DOI: 10.3897/rio.12.e183261

Authors: Enriko Kroon, Ricardo Borges, Rômulo de Jesus, Stefan Kuhn

Abstract: It has been shown that it is possible to train a (simple) neural network to classify nuclear magnetic resonance spectra by a substructures either being part of the chemical structure measured or not. We now explore the interpretability of such models using techniques from explainable AI, specifically Grad-CAM. We show that those techniques do not give ideal results in the context of NMR, which would be able to identify individual peaks. On the other hand, they enable a better interpretation of the results than those metrics just based on "right or wrong". We can also confirm the result from our previous work, that the trained network performs well for pure compounds, but its generalisability to mixtures is questionable, a limitation that could only be assumed in the original study.

HTML

XML

PDF

DOI: 10.3897/rio.10.e125086

Authors: Bekir Afsar, Kyle Eyvindson, Tuomas Rossi, Martijn Versluijs, Otso Ovaskainen

Abstract: Forests are crucial in supporting biodiversity and providing ecosystem services. Understanding forest biodiversity dynamics under different management strategies and climate change scenarios is essential for effective conservation and management. This paper introduces the Forest Biodiversity Dynamics Prototype Digital Twin (pDT), integrating forest and biodiversity models to predict the effects of management options on forest ecosystems. The primary objective is to identify optimal management strategies that promote biodiversity, focusing on conservation and adaptation to different climate conditions. We start with the case of Finnish forests and bird species and plan to expand to include more European countries and a variety of species as the pDT is further developed.

HTML

XML

PDF

DOI: 10.3897/arphapreprints.e116230

Authors: Marco Visani

Abstract: This study is centered on Natural Products (NPs) - specific chemicals synthesized by living organisms. These NPs hold significant importance in various domains, notably medicine, agriculture, and ecology. A primary resource for our research is the LOTUS database, which catalogues a vast array of NPs and their occurrence. Yet, a gap exists: there are no existing model to predict the occurrence of these NPs across different species.In our initial strategy, the occurrence of natural products was viewed as a collection of observations and their associated variables. Although simple, this strategy immediately showed its limits when dealing with the complex nature of NPs. We switched to an advanced graph-based method after seeing the necessity for a more thorough strategy to accurately represent the intricate interactions governing NPs expression. When considering species in a phylogeny or molecular pathways, the graph-based method perceives data as a network of connected entities, offering a far more logical and natural way of thinking. By employing this better methodology, we have developed a more effective approach for investigating the intricate world of Natural Products. We hope that this strategy will open up new research directions and possibly result in ground-breaking NP-related findings.

HTML

XML

PDF

DOI: 10.3897/arphapreprints.e107873

Authors: Christian Ohmann, Steve Canham, Kurt Majcen, Petr Holub, Gary Saunders, Jing Tang, Tanushree Tunstall, Philip Gribbon, Reagon Karki, Mari Kleemola, Katja Moilanen, Walter Daelemans, Pieter Fivez, Daan Broeder, Franciska de Jong, Maria Panagiotopoulou

Abstract: This publication details the workplan of the Science Project (SP) “COVID-19 metadata findability and interoperability in EOSC” (short: META-COVID) that is part of the Horizon Europe funded project EOSC Future. The COVID-19 pandemic has generated a huge variety of research activities, studies, and policies across both the life sciences (LS) and the social sciences and humanities (SSH). Useful insights from combining the data and conclusions from these different forms of research are, however, hampered by the lack of a common metadata framework with which to describe them. This is because different scientific disciplines have different ways of organising research activities. For example, the type of the research (e.g., hypothesis testing versus hypothesis generating) and the methodology chosen (e.g., experimental, survey, cohort, case study) are key elements in understanding the data generated and in supporting its secondary use. Another issue to be tackled is the integration of various sources of metadata related to parliamentary and social media metadata. In META-COVID, scientists from the LS and SSH domains gathered to discuss ways in which metadata could go beyond the description of the data itself to include the basic elements of the research process (“contextual metadata”) within the frame of the European Open Science Cloud (EOSC). The main outcomes of the SP will be: i) An inventory of metadata schemas applied across infrastructures and domains; ii) The development of a framework for a metadata model characterising the research approach and workflow across research infrastructures; iii) The application of the framework to selected COVID-19 use cases; iv) The development of an ontology of COVID-19 related topics from parliamentary data and social media.

HTML

XML

PDF

DOI: 10.3897/rio.7.e70183

Authors: Nazareno Scaccia, Taras Günther, Estibaliz Lopez de Abechuco, Matthias Filter

Abstract: In many interdisciplinary research domains, the creation of a shared understanding of relevant terms is considered the foundation for efficient cross-sector communication and interpretation of data and information. This is also true for the domain of One Health (OH) where many One Health Surveillance (OHS) documents rarely contain glossaries with a list of terms for which their specific meaning in the context of the given document is defined (Cornelia et al. 2018, Buschhardt et al. 2021). The absence of glossaries within these documents may lead to misinterpretation of surveillance results due to the wrong interpretation of terminology specifically when term definitions differ across OH sectors. Under the One Health EJP project ORION, the OHEJP Glossary was recently created. The OHEJP Glossary is a tool to improve communication and collaboration amongst OH sectors by providing an easy-to-use online resource that lists relevant OH terms and sector-specific definitions. To improve the accessibility of content from the OHEJP Glossary and support the creation of integrative glossaries in future OHS-related documents, the OHEJP Glossaryfication Web Service was created. This service can support the practical use of the OHEJP Glossary and other relevant online glossaries by OH professionals.The Glossaryfication Web Service (GWS) is an application that automatically identifies terms in any uploaded text-based document and creates a document-specific list of matching definitions in selected online glossaries. This auto-generated document-specific glossary can easily be adjusted by the user, for example, by selecting the desired definition in case multiple definitions were found for a specific term. The document-specific glossary could then be downloaded, manually adjusted and finally included into the original document where it supports the correct interpretation of terminology used. Especially in sector-specific reports, such as from animal health or public health authorities, this can be beneficial to ensure the correct interpretation by other OH sectors in the future. The GWS was developed with the open-source desktop software KNIME Analytics Platform and runs as a web service on a KNIME Web Server infrastructure. The core data processing functionality in the GWS is based on KNIME’s Text Processing extension. KNIME's JavaScript nodes provided the basis for an interactive user interface where users can easily upload their files and select between different reference glossaries, such as the OHEJP Glossary, the CDC Glossary, the WHO Glossary or the EFSA Glossary. After retrieval of the user input settings, the GWS tags words within the provided document and maps these tagged words with matching entries in the selected glossaries. As the main output, the user receives a downloadable list of matching terms with their corresponding definitions, sectorial assignments and references, which can then be added by the user to the original document. The GWS is freely accessible via this link as well as the underlying KNIME workflow.

HTML

XML

PDF

DOI: 10.3897/rio.7.e68121

Authors: Daniel Mietchen, Lane Rasberry, Thais Morata, John Sadowski, Jeanette Novakovich, James Heilman

Abstract: In this era of information overload and misinformation, it is a challenge to rapidly translate evidence-based health information to the public. Wikipedia is a prominent global source of health information with high traffic, multilingual coverage, and acceptable quality control practices. Viewership data following the Ebola crisis and during the COVID-19 pandemic reveals that a significant number of web users located health guidance through Wikipedia and related projects, including its media repository Wikimedia Commons and structured data complement, Wikidata.The basic idea discussed in this paper is to increase and expedite health institutions' global reach to the general public, by developing a specific strategy to maximize the availability of focused content into Wikimedia’s public digital knowledge archives. It was conceptualized from the experiences of leading health organizations such as Cochrane, the World Health Organization (WHO) and other United Nations Organizations, Cancer Research UK, National Network of Libraries of Medicine, and Centers for Disease Control and Prevention (CDC)'s National Institute for Occupational Safety and Health (NIOSH). Each has customized strategies to integrate content in Wikipedia and evaluate responses.We propose the development of an interactive guide on the Wikipedia and Wikidata platforms to support health agencies, health professionals and communicators in quickly distributing key messages during crisis situations. The guide aims to cover basic features of Wikipedia, including adding key health messages to Wikipedia articles, citing expert sources to facilitate fact-checking, staging text for translation into multiple languages; automating metrics reporting; sharing non-text media; anticipating offline reuse of Wikipedia content in apps or virtual assistants; structuring data for querying and reuse through Wikidata, and profiling other flagship projects from major health organizations.In the first phase, we propose the development of a curriculum for the guide using information from prior case studies. In the second phase, the guide would be tested on select health-related topics as new case studies. In its third phase, the guide would be finalized and disseminated.

HTML

XML

PDF

DOI: 10.3897/rio.3.e14837

Authors: Lynn Woolfrey

Abstract: The purpose of this project is to demonstrate that tobacco-related data from selected Africa countries can be collected and distributed from an Open Data platform. The platform and data will improve the capacity for tobacco control research in key sub-Saharan African countries, and help develop a continent-wide research approach to tobacco control. 

HTML

XML

PDF

DOI: 10.3897/rio.3.e12733

Authors: Swati Rane, Eshin Jolly, Anne Park, Hojin Jang, Cameron Craddock

Abstract: We designed a modular machine learning program that uses functional magnetic resonance imaging (fMRI) data in order to distinguish individuals with autism spectrum disorders from neurodevelopmentally normal individuals. Data was selected from the Autism Brain Imaging Dataset Exchange (ABIDE) I Preprocessed Dataset.

HTML

XML

PDF

DOI: 10.3897/rio.3.e12394

Authors: Kesshi Jordan, Anisha Keshavan, Maria Luisa Mandelli, Roland Henry

Abstract: Cluster-viz is a web application that provides a platform for cluster-based interactive quality-control of tractography algorithm outputs. This tool facilitates the creation of white matter fascicle models by employing a cluster-based approach to allow the user to select streamline bundles for inclusion/exclusion in the final fascicle model. This project was started at the 2016 Neurohackweek and BrainHack events and is still under development. We welcome contributions to the Cluster-viz github repository (https://github.com/kesshijordan/Cluster-viz).

HTML

XML

PDF

DOI: 10.3897/rio.3.e12342

Authors: Julia Huntenburg, Alexandre Abraham, João Loula, Franziskus Liem, Kamalaker Dadi, Gaël Varoquaux

Abstract: Processing neuroimaging data on the cortical surface traditionally requires dedicated heavy-weight software suites. Here, we present an initial support of cortical surfaces in Python within the neuroimaging data processing toolbox Nilearn. We provide loading and plotting functions for different surface data formats with minimal dependencies, along with examples of their application. Limitations of the current implementation and potential next steps are discussed.

HTML

XML

PDF

DOI: 10.3897/rio.3.e12346

Authors: Julia Huntenburg, Konrad Wagstyl, Christopher Steele, Thomas Funck, Richard Bethlehem, Ophélie Foubet, Benoit Larrat, Victor Borrell, Pierre-Louis Bazin

Abstract: Increasingly available high-resolution brain imaging data require specialized processing tools that can leverage their anatomical detail and handle their size. Here, we present user-friendly Python tools for cortical depth resolved analysis in such data. Our implementation is based on the CBS High-Res Brain Processing framework, and aims to make high-resolution data processing tools available to the broader community.

HTML

XML

PDF

DOI: 10.3897/rio.3.e12276

Authors: Anisha Keshavan, Christopher Madan, Esha Datta, Ian McDonough

Abstract: Mindcontrol is an open-source web-based dashboard to quality control and curate neuroimaging data. At Neurohackweek 2016, a group assembled to add new features to the Mindcontrol interface. Contributors used Python, Javascript, and Git to configure Mindcontrol for the ABIDE and CoRR open datasets, and add new types of plots to the interface. All contributions are freely available online, and the code is being actively maintained at http://www.github.com/akeshavan/mindcontrol.

HTML

XML

PDF

DOI: 10.3897/rio.3.e11825

Authors: John Chodacki, Thomas Lemberger, Jennifer Lin, Maryann Martone, Daniel Mietchen, Jessica Polka, Richard Sever, Carly Strasser

Abstract:

HTML

XML

PDF

DOI: 10.3897/rio.3.e11731

Authors: Steren Chabert, Tomás Mardones, Rodrigo Riveros, Maximiliano Godoy, Alejandro Veloz, Rodrigo Salas, Pablo Cox

Abstract: Cerebral aneurysm is a cerebrovascular disorder characterized by a bulging in a weak area in the wall of an artery that supplies blood to the brain. It is relevant to understand the mechanisms leading to the apparition of aneurysms, their growth and, more important, leading to their rupture. The purpose of this study is to study the impact on aneurysm rupture of the combination of different parameters, instead of focusing on only one factor at a time as is frequently found in the literature, using machine learning and feature extraction techniques. This discussion takes relevance in the context of the complex decision that the physicians have to take to decide which therapy to apply, as each intervention bares its own risks, and implies to use a complex ensemble of resources (human resources, OR, etc.) in hospitals always under very high work load. This project has been raised in our actual working team, composed of interventional neuroradiologist, radiologic technologist, informatics engineers and biomedical engineers, from Valparaiso public Hospital, Hospital Carlos van Buren, and from Universidad de Valparaíso – Facultad de Ingeniería and Facultad de Medicina. This team has been working together in the last few years, and is now participating in the implementation of an “interdisciplinary platform for innovation in health”, as part of a bigger project leaded by Universidad de Valparaiso (PMI UVA1402). It is relevant to emphasize that this project is made feasible by the existence of this network between physicians and engineers, and by the existence of data already registered in an orderly manner, structured and recorded in digital format. The present proposal arises from the description in nowadays literature that the actual indicators, whether based on morphological description of the aneurysm, or based on characterization of biomechanical factor or others, these indicators were shown not to provide sufficient information in order to predict by themselves the risk of rupture. Therefore, our hypothesis is that the risk of rupture lies on the combination of multiple actors. These actors together would play different roles that could be: weakening of the artery wall, increasing biomechanical stresses on the wall induced by blood flow, in addition to personal sensitivity due to family history, or personal history of comorbidity, or even seasonal variations that could gate different inflammation mechanisms. The main goal of this project is to identify relevant variables that may help in the process of predicting the risk of intracranial aneurysm rupture using machine learning and image processing techniques based on structured and non-structured data from multiple sources. We believe that the identification and the combined use of relevant variables extracted from clinical, demographical, environmental and medical imaging data sources will improve the estimation of the aneurysm rupture risk, with respect to the actual practiced method based essentially on the aneurysm size. The methodology of this work consist of four phases: (1) Data collection and storage, (2) feature extraction from multiple sources in particular from angiographic images, (3) development of the model that could describe the risk of aneurysm rupture based on the fusion and combination of the features, and (4) Identification of relevant variables related to the aneurysm rupture process. This study corresponds to an analytic transversal study with prospective and retrospective characteristics. This work will be based on publicly available health statistics data, data of weather conditions, together with clinical and demographic data of patients diagnosed with intracranial aneurysm in the Hospital Carlos van Buren. As main results of this project we are expecting to identify relevant variables extracted from images and other sources that could play a role in the risk of aneurysm rupture. The proposed model will be presented to the physicians of the Hospital Carlos van Buren, to be further implemented in this Institution according to the demonstrated impact of our results. The main results will be published in indexed journals and presented at national and international conferences.

HTML

XML

PDF

DOI: 10.3897/rio.2.e9232

Authors: Pekka Neittaanmäki, Timo Huttula, Juha Karvanen, Tom Frisk, Jouni Tuomisto, Antti Simola, Tero Tuovinen, Janne Ropponen

Abstract: Open data and models are becoming increasingly available, but there are not yet good methods and platforms to turn those into systematic evidence-based decision support. Unicorn will produce such an enviro­­nment based on existing theoretical and practical knowledge about decision support and models. This con­sortium possesses the necessary models, data, and skills to set up an environment and demonstrate its func­tionality and usefulness with several case studies related to the environmental issues, human health, and economy. The Unicorn environment will be built in a generic and systematic way so that it could even be­come an international standard for evidence-based decision support. Developing a technical environment or standard is not enough. Using the Unicorn environment is a large cul­­tural change for both researchers and decision makers, as the current decision support practices do not re­flect the principles of openness, criticism, or reuse. Therefore, this cultural change must be promoted by train­ing to use the environment, by informing the society about its possibilities, and solving a number of practi­cal and technical problems related to current practices in research institutes, ministries, and municipalities. We acknowledge these problems and offer solutions to them with an extensive interaction plan.

HTML

XML

PDF

DOI: 10.3897/rio.2.e9113

Authors: Katja Heuer, Satrajit Ghosh, Amy Robinson Sterling, Roberto Toro

Abstract:

HTML

XML

PDF

DOI: 10.3897/rio.2.e8848

Authors: Arno Klein

Abstract: Mobile phones provide a new way of collecting behavioral medical research data at a scale never before possible – Sage Bionetworks’ mPower Parkinson research app, launched at Apple’s March 9, 2015 ResearchKit announcement, is currently collecting data related to Parkinson symptoms, such as voice recordings, from thousands of registered study participants. Before making such voice data available to any qualified researcher in the world, they need to undergo quality control and editing, which is currently something only a human can do well. To achieve this goal and the required scale, we will crowdsource these tasks through Amazon's Mechanical Turk.

HTML

XML

PDF

DOI: 10.3897/rio.2.e8835

Authors: Arno Klein, Satrajit Ghosh

Abstract: The proposed research develops new computational tools to identify, diagnose, and predict treatment outcome for different mental illnesses. The research will be applied first to major depressive disorder, which affects millions of Americans, but is intended to be applied to any mental illness, such as Alzheimer’s disease, bipolar disorder, schizophrenia – indeed to analyze differences in brain structure, activity, or connectivity between any two populations.

HTML

XML

PDF

DOI: 10.3897/rio.2.e8817

Authors: Arno Klein

Abstract: We will analyze variations in brain anatomy and create the first integrated software environment to extract patterns from brains and target differences related to inter-individual variability, pathology, development, or degeneration. We will evaluate how well these differences can help diagnose and predict treatment outcome for major depressive disorder, which affects millions of Americans, but our work is intended to be applied to any mental illness, such as Alzheimer’s disease, bipolar disorder, schizophrenia – indeed to analyze differences in brain anatomy between any two populations.

HTML

XML

PDF

DOI: 10.3897/rio.2.e8816

Authors: Arno Klein

Abstract: The BigBrain, a high-resolution 3-D model of a human brain at nearly cellular resolution, is the best brain imaging data set in the world to establish a canonical space at both microscopic and macroscopic resolutions. However, for the cell-stained microstructural data to be truly useful, it needs to be segmented into cytoarchitectonic regions, a challenge no single lab could undertake. The principal aim of this proposal is to crowdsource the segmentation of cytoarchitectonic regions by means of a computer game, to transform an arduous, isolated task performed by experts into an engaging, collective activity of non-experts.

HTML

XML

PDF

DOI: 10.3897/rio.2.e8813

Authors: Arno Klein

Abstract:

HTML

XML

PDF

DOI: 10.3897/rio.2.e8815

Authors: Arno Klein, Steven Ellis

Abstract: The proposed research develops new computational tools to identify, diagnose, and predict treatment response for different mental illnesses. The research will first be applied to publicly available resting state fMRI BOLD data from patients with attentiondeficit hyperactivity disorder and autism. It will also be applied to existing clinical and biological data concerning suicidality in the context of major depressive disorder. These disorders affect millions of Americans, but these tools can be applied to any mental illness, such as Alzheimer’s disease, bipolar disorder, schizophrenia – indeed to analyze differences in brain, clinical, and biological data between any two populations.

HTML

XML

PDF