PathProfiler: Automated Quality Assessment of Retrospective Histopathology Whole-Slide Image Cohorts by Artificial Intelligence, A Case Study for Prostate Cancer Research

Research using whole slide images (WSIs) of scanned histopathology slides for the development of artificial intelligence (AI) algorithms has increased exponentially over recent years. Glass slides from large retrospective cohorts with patient follow-up data are digitised for the development and validation of AI tools. Such resources, therefore, become very important, with the need to ensure that their quality is of the standard necessary for downstream AI development. However, manual quality control of such large cohorts of WSIs by visual assessment is unfeasible, and whilst quality control AI algorithms exist, these focus on bespoke aspects of image quality, e.g. focus, or use traditional machine-learning methods such as hand-crafted features, which are unable to classify the range of potential image artefacts that should be considered. In this study, we have trained and validated a multi-task deep neural network to automate the process of quality control of a large retrospective cohort of prostate cases from which glass slides have been scanned several years after production, to determine both the usability of the images for research and the common image artefacts present. Using a two-layer approach, quality overlays of WSIs were generated from a quality assessment undertaken at patch-level at 5X magnification. From these quality overlays the slide-level quality scores were predicted and then compared to those generated by three specialist urological pathologists, with a Pearson correlation of 0.89 for overall usability (at a diagnostic level), and 0.87 and 0.82 for focus and H&E staining quality scores respectively. We subsequently applied our quality assessment pipeline to the TCGA prostate cancer cohort and to a colorectal cancer cohort, for comparison. Our model, designated as PathProfiler, indicates comparable predicted usability of images from the cohorts assessed (86-90%), and perhaps more significantly is able to predicts WSIs that could benefit from re-scanning or re-staining for quality improvement. We have shown in this study that AI can be used to automate the process of quality control of large retrospective cohorts to maximise research outputs and conclusions.

CBCT image artefacts generated by implants located inside the field of view or in the exomass

Objectives: To compare artefacts in cone-beam computed tomography (CBCT) arising from implants of different materials located either inside the field-of-view (FOV) or in the exomass, and to test different image-acquisition parameters to reduce them. Methods: CBCT scans of a human mandible prepared with either a titanium, titanium-zirconium, or zirconia implant were acquired with the Planmeca ProMax utilizing FOV sizes of 8 × 5 cm and 4 × 5 cm, which placed the implant inside the FOV (8 × 5 cm) or in the exomass (4 × 5 cm). The scanning parameters considered three conditions of metal artefact reduction (MAR), disabled, low, and high, and two kVp levels (80 and 90). The standard deviation (SD) of grey values of regions of interest was obtained. The effects of implant material, implant position, MAR condition, kVp level, and their interactions were evaluated by Analysis of Variance (α = 5%). Results: The zirconia implant produced the highest SD values (more heterogeneous grey values, corresponding to greater artefact expression), followed by titanium-zirconium, and titanium. In general, implants in the exomass produced images with higher SD values than implants inside the FOV. MAR was effective in decreasing SD values, especially from the zirconia implant, only when the implant was inside the FOV. Images with 80 kVp had higher SD values than those with 90 kVp, regardless of the other factors (p < 0.05). Conclusions: Implants in the exomass lead to greater artefact expression than when they are inside the FOV. Special attention should be paid to scanning parameters that reduce metal-related artefacts, such as MAR activation and increasing kVp. This is especially important with a zirconia implant inside the FOV.

PathML: A unified framework for whole-slide image analysis with deep learning

The inspection of stained tissue slides by pathologists is essential for the early detection, diagnosis and monitoring of disease. Recently, deep learning methods for the analysis of whole-slide images (WSIs) have shown excellent performance on these tasks, and have the potential to substantially reduce the workload of pathologists. However, successful implementation of deep learning for WSI analysis is complex and requires careful consideration of model hyperparameters, slide and image artefacts, and data augmentation. Here we introduce PathML, a Python library for performing pre- and post-processing of WSIs, which has been designed to interact with the most widely used deep learning libraries, PyTorch and TensorFlow, thus allowing seamless integration into deep learning workflows. We present the current best practices in deep learning for WSI analysis, and give a step-by-step guide using the PathML framework: from annotating and pre-processing of slides, to implementing neural network architectures, to training and post-processing. PathML provides a unified framework in which deep learning methods for WSI analysis can be developed and applied, thus increasing the accessibility of an important new application of deep learning.

Assessment of deep learning algorithms for 3D instance segmentation of confocal image datasets

Segmenting three dimensional microscopy images is essential for understanding phenomena like morphogenesis, cell division, cellular growth and genetic expression patterns. Recently, deep learning (DL) pipelines have been developed which claim to provide high accuracy segmentation of cellular images and are increasingly considered as the state-of-the-art for image segmentation problems. However, it remains difficult to define their relative performance as the concurrent diversity and lack of uniform evaluation strategies makes it difficult to know how their results compare. In this paper, we first made an inventory of the available DL methods for 3D segmentation. We next implemented and quantitatively compared a number of representative DL pipelines, alongside a highly efficient non-DL method named MARS. The DL methods were trained on a common dataset of 3D cellular confocal microscopy images. Their segmentation accuracies were also tested in the presence of different image artefacts. A new method for segmentation quality evaluation was adopted which isolates segmentation errors due to under/over segmentation. This is complemented with new visualisation strategies that make interactive exploration of segmentation quality possible. Our analysis shows that the DL pipelines have very different levels of accuracy. Two of them show high performance, and offer clear advantages in terms of adaptability to new data.

Image Artefacts in Industrial Computed Tomography

Computed tomography is a method that has been used for many years in medicine and material analysis, and recently it has also been introduced in dimensional measurements. The method has a lot of advantages compared to other 3D measurement methods, with the largest one being the possibility to perform a non-destructive measurement of an object’s inner geometry. However, it is a complex method with a large number of parameters that influence measurement results. Some of these parameters are image artefacts that occur in the scanning and reconstruction process. An artefact is any artificial feature which appears on the CT image, but does not correspond to the physical feature of an object. In order to achieve metrological traceability, it is necessary to eliminate and minimize the influence of image artefacts on measurement results. This paper presents and explains image artefacts in industrial computed tomography as the consequences of different influence parameters in the CT system.

TanzArchiv Leipzig – Disappearing Content and Traces of Past Events

The TanzArchiv Leipzig (TAL) presents itself as a precarious archive of dance that blossomed in dubious political times. It was founded when East Germany, officially known as the German Democratic Republic (GDR), was a country during 1949–1990, in which art and culture were valued as national currency ( Bourdieu 1986 ; Lohman 1994 ). Although the archive had lost its domicile as an Institution of the GDR (1989) as part of a larger Institution of the Academy of Arts (Akademie der Kunst), then it continued to act as a research centre in the Institute of the House of Literature (Haus des Buches), then renting its own premises as a foundation thereafter (ca. 1993–2010) and finally, is currently stored since 2011 as the TAL collection in the Special Collections department in the Albertina Library, at the University of Leipzig ( Reinsberg 2002 ; Ruiz [2002] ; 2018). The archival collection embraces a large collection of ‘traces’ of dance content such as manuscripts, dance scores, film, sound and image artefacts as well as objects, publications and a variety of ephemera. However, its fate as an archive of a country that no longer exists, and the question of the preservation and circulation of its content make it an ambiguous and challenging dance archive to examine in full. In this article I will focus on the description and structure of the archive, the dissemination strategies Documenta Choreologica 1 and Kurt Petermann's passion for dance transmission, through his letter correspondence within and without East European countries during the Cold War ( Boehme 1948 ; Dafova 1996 ; Guilbert 2007 ).

Data- and model-domain up/down wave separation for reverse-time migration with free-surface multiples

SUMMARY Seismic data recorded using a marine acquisition geometry contain both upgoing reflections from subsurface structures and downgoing ghost waves reflected back from the free surface. In addition to the ambiguity of propagation directions in the data, using the two-way wave equation for wavefield extrapolation of seismic imaging generates backscattered/turned waves when there are strong velocity contrasts/gradients in the model, which further increases the wavefield complexity. For reverse-time migration (RTM) of free-surface multiples, apart from unwanted crosstalk between inconsistent orders of reflections, image artefacts can also be formed along with the true reflector images from the overlapping of up/downgoing waves in the data and in the extrapolated wavefield. We present a wave-equation-based, hybrid (data- and model-domain) wave separation workflow, with vector seismic data containing pressure- and vertical-component particle velocity from dual-sensor seismic acquisition, for removing image artefacts produced by the mixture of up/downgoing waves. For imaging with free-surface multiples, the wavefield extrapolated from downgoing ghost events (reflected from the free surface) in the recorded data act as an effective source wavefield for one-order-higher free-surface multiples. Therefore, only the downgoing waves in the data should be used as the source wavefield for RTM with multiples; the recorded upgoing waves in the seismograms will be used for extrapolation of the time-reversed receiver wavefield. We use finite-difference (FD) injection for up/down separation in the data domain, to extrapolate the down- and upgoing waves of the common-source gathers for source and receiver wavefield propagation, respectively. The model-domain separation decomposes the extrapolated wavefield into upgoing (backscattered) and downgoing (transmitted) components at each subsurface grid location, to remove false images produced by cross-correlating backscattered waves along unphysical paths. We combine FD injection with the model-domain wavefield separation, for separating the wavefield into up- and downgoing components for the recorded data and for the extrapolated wavefields. Numerical examples using a simple model, and the Sigsbee 2B model, demonstrate that the hybrid up/down separation approach can effectively produce seismic images of free-surface multiples with better resolution and fewer artefacts.