skip to main content


The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Thursday, May 23 until 2:00 AM ET on Friday, May 24 due to maintenance. We apologize for the inconvenience.

Title: Reliable Dataset Identifiers Are Essential Building Blocks For Reproducible Research
10.17605/OSF.IO/AT4XE Despite increased use of digital biodiversity data in research, reliable methods to identify datasets are not widely adopted. While commonly used location-based dataset identifiers such as URLs help to easily download data today, additional identification schemes are needed to ensure long term access to datasets. We propose to augment existing location- and DOI-based identification schemes with cryptographic content-based identifiers. These content-based identifiers can be calculated from the datasets themselves using available cryptographic hashing algorithms (e.g., sha256). These algorithms take only the digital content as input to generate a unique identifier without needing a centralized identification administration. The use of content-based identifiers is not new, but a re-application of change management techniques used in the popular version control system "git". We show how content-based identifiers can be used to version datasets, to track the dataset locations, to monitor their reliability, and to efficiently detect dataset changes. We discuss the results of using our approach on datasets registered in GBIF and iDigBio from Sept 2018 to May 2020. Also, we propose how reliable, decentralized, dataset indexing and archiving systems can be devised. Lastly, we outline a modification to existing data citation practices to help work towards more reproducible and reusable research workflows.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
4th Annual Digital Data in Biodiversity Research, 1-3 June 2020
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    No systematic approach has yet been adopted to reliably reference and provide access to digital biodiversity datasets. Based on accumulated evidence, we argue that location-based identifiers such as URLs are not sufficient to ensure long-term data access. We introduce a method that uses dedicated data observatories to evaluate long-term URL reliability. From March 2019 through May 2020, we took periodic inventories of the data provided to major biodiversity aggregators, including GBIF, iDigBio, DataONE, and BHL by accessing the URL-based dataset references from which the aggregators retrieve data. Over the period of observation, we found that, for the URL-based dataset references available in each of the aggregators' data provider registries, 5% to 70% of URLs were intermittently or consistently unresponsive, 0% to 66% produced unstable content, and 20% to 75% became either unresponsive or unstable. We propose the use of cryptographic hashing to generate content-based identifiers that can reliably reference datasets. We show that content-based identifiers facilitate decentralized archival and reliable distribution of biodiversity datasets to enable long-term accessibility of the referenced datasets. 
    more » « less
  2. Obeid, I. (Ed.)
    The Neural Engineering Data Consortium (NEDC) is developing the Temple University Digital Pathology Corpus (TUDP), an open source database of high-resolution images from scanned pathology samples [1], as part of its National Science Foundation-funded Major Research Instrumentation grant titled “MRI: High Performance Digital Pathology Using Big Data and Machine Learning” [2]. The long-term goal of this project is to release one million images. We have currently scanned over 100,000 images and are in the process of annotating breast tissue data for our first official corpus release, v1.0.0. This release contains 3,505 annotated images of breast tissue including 74 patients with cancerous diagnoses (out of a total of 296 patients). In this poster, we will present an analysis of this corpus and discuss the challenges we have faced in efficiently producing high quality annotations of breast tissue. It is well known that state of the art algorithms in machine learning require vast amounts of data. Fields such as speech recognition [3], image recognition [4] and text processing [5] are able to deliver impressive performance with complex deep learning models because they have developed large corpora to support training of extremely high-dimensional models (e.g., billions of parameters). Other fields that do not have access to such data resources must rely on techniques in which existing models can be adapted to new datasets [6]. A preliminary version of this breast corpus release was tested in a pilot study using a baseline machine learning system, ResNet18 [7], that leverages several open-source Python tools. The pilot corpus was divided into three sets: train, development, and evaluation. Portions of these slides were manually annotated [1] using the nine labels in Table 1 [8] to identify five to ten examples of pathological features on each slide. Not every pathological feature is annotated, meaning excluded areas can include focuses particular to these labels that are not used for training. A summary of the number of patches within each label is given in Table 2. To maintain a balanced training set, 1,000 patches of each label were used to train the machine learning model. Throughout all sets, only annotated patches were involved in model development. The performance of this model in identifying all the patches in the evaluation set can be seen in the confusion matrix of classification accuracy in Table 3. The highest performing labels were background, 97% correct identification, and artifact, 76% correct identification. A correlation exists between labels with more than 6,000 development patches and accurate performance on the evaluation set. Additionally, these results indicated a need to further refine the annotation of invasive ductal carcinoma (“indc”), inflammation (“infl”), nonneoplastic features (“nneo”), normal (“norm”) and suspicious (“susp”). This pilot experiment motivated changes to the corpus that will be discussed in detail in this poster presentation. To increase the accuracy of the machine learning model, we modified how we addressed underperforming labels. One common source of error arose with how non-background labels were converted into patches. Large areas of background within other labels were isolated within a patch resulting in connective tissue misrepresenting a non-background label. In response, the annotation overlay margins were revised to exclude benign connective tissue in non-background labels. Corresponding patient reports and supporting immunohistochemical stains further guided annotation reviews. The microscopic diagnoses given by the primary pathologist in these reports detail the pathological findings within each tissue site, but not within each specific slide. The microscopic diagnoses informed revisions specifically targeting annotated regions classified as cancerous, ensuring that the labels “indc” and “dcis” were used only in situations where a micropathologist diagnosed it as such. Further differentiation of cancerous and precancerous labels, as well as the location of their focus on a slide, could be accomplished with supplemental immunohistochemically (IHC) stained slides. When distinguishing whether a focus is a nonneoplastic feature versus a cancerous growth, pathologists employ antigen targeting stains to the tissue in question to confirm the diagnosis. For example, a nonneoplastic feature of usual ductal hyperplasia will display diffuse staining for cytokeratin 5 (CK5) and no diffuse staining for estrogen receptor (ER), while a cancerous growth of ductal carcinoma in situ will have negative or focally positive staining for CK5 and diffuse staining for ER [9]. Many tissue samples contain cancerous and non-cancerous features with morphological overlaps that cause variability between annotators. The informative fields IHC slides provide could play an integral role in machine model pathology diagnostics. Following the revisions made on all the annotations, a second experiment was run using ResNet18. Compared to the pilot study, an increase of model prediction accuracy was seen for the labels indc, infl, nneo, norm, and null. This increase is correlated with an increase in annotated area and annotation accuracy. Model performance in identifying the suspicious label decreased by 25% due to the decrease of 57% in the total annotated area described by this label. A summary of the model performance is given in Table 4, which shows the new prediction accuracy and the absolute change in error rate compared to Table 3. The breast tissue subset we are developing includes 3,505 annotated breast pathology slides from 296 patients. The average size of a scanned SVS file is 363 MB. The annotations are stored in an XML format. A CSV version of the annotation file is also available which provides a flat, or simple, annotation that is easy for machine learning researchers to access and interface to their systems. Each patient is identified by an anonymized medical reference number. Within each patient’s directory, one or more sessions are identified, also anonymized to the first of the month in which the sample was taken. These sessions are broken into groupings of tissue taken on that date (in this case, breast tissue). A deidentified patient report stored as a flat text file is also available. Within these slides there are a total of 16,971 total annotated regions with an average of 4.84 annotations per slide. Among those annotations, 8,035 are non-cancerous (normal, background, null, and artifact,) 6,222 are carcinogenic signs (inflammation, nonneoplastic and suspicious,) and 2,714 are cancerous labels (ductal carcinoma in situ and invasive ductal carcinoma in situ.) The individual patients are split up into three sets: train, development, and evaluation. Of the 74 cancerous patients, 20 were allotted for both the development and evaluation sets, while the remain 34 were allotted for train. The remaining 222 patients were split up to preserve the overall distribution of labels within the corpus. This was done in hope of creating control sets for comparable studies. Overall, the development and evaluation sets each have 80 patients, while the training set has 136 patients. In a related component of this project, slides from the Fox Chase Cancer Center (FCCC) Biosample Repository ( -facility) are being digitized in addition to slides provided by Temple University Hospital. This data includes 18 different types of tissue including approximately 38.5% urinary tissue and 16.5% gynecological tissue. These slides and the metadata provided with them are already anonymized and include diagnoses in a spreadsheet with sample and patient ID. We plan to release over 13,000 unannotated slides from the FCCC Corpus simultaneously with v1.0.0 of TUDP. Details of this release will also be discussed in this poster. Few digitally annotated databases of pathology samples like TUDP exist due to the extensive data collection and processing required. The breast corpus subset should be released by November 2021. By December 2021 we should also release the unannotated FCCC data. We are currently annotating urinary tract data as well. We expect to release about 5,600 processed TUH slides in this subset. We have an additional 53,000 unprocessed TUH slides digitized. Corpora of this size will stimulate the development of a new generation of deep learning technology. In clinical settings where resources are limited, an assistive diagnoses model could support pathologists’ workload and even help prioritize suspected cancerous cases. ACKNOWLEDGMENTS This material is supported by the National Science Foundation under grants nos. CNS-1726188 and 1925494. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. REFERENCES [1] N. Shawki et al., “The Temple University Digital Pathology Corpus,” in Signal Processing in Medicine and Biology: Emerging Trends in Research and Applications, 1st ed., I. Obeid, I. Selesnick, and J. Picone, Eds. New York City, New York, USA: Springer, 2020, pp. 67 104. [2] J. Picone, T. Farkas, I. Obeid, and Y. Persidsky, “MRI: High Performance Digital Pathology Using Big Data and Machine Learning.” Major Research Instrumentation (MRI), Division of Computer and Network Systems, Award No. 1726188, January 1, 2018 – December 31, 2021. https://www. [3] A. Gulati et al., “Conformer: Convolution-augmented Transformer for Speech Recognition,” in Proceedings of the Annual Conference of the International Speech Communication Association (INTERSPEECH), 2020, pp. 5036-5040. [4] C.-J. Wu et al., “Machine Learning at Facebook: Understanding Inference at the Edge,” in Proceedings of the IEEE International Symposium on High Performance Computer Architecture (HPCA), 2019, pp. 331–344. [5] I. Caswell and B. Liang, “Recent Advances in Google Translate,” Google AI Blog: The latest from Google Research, 2020. [Online]. Available: [Accessed: 01-Aug-2021]. [6] V. Khalkhali, N. Shawki, V. Shah, M. Golmohammadi, I. Obeid, and J. Picone, “Low Latency Real-Time Seizure Detection Using Transfer Deep Learning,” in Proceedings of the IEEE Signal Processing in Medicine and Biology Symposium (SPMB), 2021, pp. 1 7. https://www.isip. [7] J. Picone, T. Farkas, I. Obeid, and Y. Persidsky, “MRI: High Performance Digital Pathology Using Big Data and Machine Learning,” Philadelphia, Pennsylvania, USA, 2020. [8] I. Hunt, S. Husain, J. Simons, I. Obeid, and J. Picone, “Recent Advances in the Temple University Digital Pathology Corpus,” in Proceedings of the IEEE Signal Processing in Medicine and Biology Symposium (SPMB), 2019, pp. 1–4. [9] A. P. Martinez, C. Cohen, K. Z. Hanley, and X. (Bill) Li, “Estrogen Receptor and Cytokeratin 5 Are Reliable Markers to Separate Usual Ductal Hyperplasia From Atypical Ductal Hyperplasia and Low-Grade Ductal Carcinoma In Situ,” Arch. Pathol. Lab. Med., vol. 140, no. 7, pp. 686–689, Apr. 2016. 
    more » « less
  3. Abstract Commonly used data citation practices rely on unverifiable retrieval methods which are susceptible to content drift, which occurs when the data associated with an identifier have been allowed to change. Based on our earlier work on reliable dataset identifiers, we propose signed citations, i.e., customary data citations extended to also include a standards-based, verifiable, unique, and fixed-length digital content signature. We show that content signatures enable independent verification of the cited content and can improve the persistence of the citation. Because content signatures are location- and storage-medium-agnostic, cited data can be copied to new locations to ensure their persistence across current and future storage media and data networks. As a result, content signatures can be leveraged to help scalably store, locate, access, and independently verify content across new and existing data infrastructures. Content signatures can also be embedded inside content to create robust, distributed knowledge graphs that can be cited using a single signed citation. We describe applications of signed citations to solve real-world data collection, identification, and citation challenges. 
    more » « less
  4. Abstract

    Insect populations are changing rapidly, and monitoring these changes is essential for understanding the causes and consequences of such shifts. However, large‐scale insect identification projects are time‐consuming and expensive when done solely by human identifiers. Machine learning offers a possible solution to help collect insect data quickly and efficiently.

    Here, we outline a methodology for training classification models to identify pitfall trap‐collected insects from image data and then apply the method to identify ground beetles (Carabidae). All beetles were collected by the National Ecological Observatory Network (NEON), a continental scale ecological monitoring project with sites across the United States. We describe the procedures for image collection, image data extraction, data preparation, and model training, and compare the performance of five machine learning algorithms and two classification methods (hierarchical vs. single‐level) identifying ground beetles from the species to subfamily level. All models were trained using pre‐extracted feature vectors, not raw image data. Our methodology allows for data to be extracted from multiple individuals within the same image thus enhancing time efficiency, utilizes relatively simple models that allow for direct assessment of model performance, and can be performed on relatively small datasets.

    The best performing algorithm, linear discriminant analysis (LDA), reached an accuracy of 84.6% at the species level when naively identifying species, which was further increased to >95% when classifications were limited by known local species pools. Model performance was negatively correlated with taxonomic specificity, with the LDA model reaching an accuracy of ~99% at the subfamily level. When classifying carabid species not included in the training dataset at higher taxonomic levels species, the models performed significantly better than if classifications were made randomly. We also observed greater performance when classifications were made using the hierarchical classification method compared to the single‐level classification method at higher taxonomic levels.

    The general methodology outlined here serves as a proof‐of‐concept for classifying pitfall trap‐collected organisms using machine learning algorithms, and the image data extraction methodology may be used for nonmachine learning uses. We propose that integration of machine learning in large‐scale identification pipelines will increase efficiency and lead to a greater flow of insect macroecological data, with the potential to be expanded for use with other noninsect taxa.

    more » « less
  5. The Global Biodiversity Information Facility (GBIF 2022a) has indexed more than 2 billion occurrence records from 70,147 datasets. These datasets often include "hidden" biotic interaction data because biodiversity communities use the Darwin Core standard (DwC, Wieczorek et al. 2012) in different ways to document biotic interactions. In this study, we extracted biotic interactions from GBIF data using an approach similar to that employed in the Global Biotic Interactions (GloBI; Poelen et al. 2014) and summarized the results. Here we aim to present an estimation of the interaction data available in GBIF, showing that biotic interaction claims can be automatically found and extracted from GBIF. Our results suggest that much can be gained by an increased focus on development of tools that help to index and curate biotic interaction data in existing datasets. Combined with data standardization and best practices for sharing biotic interactions, such as the initiative on plant-pollinators interaction (Salim 2022), this approach can rapidly contribute to and meet open data principles (Wilkinson 2016). We used Preston (Elliott et al. 2020), open-source software that versions biodiversity datasets, to copy all GBIF-indexed datasets. The biodiversity data graph version (Poelen 2020) of the GBIF-indexed datasets used during this study contains 58,504 datasets in Darwin Core Archive (DwC-A) format, totaling 574,715,196 records. After retrieval and verification, the datasets were processed using Elton. Elton extracts biotic interaction data and supports 20+ existing file formats, including various types of data elements in DwC records. Elton also helps align interaction claims (e.g., host of, parasite of, associated with) to the Relations Ontology (RO, Mungall 2022), making it easier to discover datasets across a heterogeneous collection of datasets. Using specific mapping between interaction claims found in the DwC records to the terms in RO*1, Elton found 30,167,984 potential records (with non-empty values for the scanned DwC terms) and 15,248,478 records with recognized interaction types. Taxonomic name validation was performed using Nomer, which maps input names to names found in a variety of taxonomic catalogs. We only considered an interaction record valid where the interaction type could be mapped to a term in RO and where Nomer found a valid name for source and target taxa. Based on the workflow described in Fig. 1, we found 7,947,822 interaction records (52% of the potential interactions). Most of them were generic interactions ( interacts_ with , 87.5%), but the remaining 12.5% (993,477 records) included host-parasite and plant-animal interactions. The majority of the interactions records found involved plants (78%), animals (14%) and fungi (6%). In conclusion, there are many biotic interactions embedded in existing datasets registered in large biodiversity data indexers and aggregators like iDigBio, GBIF, and BioCASE. We exposed these biotic interaction claims using the combined functionality of biodiversity data tools Elton (for interaction data extraction), Preston (for reliable dataset tracking) and Nomer (for taxonomic name alignment). Nonetheless, the development of new vocabularies, standards and best practice guides would facilitate aggregation of interaction data, including the diversification of the GBIF data model (GBIF 2022b) for sharing biodiversity data beyond occurrences data. That is the aim of the TDWG Interest Group on Biological Interactions Data (TDWG 2022). 
    more » « less