Search

Semantic segmentation has been proposed as a tool to accelerate the processing of natural history collection images. However, developing a flexible and resilient segmentation network requires an approach for adaptation which allows processing different datasets with minimal training and validation. This paper presents a cross-validation approach designed to determine whether a semantic segmentation network possesses the flexibility required for application across different collections and institutions. Consequently, the specific objectives of cross-validating the semantic segmentation network are to (a) evaluate the effectiveness of the network for segmenting image sets derived from collections different from the one in which the network was initially trained on; and (b) test the adaptability of the segmentation network for use in other types of collections. The resilience to data variations from different institutions and the portability of the network across different types of collections are required to confirm its general applicability. The proposed validation method is tested on the Natural History Museum semantic segmentation network, designed to process entomological microscope slides. The proposed semantic segmentation network is evaluated through a series of cross-validation experiments designed to test using data from two types of collections: microscope slides (from three institutions) and herbarium sheets (from seven institutions). The main contribution of this work is the method, software and ground truth sets created for this cross-validation as they can be reused in testing similar segmentation proposals in the context of digitization of natural history collections. The cross-validation of segmentation methods should be a required step in the integration of such methods into image processing workflows for natural history collections.

This poster illustrates the adoption and customisation of a process for aligning existing data objects with the FAIR data principles (Findability, Accessibility, Interoperability, and Reuse). This is a preparatory step for creating the Catalysis Data Infrastructure (CDI), an online repository that stores persistent relationships between different types of data objects generated by UK Catalysis Hub projects. Achieving data ‘FAIRness’ eases data integration, which in turn supports cross-disciplinary sharing and use of data. In Illustrating this, we aim to also highlight in practice the benefits of the alignment to the FAIR data principles.

The need to increase global accessibility to specimens while preserving the physical specimens by reducing their handling motivates digitisation. Digitisation of natural history collections has evolved from recording of specimens’ catalogue data to including digital images and 3D models of specimens. The sheer size of the collections requires developing high throughput digitisation workflows, as well as novel acquisition systems, image standardisation, curation, preservation, and publishing. For instance, herbarium sheet digitisation workflows (and fast digitisation stations) can digitise up to 6,000 specimens per day; operating digitisation stations in parallel can increase that capacity. However, other activities of digitisation workflows still rely on manual processes which throttle the speed with which images can be published. Image quality control and information extraction from images can benefit from greater automation. This presentation explores the advantages of applying semantic segmentation (Fig. 1) to improve and automate image quality management (IQM) and information extraction from images (IEFI) of physical specimens. Two experiments were designed to determine if IQM and IEFI activities can be improved by using segments instead of full images. The time for segmenting full images needs to be considered for both IQM and IEFI. A semantic segmentation method developed by the Natural History Museum (Durrant and Livermore 2018) adapted for segmenting herbarium sheet images (Dillen et al. 2019) can process 50 images in 12 minutes. The IQM experiments evaluated the application of three quality attributes to full images and to image segments: colourfulness (Fig. 2), contrast (Fig. 3) and sharpness (Fig. 4). Evaluating colourfulness is an alternative to colour quantization algorithms such as RMSE and Delta E (Hasler and Suesstrunk 2003, Palus 2006), the method produces a value indicating if the image degrades after processing. Contrast measures the difference in luminance or colour that makes an object distinguishable. Contrast is determined by the difference in colour and brightness of the object and other objects within the same field of view (Matkovic et al. 2005, Präkel 2010). Sharpness encompasses the concepts of resolution and acutance (Bahrami and Kot 2014, Präkel 2010). Sharpness influences specimen appearance and readability of information from labels and barcodes. Evaluating the criteria on 56 barcodes and 50 colour charts segments extracted from fifty images took 34 minutes (8 minutes for the barcodes and 26 minutes for colour charts). The evaluation on the corresponding full images took 100 minutes. The processing of individual segments and full images provided results equivalent to subjective manual quality management. The IEFI experiments compared the performance of four optical character recognition (OCR) programs applied to full images (Drinkwater et al. 2014) against individual segments. The four OCR programs evaluated were Tesseract 4.X, Tesseract 3.X, Abby FineReader Engine 12, and Microsoft OneNote 2013. The test was based on a set of 250 herbarium sheet images and 1,837 segments extracted from them. The results from the experiments show that there is an average OCR speed-up of 49% when using segmented images when compared to processing times for full images (Table 1). Similarly, there was an average increase of 13% in line correctness (information from lines is ordered and not fragmented (Fig. 5, Table 2 ). Additionally, the results are useful for comparing the four OCR programs, with Tesseract 3.x offering shortest processing time, while Tesseract 4.X achieving the highest scores for line accuracy (including hand written text recognition). The results suggest that IEFI could be improved by performing OCR using segments rather than whole images, leading to faster processing and more accurate outputs. The findings support the feasibility of further automation of digitisation workflows for natural history collections. In addition to increasing the accuracy and speed of IQM and IEFI activities, the explored approaches can be packaged and published, enabling automated quality management and information extraction to be offered as a service, taking advantage of cloud platforms and workflow engines.

Digitisation of Natural History Collections (NHC) has evolved from transcription of specimen catalogues in databases to web portals providing access to data, digital images, and 3D models of specimens. These portals increase global accessibility to specimens and help preserve the physical specimens by reducing their handling. The size of the NHC requires developing high-throughput digitisation workflows, as well as research into novel acquisition systems, image standardisation, curation, preservation, and publishing. Nowadays, herbarium sheet digitisation workflows (and fast digitisation stations) can digitise up to 6,000 specimens per day. Operating those digitisation stations in parallel, can increase the digitisation capacity. The high-resolution images obtained from these specimens, and their volume require substantial bandwidth, and disk space and tapes for storage of original digitised materials, as well as availability of computational processing resources for generating derivatives, information extraction, and publishing. While large institutions have dedicated digitisation teams that manage the whole workflow from acquisition to publishing, other institutions cannot dedicate resources to support all digitisation activities, in particular long-term storage. National and European e-infrastructures can provide an alternative solution by supporting different parts of the digitisation workflows. In the context of the Innovation and consolidation for large scale digitisation of natural heritage (ICEDIG Project 2018), three different e-infrastructures providing long-term storage have been analysed through three pilot studies: EUDAT-CINES, Zenodo, and National Infrastructures. The EUDAT-CINES pilot centred on transferring large digitised herbarium collections from the National Museum of Natural History France (MNHN) to the storage infrastructure provided by the Centre Informatique National de l’Enseignement Supérieur (CINES 2014), a European trusted digital repository. The upload, processing, and access services are supported by a combination of services provided by the European Collaborative Data Infrastructure (EUDAT CDI 2019) and CINES . The Zenodo pilot included the upload of herbarium collections from Meise Botanic Garden (APM) and other European herbaria into the Zenodo repository (Zenodo 2019). The upload, processing and access services are supported by Zenodo services, accessed by APM. The National Infrastructures pilot facilitated the upload of digital assets derived from specimens of herbarium and entomology collections held at the Finnish Museum of Natural History (LUOMUS) into the Finnish Biodiversity Information Facility (FinBIF 2019). This pilot concentrates on simplifying the integration of digitisation facilities to Finnish national e-infrastructures, using services developed by LUOMUS to access FinBIF resources. The data models employed in the pilots allow defining data schemas according to the types of collection and specimen images stored. For EUDAT-CINES, data were composed of the specimen data and its business metadata (those the institution making the deposit, in this case MNHN, considers relevant for the data objects being stored), enhanced by archiving metadata, added during the archiving process (institution, licensing, identifiers, project, archiving date, etc). EUDAT uses ePIC identifiers (ePIC 2019) to identify each deposit. The Zenodo pilot was designed to allow defining specimen data and metadata supporting indexing and access to resources. Zenodo uses DataCite Digital Object Identifiers (DOI) and the underlying data types as the main identifiers for the resources, augmented with fields based on standard TDWG vocabularies. FinBIF compiles Finnish biodiversity information to one single service for open access sharing. In FinBIF, HTTP URI based identifiers are used for all data, which link the specimen data with other information, such as images. The pilot infrastructure design reports describe features, capacities, functions and costs for each model, in three specific contexts are relevant for the implementation of the Distributed Systems of Scientific Collections (DiSSCo 2019) research infrastructure, informing the options for long-term storage and archiving digitised specimen data. The explored options allow preservation of assets and support easy access. In a wider context, the results provide a template for service evaluation in the European Open Science Cloud (EOSC 2019) which can guide similar efforts.

There are many software tools designed to support research activities such as data analysis and in silico experiments in the biotechnology field. This paper presents an adaptation of BioVel portal called "Laboratorio Virtual de Biotecnologia" (LVB), that can be used to introduce students to scientific workflows. The paper reports the preliminary results of a usability case study carried out on the Spanish version of this academic resource. This case study has three main stages: introduction, thinking aloud and closing. The paper suggests a set of services and functional requirements for the deployment of LVB in order to provide relevant services for biotechnology students.

This paper presents SCRAM---CK, a method to elicit requirements by means of strong user involvement supported by prototyping activities. The method integrates two existing approaches, SCRAM and CK theory. SCRAM provides the framework for requirements management, while CK theory provides a framework for reasoning about design and its evolution. The method is demonstrated with the definition and refining of requirements for the BioVeL web toolkit. The objective of BioVeL is to allow scientists to understand, run, modify and construct workflows for data analysis with minimal training using a web-based interface. The proposed method is supported by prototyping activities for gathering user feedback, and refining requirements and design proposals. Using this method, the prototypes evolved from simple workflow execution enablers to include more complex functionalities for reviewing, modifying and building workflows in later versions. This paper presents a contribution to the application of techniques for requirements engineering. SCRAM---CK is an amalgamated method that combines a user-centred continuous refinement approach with support for design evolution through prototyping. The paper also shows the influence of the requirements engineering process in the evolution of design proposals.

The Taverna workflow tool suite (http://www.taverna.org.uk) is designed to combine distributed Web Services and/or local tools into complex analysis pipelines. These pipelines can be executed on local desktop machines or through larger infrastructure (such as supercomputers, Grids, or cloud environments), using the Taverna Server. In bioinformatics, Taverna workflows are typically used in the areas of high-throughput omics analyses (for example proteomics, or transcriptomics), or for evidence gathering methods involving text mining or data mining. Through Taverna, scientists have access to several thousand different tools and resources that are freely available from a large range of Life Science institutions. Once constructed, the workflows are reusable, executable bioinformatics protocols that can be shared, reused and repurposed. A repository of public workflows is available at http://www.myexperiment.org. This article provides an update to the Taverna tool suite, highlighting new features and developments since those reported in a previous Web Server issue of this journal.

Small and medium enterprises (SMEs) typically rely on emails to interact with other SMEs in a collaborative network. However, with the use of different document standards and interpretation techniques establishing a shared understanding of business documents (or emails) can be difficult and raise interoperability issues. In this paper we introduce Commius -- a zero cost enterprise system that aims to achieve seamless interoperability among collaborative SMEs. At the heart of Commius is a semantic alignment mechanism capable of automatically interpreting a document received by the system. However, when the automatic interpretation fails, Commius employs agent-inspired negotiation techniques to help achieve common understanding of the document with other Commius system (installed at sender or collaborative SME). This paper presents details of Commius's email negotiation mechanism with the help of a running scenario that demonstrates how Commius-based negotiation can support SMEs achieve semantic alignment in collaborative networks.

The definition of a digital specimen is proposed to encompass the digital representation(s) of physical specimens from natural science collections. The digital specimen concept is intended to define a representation (digital object) that brings together an array of heterogeneous data types, which are themselves alternative physical specimen representations. In this case, the digital specimen (DS) holds references to specimen data from a collection management system, images, 3D models, research articles, DNA sequences, collector information, among many other data types. The proposal is to create persistent relationships between the DS and other categories of digital objects (e.g. resource types mentioned above, collections, storage platforms, organisations, databases, and provenance data). Complying with FAIR data principles (findability, accessibility, interoperability, and reuse), i.e., achieving data ‘FAIRness’, eases data integration, which is needed for cross-disciplinary linking and combination of data from different domains, making the DS as a comprehensive package of information about a specimen. Implementation and access to a digital specimen repository (DSR) as a Digital Object Architecture (Sharp 2016) component demonstrates the alignment of the DS concept and FAIR data principles (Wilkinson et al. 2016, Kahn and Wilensky 2006). The DSR fulfills four roles: data producer, resource manager, data publisher, and collaboration space. As data producer, the DSR allows acquisition and curation (indexing, storage) of DSs linking primary data, models, analyses, and other digital object types. As resource manager, the DSR manages access to distributed platforms, ranging from acquisition networks (digitisation stations, museums, herbariums) to processing services, advanced computational resources, data asset storage systems, and specialised servers. As data publisher, the DSR provides access to data assets from national and transnational data archives. As collaboration space, the DSR supports users’ accessing, sharing and (re)using data assets, and derived data products and services. Adopting the collaboration space and data publisher roles, the DSR implements interfaces that expose the DSs to the research community, fulfilling the FAIR findability, accessibility, and reuse principles. Adopting the data producer and resource manager roles, the DSR creates meaningful and persistent relationships required to link DSs and other types of digital objects, fulfilling the FAIR interoperability principle. A prototype DSR based on the Cordra digital object repository has been deployed (Corporation for National Research Initiatives (CNRI) 2018, Reilly and Tupelo-Schneck 2010). The advantages of Cordra are: rapid deployment, customisable object model, creation of relations between digital objects, and application program interfaces for programmatic access. Rapid deployment of the DSR provides a tangible target for discussing the implementation of the DS concept. The customisable object model enables the refinement and enhancing of the definition of DS in response to feedback from colleagues who have accessed the DSR and used its contents. Creating relations between digital objects enables flexible linking to digital objects stored in different repositories. Accessing the DSR programmatically through APIs enables extending the use of the repository in different platforms (e.g. mobile devices) as well as integration with other repositories and services. As well as supporting a HTTP-oriented API, Cordra implements Digital Object Interface Protocol (DONA Foundation 2018), allowing the definition of operations to act directly on selected DSs in the repository. The DSR prototype has been demonstrated by providing access to the repository administrative interface and with a custom interface designed to facilitate access by different user groups, such as collection curators, researchers, teachers, and students. The client interface has been designed to demonstrate a subset of the functionalities derived from user stories, which describe software features from the end-user perspective. Demonstrating the DSR capabilities as proposed, will inform the refinement of the design of the DS model and provide early feedback about the needed software features.