Novel context-based segmentation algorithms for intelligent microscopy

Lead Research Organisation: University of Birmingham
Department Name: Dentistry

Abstract

The aim of this proposal is to develop new computer methods to analyse, quantify and understand the information contained in images of cells and tissues obtained using digital microscopy. This will have wide applications in many areas of biomedicine, especially in histopathology, where it can be used to diagnose cancer, predict the potential behaviour of malignant disease, and implement the most suitable form of treatment.
Currently, decisions about histopathological diagnosis rely to a great extent on expert observers and their experience, but this has the disadvantage that the inevitable element of subjectivity in visual observation makes it difficult to reach quantitative or reproducible judgements.
In this project we will design 'context-based' imaging programs to help advance automated analysis and diagnosis of cancer. By 'context-based' here we understand the use of data constructs that (1) allow the structure and relations of cells and tissues in biopsy samples to be modelled in a way that enables computer programs to subsequently 'reason' about the image contents, and (2) allow methods of data extraction that are both quantitative and reproducible.
This will be achieved by using a spatial logic called Discrete Mereotopology (DM). We already have proof-of-concept software and a peer-reviewed preliminary publication which demonstrate the efficacy of this approach for encoding and querying relations between the biologically-relevant entities (e.g. cell nucleus, tissue layers, staining patterns) and the image segments detected to correspond to them, thereby enabling histologically relevant models (e.g. cells, tissue types and voids) to be formulated that can be operated on at a level that has not been possible before. This is a departure from traditional pixel-based routines, leading to region-based algorithms that are histologically relevant to the range of images and structures that are expected in histological imaging.
This logic-based approach to image analysis brings several advantages. Firstly, it provides a robust, rigorous mathematical foundation for software development. Secondly, it allows histological images to be systematically interpreted in terms of histologically-relevant entities, not just pixels. Thirdly, it enables symbolic and automated reasoning programs to be used alongside numerical methods.
We will also incorporate immunohistochemical markers to our histologically relevant models. A key innovation in the proposal is the use of DM to explicitly model the histological localisation of molecular markers, which has never been done before. This will allow a rational evaluation of immunohistochemical patterns and subsequent development of histological predictor markers associated with tumour behaviour and outcome, i.e., what patterns are observed in what kinds of neoplasm, where in tissues the markers are expressed, and how characteristic the patterns are to recorded case outcomes.
Finally we will develop markers of data quality to label structures according to how well they approach the expected models they represent. In this way indications of 'confidence in the results' can be associated to the models found in microscopy images. So far this feature is not provided by any form of histological imaging.
We believe that the approach outlined here is both translational and invaluable in most biological areas using microscopy, where quantitative results are required to make sound evidence-based decisions.

Planned Impact

This is a cross-disciplinary translational project where the immediate beneficiaries are cancer patients and the healthcare profession. The overall aim of this research is to address clinical needs in the performance of imaging programs for understanding biological data from microscopy images. While this proposed project specifically targets the diagnosis and prognosis of oropharyngeal carcinomas, the methods adopted are by no means restricted to this type of cancer, or even cancer in general, but will be applicable to most biological areas dealing with interpretation of microscopy images of cells and tissues. Deliverables from this project will make an impact by enabling imaging technologies to elevate the interpretation of biological data (histological imagery) to levels that so far have been difficult to establish.

For the healthcare profession this extends to impact arising from new software tools that enable reliable data extraction and mechanical reasoning on microscopy image contents in ways that have not been possible before. These will provide the means to implement high-throughput systems to analyse more data than is currently feasible, to develop new histological diagnostic standards based on quantifiable evidence-based research, and to create better diagnostic and prognostic models of disease.

The expectations of impact in society are high.
1) Faster and more reliable diagnosis of cancers, resulting in quicker turnaround, with fewer delays in instituting a definitive therapy or adjuvant therapy post-operatively. As outcomes have been shown to be associated with time delay to therapy, faster diagnosis and quicker treatment times are likely to result in better survival rates.
2) Studies have shown that the period of awaiting a definitive diagnosis is the time of highest stress for patients. Faster turnaround in analysing samples will help to ameliorate this.
3) Higher reliability of tests will result in fewer incorrect diagnoses, with less harm to patients and lower health, social and economic costs of implementing incorrect treatments.
4) There is a shortage of histopathologists in the UK. By enabling their work to be more efficient through automation, this will result in a decrease in the delays caused by overstretched resources, and consequently in lower costs, that would translate into efficient use of resources for the NHS.

In terms of benefits to academia and commercial research, the work will advance context-based imaging algorithms that are sensitive to the spatial localisation of features in tissues as it will make it possible to pinpoint not only what cell and tissue patterns are detected, but also where within the samples those features of interest are localised. The research will also provide key insights to the qualitative expression of immunohistochemical markers in images to enable those markers to be related to tissue biological behaviour and therefore to the prognosis of tumours.
 
Description * We developed two automated methods for extracting cell nuclei from microscopy images, inspired by strategies observed in humans when adjusting manually sliders controls. One of the methods achieved the best performance versus a gold standard, when compared to other 23 existing thresholding methods. We also developed a novel boundary concavity-based method for separation of nuclei clusters in images to infer merged regions.

* We digitised samples of oropharyngeal carcinomas in tissue micro array format (the database is currently 8772 images including 3 staining methods). We developed methods to reduce the computational cost of analysing the images by automatically eliminating unnecessary background space. On a subset of those, hand-generated gold standards were produced for 3 types of image components and various segmentation and clustering procedures were implemented. Preliminary results indicate that the 3 types of components can be separated using unsupervised clustering methods with ~80% accuracy without human interaction, opening a possible avenue for automated sample pre-screening and data driven analysis.

* We have developed a model-based method to correct segmentation errors using relations between regions. We found that the number of operators and operations needed for certain corrections dictate the complexity of the procedure which could become computationally expensive unless other constraints (shape and size) are taken into consideration. We found links between these procedures and the ontological levels referred earlier to restrict the amount of computation necessary.

* We found that ordering a sequence of spatial configurations to meet certain transition constraints is NP-complete when using the RCC8 relation set. Consequently, a framework was proposed where the temporal aspect of that sequence is constrained by a Point Algebra network. This again showed that the task remains NP complete.

* We formulated an ontological histological imaging framework, integrating quantitative and algorithmic analyses of cells and tissue images. This enables hypothesising about biologically meaningful models from microscopy images. This process can be understood as a progression through various ontological levels, each populated by entities related in systematic ways to entities at other levels. This also provides a useful basis for classifying the various kinds of artefacts and anomalies that arise at different stages of the imaging pipeline and we are currently working on publications to present these ideas.

* Collaborative work with researchers in Brazil on Connectivity Descriptors resulted in a new generic multiscale texture image analysis method. This provided the best performance in texture classification when compared to other state of the art methods on three well established databases. We also investigated the Bouligand-Minkowski descriptors for classifying tissue architecture in three types jaw cysts (radicular cyst, solitary- and syndromic- odontogenic keratocysts). These descriptors submitted to a machine learning algorithm were able to discriminate between radicular and odontogenic keratocysts in 98% of the images, and 68% of two subtypes of keratocysts, improving over previously reported classification rates in the literature. Current collaboration on deep learning is extending this stream of research.

* Our algorithms for tissue segmentation/architectural analysis were applied to a biodiversity problem in the analysis of plant taxonomy through histology. Geometrical measures from cells and tissue layers were combined into vectors of features for species categorization. This achieved a classification success rate of 91.7% when applied to a histological database of 10 plant species from Brazil, outperforming other classical shape-based approaches in the literature.* We developed an imaging workflow to detect human papilloma virus status in tissue microarray images of in situ hibridisation tests, in oropharyngeal carcinomas with a 90% accuracy compared to expert recorded scores on the same dataset.

* Developed a new fast computational algorithm for Region Connection Calculus with a performance increase in execution speed of more than 1400 times than our previously published algorithm.

* Developed an unsupervised tissue detecting algorithm based on multiscale superpixel clustering that on average performs better than single scale analyses.

* We expanded the operators necessary to perform tissue segmentation corrections to include topological and set-theoretic operators.

* We developed a Deep Learning network to detect epithelium in tissue micro array images of in situ hybridisation solely based on single channel counterstain data. This is necessary to be able to automate detection of in situ hybridisation products such as human papillomavirus (HPV) infection. This is now being extended to integrate Discrete Mereotopology to detect in situ HPV products in the epithelial compartment of tissues.
Exploitation Route We have not yet released all our software, pending on publications, but we anticipate sharing our software tools with imaging community when the outcome of the peer review of the papers is received.
Sectors Digital/Communication/Information Technologies (including Software),Education,Healthcare,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Description Spatial reasoning methods in histological imaging have now been integrated (since 2017) to be presented at the Image Analysis Training School (IATS) run by Nottingham University Molecular Pathology Node. This year this will be also presented at the IATS in Vienna and Nottingham. This highlights the interest and potential for changes in practice in the methods for histological imaging developed during this project.
First Year Of Impact 2017
Sector Healthcare
 
Description Brazil- Sao Paulo University & Campinas University 
Organisation State University of Campinas
Department Center for Computational Engineering & Sciences
Country Brazil 
Sector Academic/University 
PI Contribution Developed new imaging algorithms for image texture recognition. Applied tissue architecture analysis to histological images of plants Investigated the potential for histological image classification using Bouligand-Minkowski descriptors for discrimination of jaw cysts images.
Collaborator Contribution The collaboration with Sao Paulo University (Institute of Physics) and later Campinas University (Dept of Mathematics) was forged from a previous visit by one of the Brazilian collaborators and continues to date. The collaboration has brought into our work expertise in computational texture analysis and classification of histological images. It has provided an excellent opportunity to investigate the generic properties of our algorithms in different domains. For example, some of our methods in tissue architecture analysis were applied to data from the Brazil team for taxonomic analysis of plant species.
Impact Multidisciplinary collaboration (Computational Physics, Pathology, Imaging) doi: 10.1016/j.compbiomed.2016.12.003 Multidisciplinary, Physics, Computer Science, Histopathology doi: 10.1016/j.patrec.2016.09.013, Multidisciplinary, Computer Science, Image Analysis doi: 10.1139/cjb-2015-0075 Multidisciplinary, Computer Science, Histopathology, Botany
Start Year 2015
 
Description Brazil- Sao Paulo University & Campinas University 
Organisation University of Sao Paulo
Department Physics Institute of São Carlos
Country Brazil 
Sector Academic/University 
PI Contribution Developed new imaging algorithms for image texture recognition. Applied tissue architecture analysis to histological images of plants Investigated the potential for histological image classification using Bouligand-Minkowski descriptors for discrimination of jaw cysts images.
Collaborator Contribution The collaboration with Sao Paulo University (Institute of Physics) and later Campinas University (Dept of Mathematics) was forged from a previous visit by one of the Brazilian collaborators and continues to date. The collaboration has brought into our work expertise in computational texture analysis and classification of histological images. It has provided an excellent opportunity to investigate the generic properties of our algorithms in different domains. For example, some of our methods in tissue architecture analysis were applied to data from the Brazil team for taxonomic analysis of plant species.
Impact Multidisciplinary collaboration (Computational Physics, Pathology, Imaging) doi: 10.1016/j.compbiomed.2016.12.003 Multidisciplinary, Physics, Computer Science, Histopathology doi: 10.1016/j.patrec.2016.09.013, Multidisciplinary, Computer Science, Image Analysis doi: 10.1139/cjb-2015-0075 Multidisciplinary, Computer Science, Histopathology, Botany
Start Year 2015
 
Description Universite´ d'Artois, Lens, France 
Organisation Artois University
PI Contribution Our experience in histological imaging and spatial logics brought a real-world application to the development of efficient graph search algorithms used in theoretical Qualitative Spatial Reasoning constraint-based modelling.
Collaborator Contribution The collaborators wrote and released software (Python code) for RCC8 relational consistency checking. This is important for the work we do in relation to situations when predicted sets of spatial (or spatio-temporal) relations have models in out labelled segmented images.
Impact Multidisciplinary research (Spatial and Temporal Reasoning, Computer Science, Histological Imaging). Paper published: 10.1007/978-3-319-23868-5_10
Start Year 2015
 
Description University of Naples 
Organisation University of Naples
Country Italy 
Sector Academic/University 
PI Contribution The collaboration with Italy is still ongoing and applied histological imaging to the analysis of Haemic Neoplasia cell populations in molluscs.
Collaborator Contribution We developed workflows to quantify the nuclear morphometry and DNA content of neoplastic cells of mussels Mytilus galloprovincialis affected by a not fully characterised type of cancer (Haemic Neoplasia). We showed that haemocytes in this disease have distinct morphological features and DNA abnormalities (aneuploidy) that are suggested as markers of disease progression. A set of 120,166 nuclei were analysed and 21 morphological parameters from normal and neoplastic nuclei. Eighteen of these parameters were different and allowed us to discriminate between normal and neoplastic cells (87.6% and 89.2% respectively). Those differences allowed identifying two distinctive populations of neoplastic nuclei, occasionally in the same individuals at a given phase of the disease.
Impact Multidisciplinary research (Veterinary Pathology, Microscopy, Imaging). Paper published: Carella F, De Vico G, Landini G. Nuclear morphometry and ploidy of normal and neoplastic haemocytes in mussels. PloS One 12 (3): 1-18, 2017, e0173219. doi:10.1371/journal.pone.0173219 Conference presentations: Carella F, De Vico G, Landini G. A morphometric study of normal, inflammatory and neoplastic haemocytes in mediterranean mussel Mytilus Galloprovincialis. 17th International Conference on Diseases of Fish & Shellfish, Las Palmas de Gran Canaria, 7-11 Sep, 2015. Carella F, Landini G, Devico G. Haemic Neoplasia. 2nd International Symposium on the Advances in Marine Mussel Research, Montpellier, Sep 7-8, 2017.
Start Year 2015
 
Title Automatic thresholding plugins for ImageJ 
Description The procedures implemented are based on a reinterpretation of the strategy observed in human operators when adjusting thresholds manually and interactively by means of 'slider' controls. Users most often find the "right" threshold value by repeatedly over- and under-thresholding the image until the thresholded phase more or less coincides with the sought objects in the image. The advantage of the computerised versions presented here is that the methods do not suffer from the uncertainty of "when to stop adjusting" the threshold. Two different methods were implemented. The first one is a simple global thresholding procedure suitable for single or multiple global thresholds (i.e. one value for the whole image). The procedure consists of searching the greyscale space for a threshold value that generates a phase whose boundary coincides with the largest gradients in the original image. The plugin is called Threshold_Global_Gradient and it has the option of using two possible measures of the phase gradient (the mean and the total gradient). The second method is a more complex variation of the same principle, but operates on the discrete connected components of the thresholded phase (i.e. candidate binary regions that might represent the intended object in the image) independently, therefore becoming an adaptive local thresholding procedure which operates on regions, rather than on pixels of local image subsets as is the case in the vast majority of local thresholding methods in the literature. 
Type Of Technology Software 
Year Produced 2016 
Open Source License? Yes  
Impact Provides an automated ways of nuclear thresholding/segmentation in sections of tissues. In terms of performance on our samples, the software provides better (more accurate and cleaner) results than other currently methods available to perform the same task. For details on performance, see Landini G, Randell DA, Fouad S, Galton A. Automatic thresholding from the gradients of region boundaries. Journal of Microscopy, 2016. 
URL http://www.mecourse.com/landinig/software/threshgrad/threshgrad.html
 
Title Consensus Clustering Ensemble for ImageJ 
Description A plugin featuring an algorithm that combines the results of multiple based clusterings into a single consensual result. Includes ImageJ macro code for segmenting relevant regions to be submitted to various clustering methods. Requires the Weka library. http://www.cs.waikato.ac.nz/~ml/weka/downloading.html and the ImageJ Morphology collection http://www.mecourse.com/landinig/software/software.html 
Type Of Technology Software 
Year Produced 2017 
Open Source License? Yes  
Impact The plugin allows algorithmic segmentation of histologically relevant regions of tissues stained with haematoxylin and eosin (or other stains) in digitized histopathological images is a critical step towards computer-assisted diagnosis and analysis. Consensus Clustering is used here for the unsupervised recognition of tissue compartments, that consider the consensual opinion of a group of individual clustering algorithms. Unlike most unsupervised segmentation analyses, which depend on a single clustering method, the CC learning models allow for more robust and stable detection of tissue regions. The unsupervised results obtained with the algorithm outperform those obtained with individual clustering algorithms. 
URL https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0188717
 
Title Fast Region Connection Calculus software suite 
Description A suite of ImageJ plugins and macros to perform spatial reasoning using the region connection calculus (RCC) using Discrete Mereotopology (DM) principles and applied to single region pairs embedded in different images, as well as multiple regions pairs in different images. Other analyses include a new type of morphological operations: minimal opening, minimal closing, as well as methods to extract topological properties of image regions, querying images for determined spatial relations between regions, e.g. cellular inclusions and cell model (nuclear cytoplasmic) relations. We developed principles to detect and correct segmentation artefacts using RCC and DM. 
Type Of Technology Software 
Year Produced 2018 
Open Source License? Yes  
Impact The new programmes to perform region calculus analysis are approximately up to 1400 times faster than our initial implementation, which are essential for applications in high throughput environments. Software is currently available from the authors. We are awaiting the results of the review of a paper describing the methods to release the software under open source license. 
 
Title Mathematical Morphology plugins for ImageJ 
Description A collection of more than 100 ImageJ plugins to perform various mathematical morphology operations. Several of the plugins in the collection existed before this project, but new versions were written with improvements and speeded up to allow Fast Region Connection Calculus software described elsewhere. We are awaiting the results of the review of a paper describing the methods to release the software under open source license. An older version of the software can be downloaded from http://www.mecourse.com/landinig/software/software.html 
Type Of Technology Software 
Year Produced 2018 
Open Source License? Yes  
Impact New as well as improved versions of mathematical morphology plugins have enabled writing imaging procedures that can be applied in high throughput applications using open source platforms (Fiji, ImageJ). 
 
Description Biosciences lecture 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Undergraduate students
Results and Impact Lecture to Biosciences students (year 2), Birmingham University (7 Feb 2017) on epithelial tissue including architectural features, and pathologies of the oral mucosa. This included a showcase of diagnostic problems faced by histopathologists and some of the solutions we are pursuing in characterising those tissues.
Year(s) Of Engagement Activity 2017
 
Description CORLAS Congress, Beijing 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Mini symposium on Head and Neck at the CORLAS congress (Sept 16-19, 2018, Beijing): Prof Mehanna discussed the perspective of margins for open versus endoscopic surgery in view of the current evidence available and the prospect of newly developed digital technologies for accurate evaluation of histological material.
Year(s) Of Engagement Activity 2018
URL http://www.collegium2018.com.cn/
 
Description Carella F, Devico J, Landini G. A morphometric study of normal, inflammatory and neoplastic haemocytes in mediterranean mussel Mytilus Galloprovincialis. Presented at 17th International Conference on Diseases of Fish and Shellfish 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Collaborative research with Veterinary Pathology. Presented preliminary results of digital imaging analysis applied to DNA cytometry in haemocytes in mussels. Sparked further work which resulted in a journal publication.
Year(s) Of Engagement Activity 2015
URL http://www.barcelocongresos.com.es/archivo/2015EAFP/
 
Description Engagement with Industry: GSK Dental Discovery Day 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Industry/Business
Results and Impact Showcase of Preclinical and Clinical Research at Birmingham School of Dentistry to GlaxoSmithKline (Regional Clinical Operations Director for EU, Innovation Snr Scientist, Media Affairs, Clinical Director for Oral Health, Category Medical Affairs Principal Scientist Oral Health, College Business Engagement Partner.Business Engagement Officer, others) . Presented our current progress on Intelligent Microscopy in relation to cancer diagnosis and applications of the techniques to other areas of biological imaging.
Year(s) Of Engagement Activity 2016
 
Description Fouad S, Landini G, Randell D, Galton A, Mehanna H. Unsupervised and semi-supervised learning frameworks for the epithelium and stroma segmentation in histopathological images. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Galton A, AI - a (very) quick overview. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Galton A, Landini G, Randell D, Fouad S. Ontology levels in histological imaging. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description ICBO 2017 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Poster and Flash Presentation at the 8th International Conference on Biomedical Ontology (ICBO) 2017 meeting: Galton A, Fouad S, Landini G, Randell D. Errors and Artefacts in Histopathological Imaging.
Year(s) Of Engagement Activity 2017
URL http://ceur-ws.org/Vol-2137/paper_11.pdf
 
Description Invited talk at the Image Analysis Training School, Nottingham University 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact About 25 attendants: pathologists, imaging scientist, industry/manufacturers. AidPath sponsored event, part of the Image Analysis Training School at Nottingham University. Focus on Digital Pathology.
http://aidpath.eu/wp-content/uploads/2017/09/Image-Analysis-Training-School-2017.pdf
Year(s) Of Engagement Activity 2017
URL https://www.nottingham.ac.uk/research/groups/pathology/events/2017-image-analysis-training-school.as...
 
Description Landini G, Randell D, Galton A, Fouad S. Discrete Mereotopology in histological imaging. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Landini G, Randell DA, Fouad SA. Histopathological Problems and Imaging. Invited seminar, Warwick University, Department of Computer Science. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Postgraduate students
Results and Impact Presented principles of mererotopology and intelligent microscopy applied to histological imaging and discussed common problems in the subject area.

Had the opportunity to show our approach to spatial reasoning in histological imaging problems as well as learning about the activities ongoing at Warwick University.
Set up excellent contacts for future collaboration.
Year(s) Of Engagement Activity 2015
 
Description Ontological Levels in Histological Imaging. Presented by Dr G. Galton at Exeter Imaging Network. 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Postgraduate students
Results and Impact A presentation was made to about 30 people in the Exeter Imaging Network (part of GW4 partnership) on our work on ontologies in histological imaging. Interest was expressed in the work and a request for a paper recently submitted to FOIS 2016.
Year(s) Of Engagement Activity 2016
 
Description Organised the AI & Digital Microscopy Meeting, Exeter 25-26 June, 2018 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We organised a two day meeting on "AI and Digital Microscopy" at Exeter University. It was a multidisciplinary meeting, with participants involved in the analysis of bio-medical and materials science imagery.
There meeting involvedwere 2 key note lectures on AI and imaging problems, 7 presentations on research funded by this projects, 11 presentations by attendants on various subjects of imaging and AI, and two discussion sessions.
Year(s) Of Engagement Activity 2018
URL http://mecourse.com/landinig/ai/
 
Description Randell D, Galton A, Landini G, Fouad S, Mehanna H. Resegmenting Digitised Histological Images using Discrete Mereotopology. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Randell D, Galton A, Landini G, Fouad S. Discrete Mereotopology: an introduction. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Seminar on Histological Imaging, Computer Science Department, Aberystwyth University. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Seminar talk to Computer Science department.
Year(s) Of Engagement Activity 2016
 
Description Song T-H, Landini G, Randell D, Mehanna H.Tissue segmentation for HPV status assessment. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018
 
Description Spatial reasoning for histological imaging - Image Analysis Training School 2018, Nottingham University 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact About 20 attendants: pathologists, imaging scientist, industry/manufacturers. NMPN (Nottingham Molecular Pathology Node) sponsored event, part of the Image Analysis Training School at Nottingham University. Focus on Digital Pathology.
Year(s) Of Engagement Activity 2018
URL https://www.nmpn.info/imageanalysis2018
 
Description Sui T, Landini G. Advanced time-resolved microscopy reveals the nanostructure and function of human dental tissue. AI & Digital Microscopy, Exeter, Jun 25-26, 2018. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Presented paper/talk
Year(s) Of Engagement Activity 2018