and anal pathologies. Six human adult en bloc cadaveric specimens (three males, three females) were obtained from the University of Leeds GIFT Research Tissue Programme. Paraffin-embedded mega blocks containing the ASC were produced and serially sectioned at 250 µm intervals. Whole mount microscopic sections were histologically stained and digitally scanned. The intersphincteric plane was shown to be potentially very variable. In some places adipose tissue is located between the external anal sphincter (EAS) and internal anal sphincter (IAS), whereas in others the LM interdigitates to obliterate the plane. Elsewhere the LM is (partly) absent with the intersphincteric plane lying on the IAS. The LM gave rise to the formation of the submucosae and corrugator ani muscles by penetrating the IAS and EAS. In four of six specimens, striated muscle fibers from the EAS curled around the distal IAS reaching the anal submucosa. The ASC formed a complex structure, varying between individuals with an inconstant LM affecting the potential location of the intersphincteric plane as well as a high degree of intermingling striated and smooth muscle fibers potentially further disrupting the plane. The complexity of identifying the correct pathological staging of low rectal cancer is also demonstrated. Clin. Anat. 33:567-577, 2020. © 2019 Wiley Periodicals, Inc.

Three-dimensional reconstruction with serial whole-mount sections of oral tongue squamous cell carcinoma: A preliminary study.

OBJECTIVES: Margin status and invasion pattern are prognostic factors for oral tongue squamous cell carcinoma (OTSCC). Current methods to identify these factors are limited to 2D observation; it is necessary to explore 3D reconstruction with whole-mount sample to improve the accuracy of analysis. This study aimed to study the tissue preparation, section generation, and 3D reconstruction with whole-mount OTSCC specimen. STUDY DESIGN: Two OTSCC samples were retrieved from Nanjing Stomatological Hospital, Medical School of Nanjing University. One sample was sliced into 3 equal-sized pieces and subjected to different processing schedules to determine the best method. The second sample was processed accordingly. Serial whole-mount sections of the second sample were generated, stained with HE/anticytokine antibody in intersection manner, and scanned into digital images. Digital images were aligned and reconstructed into 3D images with Hetero Genius Medical Image Manager 3D Pathology Add-On [HGMIM3D]. RESULTS: Successful serial whole-mount sections of comparable quality to traditional sections were generated. Three-dimensional images with serial whole-mount sections were successfully generated. CONCLUSIONS: Whole-mount histopathological 3D reconstruction of OTSCC was successfully generated, providing a solid foundation for comprehensive margin and invasion analysis. Although future study and improvement were needed, whole-mount histopathological 3D reconstruction proved to be a promising method in OTSCC study.

Quantitative fat and R2* mapping in vivo to measure lipid-rich necrotic core and intraplaque hemorrhage in carotid atherosclerosis.

PURPOSE: The aim of this work was to quantify the extent of lipid-rich necrotic core (LRNC) and intraplaque hemorrhage (IPH) in atherosclerotic plaques. METHODS: Patients scheduled for carotid endarterectomy underwent four-point Dixon and T1-weighted magnetic resonance imaging (MRI) at 3 Tesla. Fat and R2* maps were generated from the Dixon sequence at the acquired spatial resolution of 0.60 × 0.60 × 0.70 mm voxel size. MRI and three-dimensional (3D) histology volumes of plaques were registered. The registration matrix was applied to segmentations denoting LRNC and IPH in 3D histology to split plaque volumes in regions with and without LRNC and IPH. RESULTS: Five patients were included. Regarding volumes of LRNC identified by 3D histology, the average fat fraction by MRI was significantly higher inside LRNC than outside: 12.64 ± 0.2737% versus 9.294 ± 0.1762% (mean ± standard error of the mean [SEM]; P < 0.001). The same was true for IPH identified by 3D histology, R2* inside versus outside IPH was: 71.81 ± 1.276 s-1 versus 56.94 ± 0.9095 s-1 (mean ± SEM; P < 0.001). There was a strong correlation between the cumulative fat and the volume of LRNC from 3D histology (R2 = 0.92) as well as between cumulative R2* and IPH (R2 = 0.94). CONCLUSION: Quantitative mapping of fat and R2* from Dixon MRI reliably quantifies the extent of LRNC and IPH. Magn Reson Med 78:285-296, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

3D Buried Utility Location Using A Marching-Cross-Section Algorithm for Multi-Sensor Data Fusion.

We address the problem of accurately locating buried utility segments by fusing data from multiple sensors using a novel Marching-Cross-Section (MCS) algorithm. Five types of sensors are used in this work: Ground Penetrating Radar (GPR), Passive Magnetic Fields (PMF), Magnetic Gradiometer (MG), Low Frequency Electromagnetic Fields (LFEM) and Vibro-Acoustics (VA). As part of the MCS algorithm, a novel formulation of the extended Kalman Filter (EKF) is proposed for marching existing utility tracks from a scan cross-section (scs) to the next one; novel rules for initializing utilities based on hypothesized detections on the first scs and for associating predicted utility tracks with hypothesized detections in the following scss are introduced. Algorithms are proposed for generating virtual scan lines based on given hypothesized detections when different sensors do not share common scan lines, or when only the coordinates of the hypothesized detections are provided without any information of the actual survey scan lines. The performance of the proposed system is evaluated with both synthetic data and real data. The experimental results in this work demonstrate that the proposed MCS algorithm can locate multiple buried utility segments simultaneously, including both straight and curved utilities, and can separate intersecting segments. By using the probabilities of a hypothesized detection being a pipe or a cable together with its 3D coordinates, the MCS algorithm is able to discriminate a pipe and a cable close to each other. The MCS algorithm can be used for both post- and on-site processing. When it is used on site, the detected tracks on the current scs can help to determine the location and direction of the next scan line. The proposed "multi-utility multi-sensor" system has no limit to the number of buried utilities or the number of sensors, and the more sensor data used, the more buried utility segments can be detected with more accurate location and orientation.

Comparison of the Diagnostic Performance of Four Quantitative Myocardial Perfusion Estimation Methods Used in Cardiac MR Imaging: CE-MARC Substudy.

PURPOSE: To compare the diagnostic performance of four tracer kinetic analysis methods to quantify myocardial perfusion from magnetic resonance (MR) imaging cardiac perfusion data sets in terms of their ability to lead to the diagnosis of myocardial ischemia. MATERIALS AND METHODS: The study was approved by the regional ethics committee, and all patients gave written consent. A representative sample of 50 patients with suspected ischemic heart disease was retrospectively selected from the Clinical Evaluation of Magnetic Resonance Imaging in Coronary Heart Disease trial data set. Quantitative myocardial blood flow (MBF) was estimated from rest and adenosine stress MR imaging perfusion data sets by using four established methods. A matching diagnosis of both an inducible defect as assessed with single photon emission computed tomography and a luminal stenosis of 70% or more as assessed with quantitative x-ray angiography was used as the reference standard for the presence of myocardial ischemia. Diagnostic performance was evaluated with receiver operating characteristic (ROC) curve analysis for each method, with stress MBF and myocardial perfusion reserve (MPR) serving as continuous measures. RESULTS: Area under the ROC curve with stress MBF and MPR as the outcome measures, respectively, was 0.86 and 0.92 for the Fermi model, 0.85 and 0.87 for the uptake model, 0.85 and 0.80 for the one-compartment model, and 0.87 and 0.87 for model-independent deconvolution. There was no significant difference between any of the models or between MBF and MPR, except that the Fermi model outperformed the one-compartment model if MPR was used as the outcome measure (P = .02). CONCLUSION: Diagnostic performance of quantitative myocardial perfusion estimates is not affected by the tracer kinetic analysis method used.

Three-dimensional reconstruction of ductal carcinoma in situ with virtual slides.

AIMS: This study aimed to assess the feasibility of using virtual slides to create 3D histopathological reconstructions to aid in the study of the biology of DCIS. METHODS: Four µm thick serial sections of formalin fixed paraffin embedded tissue from three cases were cut and mounted onto glass slides, stained with haematoxylin and eosin, then scanned. The three image stacks comprised 30, 115 and 100 scanned sections creating a similar number of virtual slides. The virtual slides were registered using custom 3D software to create 3D tissue volumes. The volumes were annotated to highlight distinct features and 3D visualisations (segmentations) were created to study these features in 3D. RESULTS: The most time-intensive step was the manual annotation of virtual slides 3D histopathological reconstructions were created of (i) DCIS surrounded by adjacent invasion; (ii) pure DCIS and (iii) a 'normal' lobule. CONCLUSION: 3D in silico reconstructions of DCIS were created and more extensive studies can now be done within a realistic timescale. We have identified structural similarities between a benign lobule and DCIS which support the view that much DCIS, apparently in a 'duct' is contained within and expanded lobule. This method has the potential to provide insights into the biology of DCIS.

New insights into the lymphovascular microanatomy of the colon and the risk of metastases in pT1 colorectal cancer obtained with quantitative methods and three-dimensional digital reconstruction.

AIMS: UK faecal occult blood test screening has tripled the proportion of pT1 colorectal cancers. The risk of metastasis is predicted by depth of invasion, suggesting that access to deep lymphovascular vessels is important. The aim of this study was to quantify the distribution and size of the submucosal vasculature, and generate a novel three-dimensional (3D) model to validate the findings. METHODS AND RESULTS: Thirty samples of normal large bowel wall were immunostained with CD31, a vascular endothelium marker, to identify blood vessels, which were quantified and digitally analysed for their number, circumference, area and diameter in the deep mucosa and submucosa (Sm1, Sm2, and Sm3). The model required serial sections, a double immunostain (using CD31 and D2-40), and 3D reconstruction. Significant differences were shown between submucosal layers in the number, circumference and area of vessels (P < 0.001). Blood vessels were most numerous in the mucosa (11.79 vessels/0.2 mm(2)) but smaller [median area of 247 µm(2) , interquartile range (IQR) 162-373 µm(2)] than in Sm2, where they were fewer in number (6.92 vessels/0.2 mm(2)) but considerably larger (2086 µm(2), IQR 1007-4784 µm(2)). The 3D model generated novel observations on lymphovascular structures. CONCLUSIONS: The number and size of blood vessels do not increase with depth of submucosa, as hypothesized. The distribution of vessels suggests that we should investigate the area or volume of submucosal invasion rather than the depth.

Comparison of diffusion tensor imaging by cardiovascular magnetic resonance and gadolinium enhanced 3D image intensity approaches to investigation of structural anisotropy in explanted rat hearts.

BACKGROUND: Cardiovascular magnetic resonance (CMR) can through the two methods 3D FLASH and diffusion tensor imaging (DTI) give complementary information on the local orientations of cardiomyocytes and their laminar arrays. METHODS: Eight explanted rat hearts were perfused with Gd-DTPA contrast agent and fixative and imaged in a 9.4T magnet by two types of acquisition: 3D fast low angle shot (FLASH) imaging, voxels 50 × 50 × 50 µm, and 3D spin echo DTI with monopolar diffusion gradients of 3.6 ms duration at 11.5 ms separation, voxels 200 × 200 × 200 µm. The sensitivity of each approach to imaging parameters was explored. RESULTS: The FLASH data showed laminar alignments of voxels with high signal, in keeping with the presumed predominance of contrast in the interstices between sheetlets. It was analysed, using structure-tensor (ST) analysis, to determine the most (v1(ST)), intermediate (v2(ST)) and least (v3(ST)) extended orthogonal directions of signal continuity. The DTI data was analysed to determine the most (e1(DTI)), intermediate (e2(DTI)) and least (e3(DTI)) orthogonal eigenvectors of extent of diffusion. The correspondence between the FLASH and DTI methods was measured and appraised. The most extended direction of FLASH signal (v1(ST)) agreed well with that of diffusion (e1(DTI)) throughout the left ventricle (representative discrepancy in the septum of 13.3 ± 6.7°: median ± absolute deviation) and both were in keeping with the expected local orientations of the long-axis of cardiomyocytes. However, the orientation of the least directions of FLASH signal continuity (v3(ST)) and diffusion (e3(ST)) showed greater discrepancies of up to 27.9 ± 17.4°. Both FLASH (v3(ST)) and DTI (e3(DTI)) where compared to directly measured laminar arrays in the FLASH images. For FLASH the discrepancy between the structure-tensor calculated v3(ST) and the directly measured FLASH laminar array normal was of 9 ± 7° for the lateral wall and 7 ± 9° for the septum (median ± inter quartile range), and for DTI the discrepancy between the calculated v3(DTI) and the directly measured FLASH laminar array normal was 22 ± 14° and 61 ± 53.4°. DTI was relatively insensitive to the number of diffusion directions and to time up to 72 hours post fixation, but was moderately affected by b-value (which was scaled by modifying diffusion gradient pulse strength with fixed gradient pulse separation). Optimal DTI parameters were b = 1000 mm/s(2) and 12 diffusion directions. FLASH acquisitions were relatively insensitive to the image processing parameters explored. CONCLUSIONS: We show that ST analysis of FLASH is a useful and accurate tool in the measurement of cardiac microstructure. While both FLASH and the DTI approaches appear promising for mapping of the alignments of myocytes throughout myocardium, marked discrepancies between the cross myocyte anisotropies deduced from each method call for consideration of their respective limitations.

3-D tissue modelling and virtual pathology as new approaches to study ductal carcinoma in situ.

Widespread screening mammography programmes mean that ductal carcinoma in situ (DCIS), a pre-invasive breast lesion, is now more frequently diagnosed. However, not all diagnosed DCIS lesions progress to invasive breast cancer, which presents a dilemma for clinicians. As such, there is much interest in studying DCIS in the laboratory, in order to help understand more about its biology and determine the characteristics of those that progress to invasion. Greater knowledge would lead to targeted and better DCIS treatment. Here, we outline some of the models available to study DCIS, with a particular focus on animal-free systems.

Computational methods for metastasis detection in lymph nodes and characterization of the metastasis-free lymph node microarchitecture: A systematic-narrative hybrid review.

Background: Histological examination of tumor draining lymph nodes (LNs) plays a vital role in cancer staging and prognostication. However, as soon as a LN is classed as metastasis-free, no further investigation will be performed and thus, potentially clinically relevant information detectable in tumor-free LNs is currently not captured. Objective: To systematically study and critically assess methods for the analysis of digitized histological LN images described in published research. Methods: A systematic search was conducted in several public databases up to December 2023 using relevant search terms. Studies using brightfield light microscopy images of hematoxylin and eosin or immunohistochemically stained LN tissue sections aiming to detect and/or segment LNs, their compartments or metastatic tumor using artificial intelligence (AI) were included. Dataset, AI methodology, cancer type, and study objective were compared between articles. Results: A total of 7201 articles were collected and 73 articles remained for detailed analyses after article screening. Of the remaining articles, 86% aimed at LN metastasis identification, 8% aimed at LN compartment segmentation, and remaining focused on LN contouring. Furthermore, 78% of articles used patch classification and 22% used pixel segmentation models for analyses. Five out of six studies (83%) of metastasis-free LNs were performed on publicly unavailable datasets, making quantitative article comparison impossible. Conclusions: Multi-scale models mimicking multiple microscopy zooms show promise for computational LN analysis. Large-scale datasets are needed to establish the clinical relevance of analyzing metastasis-free LN in detail. Further research is needed to identify clinically interpretable metrics for LN compartment characterization.

Immune subtyping of melanoma whole slide images using multiple instance learning.

Determining early-stage prognostic markers and stratifying patients for effective treatment are two key challenges for improving outcomes for melanoma patients. Previous studies have used tumour transcriptome data to stratify patients into immune subgroups, which were associated with differential melanoma specific survival and potential predictive biomarkers. However, acquiring transcriptome data is a time-consuming and costly process. Moreover, it is not routinely used in the current clinical workflow. Here, we attempt to overcome this by developing deep learning models to classify gigapixel haematoxylin and eosin (H&E) stained pathology slides, which are well established in clinical workflows, into these immune subgroups. We systematically assess six different multiple instance learning (MIL) frameworks, using five different image resolutions and three different feature extraction methods. We show that pathology-specific self-supervised models using 10x resolution patches generate superior representations for the classification of immune subtypes. In addition, in a primary melanoma dataset, we achieve a mean area under the receiver operating characteristic curve (AUC) of 0.80 for classifying histopathology images into 'high' or 'low immune' subgroups and a mean AUC of 0.82 in an independent TCGA melanoma dataset. Furthermore, we show that these models are able to stratify patients into 'high' and 'low immune' subgroups with significantly different melanoma specific survival outcomes (log rank test, P< 0.005). We anticipate that MIL methods will allow us to find new biomarkers of high importance, act as a tool for clinicians to infer the immune landscape of tumours and stratify patients, without needing to carry out additional expensive genetic tests.

Toward routine use of 3D histopathology as a research tool.

Three-dimensional (3D) reconstruction and examination of tissue at microscopic resolution have significant potential to enhance the study of both normal and disease processes, particularly those involving structural changes or those in which the spatial relationship of disease features is important. Although other methods exist for studying tissue in 3D, using conventional histopathological features has significant advantages because it allows for conventional histopathological staining and interpretation techniques. Until now, its use has not been routine in research because of the technical difficulty in constructing 3D tissue models. We describe a novel system for 3D histological reconstruction, integrating whole-slide imaging (virtual slides), image serving, registration, and visualization into one user-friendly package. It produces high-resolution 3D reconstructions with minimal user interaction and can be used in a histopathological laboratory without input from computing specialists. It uses a novel method for slice-to-slice image registration using automatic registration algorithms custom designed for both virtual slides and histopathological images. This system has been applied to >300 separate 3D volumes from eight different tissue types, using a total of 5500 virtual slides comprising 1.45 TB of primary image data. Qualitative and quantitative metrics for the accuracy of 3D reconstruction are provided, with measured registration accuracy approaching 120 µm for a 1-cm piece of tissue. Both 3D tissue volumes and generated 3D models are presented for four demonstrator cases.

Image analysis of cutaneous melanoma histology: a systematic review and meta-analysis.

The current subjective histopathological assessment of cutaneous melanoma is challenging. The application of image analysis algorithms to histological images may facilitate improvements in workflow and prognostication. To date, several individual algorithms applied to melanoma histological images have been reported with variations in approach and reported accuracies. Histological digital images can be created using a camera mounted on a light microscope, or through whole slide image (WSI) generation using a whole slide scanner. Before any such tool could be integrated into clinical workflow, the accuracy of the technology should be carefully evaluated and summarised. Therefore, the objective of this review was to evaluate the accuracy of existing image analysis algorithms applied to digital histological images of cutaneous melanoma. Database searching of PubMed and Embase from inception to 11th March 2022 was conducted alongside citation checking and examining reports from organisations. All studies reporting accuracy of any image analysis applied to histological images of cutaneous melanoma, were included. The reference standard was any histological assessment of haematoxylin and eosin-stained slides and/or immunohistochemical staining. Citations were independently deduplicated and screened by two review authors and disagreements were resolved through discussion. The data was extracted concerning study demographics; type of image analysis; type of reference standard; conditions included and test statistics to construct 2 × 2 tables. Data was extracted in accordance with our protocol and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Diagnostic Test Accuracy (PRISMA-DTA) Statement. A bivariate random-effects meta-analysis was used to estimate summary sensitivities and specificities with 95% confidence intervals (CI). Assessment of methodological quality was conducted using a tailored version of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The primary outcome was the pooled sensitivity and specificity of image analysis applied to cutaneous melanoma histological images. Sixteen studies were included in the systematic review, representing 4,888 specimens. Six studies were included in the meta-analysis. The mean sensitivity and specificity of automated image analysis algorithms applied to melanoma histological images was 90% (CI 82%, 95%) and 92% (CI 79%, 97%), respectively. Based on limited and heterogeneous data, image analysis appears to offer high accuracy when applied to histological images of cutaneous melanoma. However, given the early exploratory nature of these studies, further development work is necessary to improve their performance.

Completing the view - histologic insights from circular AAA specimen including 3D imaging : A methodologic approach towards histologic analysis of circumferential AAA samples.

Abdominal aortic aneurysm (AAA) is a pathologic enlargement of the infrarenal aorta with an associated risk of rupture. However, the responsible mechanisms are only partially understood. Based on murine and human samples, a heterogeneous distribution of characteristic pathologic features across the aneurysm circumference is expected. Yet, complete histologic workup of the aneurysm sac is scarcely reported. Here, samples from five AAAs covering the complete circumference partially as aortic rings are investigated by histologic means (HE, EvG, immunohistochemistry) and a new method embedding the complete ring. Additionally, two different methods of serial histologic section alignment are applied to create a 3D view. The typical histopathologic features of AAA, elastic fiber degradation, matrix remodeling with collagen deposition, calcification, inflammatory cell infiltration and thrombus coverage were distributed without recognizable pattern across the aneurysm sac in all five patients. Analysis of digitally scanned entire aortic rings facilitates the visualization of these observations. Immunohistochemistry is feasible in such specimen, however, tricky due to tissue disintegration. 3D image stacks were created using open-source and non-generic software correcting for non-rigid warping between consecutive sections. Secondly, 3D image viewers allowed visualization of in-depth changes of the investigated pathologic hallmarks. In conclusion, this exploratory descriptive study demonstrates a heterogeneous histomorphology around the AAA circumference. Warranting an increased sample size, these results might need to be considered in future mechanistic research, especially in reference to intraluminal thrombus coverage. 3D histology of such circular specimen could be a valuable visualization tool for further analysis.

Automated detection and delineation of lymph nodes in haematoxylin & eosin stained digitised slides.

Treatment of patients with oesophageal and gastric cancer (OeGC) is guided by disease stage, patient performance status and preferences. Lymph node (LN) status is one of the strongest prognostic factors for OeGC patients. However, survival varies between patients with the same disease stage and LN status. We recently showed that LN size from patients with OeGC might also have prognostic value, thus making delineations of LNs essential for size estimation and the extraction of other imaging biomarkers. We hypothesized that a machine learning workflow is able to: (1) find digital H&E stained slides containing LNs, (2) create a scoring system providing degrees of certainty for the results, and (3) delineate LNs in those images. To train and validate the pipeline, we used 1695 H&E slides from the OE02 trial. The dataset was divided into training (80%) and validation (20%). The model was tested on an external dataset of 826 H&E slides from the OE05 trial. U-Net architecture was used to generate prediction maps from which predefined features were extracted. These features were subsequently used to train an XGBoost model to determine if a region truly contained a LN. With our innovative method, the balanced accuracies of the LN detection were 0.93 on the validation dataset (0.83 on the test dataset) compared to 0.81 (0.81) on the validation (test) datasets when using the standard method of thresholding U-Net predictions to arrive at a binary mask. Our method allowed for the creation of an "uncertain" category, and partly limited false-positive predictions on the external dataset. The mean Dice score was 0.73 (0.60) per-image and 0.66 (0.48) per-LN for the validation (test) datasets. Our pipeline detects images with LNs more accurately than conventional methods, and high-throughput delineation of LNs can facilitate future LN content analyses of large datasets.

Tracking with virtual slides: a tool to study diagnostic error in histopathology.

AIMS: To determine the reasons for diagnostic error by virtual slides which allow unsupervised study of diagnosis and error. METHODS AND RESULTS: Software was developed to produce visualizations of the diagnostic track followed by pathologists as they viewed virtual slides. These showed the diagnostic path in four dimensions (x, y, time and zoom), areas studied for >1000 ms, and included pathologists' comments about the areas viewed. The system was used to study two trainee and two expert pathologists diagnosing 60 Barrett's oesophageal biopsy specimens. Comparisons of the diagnostic tracks showed the reason for errors. Forty-six cases had an expert consensus diagnosis. The trainees made errors in 21 and 15 cases, respectively, of which 11 and nine were clinically significant. Errors were made across the whole spectrum of diagnoses from negative to intramucosal carcinoma. Detailed examination of the tracks showed that in all errors there was incorrect interpretation of information; in three errors there was an additional failure to identify diagnostic features. CONCLUSIONS: Tracking with virtual slides is a useful tool in studying diagnosis and error, which has the potential for use in training and assessment.

A physics based method for combining multiple anatomy models with application to medical simulation.

We present a physics based approach to the construction of anatomy models by combining components from different sources; different image modalities, protocols, and patients. Given an initial anatomy, a mass-spring model is generated which mimics the physical properties of the solid anatomy components. This helps maintain valid spatial relationships between the components, as well as the validity of their shapes. Combination can be either replacing/modifying an existing component, or inserting a new component. The external forces that deform the model components to fit the new shape are estimated from Gradient Vector Flow and Distance Transform maps. We demonstrate the applicability and validity of the described approach in the area of medical simulation, by showing the processes of non-rigid surface alignment, component replacement, and component insertion.