| 1. |
- Akbar, Muhammad Usman, 1990-, et al.
(författare)
-
Brain tumor segmentation using synthetic MR images - A comparison of GANs and diffusion models
- 2024
-
Ingår i: Scientific Data. - : Nature Publishing Group. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Large annotated datasets are required for training deep learning models, but in medical imaging data sharing is often complicated due to ethics, anonymization and data protection legislation. Generative AI models, such as generative adversarial networks (GANs) and diffusion models, can today produce very realistic synthetic images, and can potentially facilitate data sharing. However, in order to share synthetic medical images it must first be demonstrated that they can be used for training different networks with acceptable performance. Here, we therefore comprehensively evaluate four GANs (progressive GAN, StyleGAN 1–3) and a diffusion model for the task of brain tumor segmentation (using two segmentation networks, U-Net and a Swin transformer). Our results show that segmentation networks trained on synthetic images reach Dice scores that are 80%–90% of Dice scores when training with real images, but that memorization of the training images can be a problem for diffusion models if the original dataset is too small. Our conclusion is that sharing synthetic medical images is a viable option to sharing real images, but that further work is required. The trained generative models and the generated synthetic images are shared on AIDA data hub.
|
|
| 2. |
- Ammar, Yosr, et al.
(författare)
-
Long-term dataset for contaminants in fish, mussels, and bird eggs from the Baltic Sea
- 2024
-
Ingår i: Scientific Data. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Widespread persistent contaminants are a global environmental problem. In the Baltic Sea, wildlife contamination was first noticed in the 1960s, prompting the Swedish Environmental Protection Agency to establish a comprehensive Swedish National Monitoring Programme for Contaminants in Marine Biota (MCoM) in 1978 run by the Swedish Museum of Natural History. Eight species have been analysed, four fish species (Atlantic herring, Atlantic cod, European perch, viviparous eelpout), one bivalve species (blue mussel), and egg from three bird species (common guillemot, common tern, Eurasian oystercatcher). Here, we present a dataset containing MCoM data from its start until 2021. It includes 36 sets of time-series, each analysed for more than 100 contaminants. The longest time-series is for common guillemot and starts in 1968. We describe the structure of MCoM including historic changes to the number of stations, sample treatment, analytical methods, instruments, and laboratories. The MCoM data is available at the Bolin Centre repository and on GitHub through our R package mcomDb. The latter will be updated yearly with new MCoM records.
|
|
| 3. |
- Ammar, Yosr, et al.
(författare)
-
Long-term dataset for contaminants in fish, mussels, and bird eggs from the Baltic Sea
- 2024
-
Ingår i: Scientific Data. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Widespread persistent contaminants are a global environmental problem. In the Baltic Sea, wildlife contamination was first noticed in the 1960s, prompting the Swedish Environmental Protection Agency to establish a comprehensive Swedish National Monitoring Programme for Contaminants in Marine Biota (MCoM) in 1978 run by the Swedish Museum of Natural History. Eight species have been analysed, four fish species (Atlantic herring, Atlantic cod, European perch, viviparous eelpout), one bivalve species (blue mussel), and egg from three bird species (common guillemot, common tern, Eurasian oystercatcher). Here, we present a dataset containing MCoM data from its start until 2021. It includes 36 sets of time-series, each analysed for more than 100 contaminants. The longest time-series is for common guillemot and starts in 1968. We describe the structure of MCoM including historic changes to the number of stations, sample treatment, analytical methods, instruments, and laboratories. The MCoM data is available at the Bolin Centre repository and on GitHub through our R package mcomDb. The latter will be updated yearly with new MCoM records.
|
|
| 4. |
- Berner, Logan T., et al.
(författare)
-
The Arctic plant aboveground biomass synthesis dataset
- 2024
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Plant biomass is a fundamental ecosystem attribute that is sensitive to rapid climatic changes occurring in the Arctic. Nevertheless, measuring plant biomass in the Arctic is logistically challenging and resource intensive. Lack of accessible field data hinders efforts to understand the amount, composition, distribution, and changes in plant biomass in these northern ecosystems. Here, we present The Arctic plant aboveground biomass synthesis dataset, which includes field measurements of lichen, bryophyte, herb, shrub, and/or tree aboveground biomass (g m−2) on 2,327 sample plots from 636 field sites in seven countries. We created the synthesis dataset by assembling and harmonizing 32 individual datasets. Aboveground biomass was primarily quantified by harvesting sample plots during mid- to late-summer, though tree and often tall shrub biomass were quantified using surveys and allometric models. Each biomass measurement is associated with metadata including sample date, location, method, data source, and other information. This unique dataset can be leveraged to monitor, map, and model plant biomass across the rapidly warming Arctic.
|
|
| 5. |
- Clements, Harley S., et al.
(författare)
-
The bii4africa dataset of faunal and floral population intactness estimates across Africa's major land uses
- 2024
-
Ingår i: Scientific data. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Sub-Saharan Africa is under-represented in global biodiversity datasets, particularly regarding the impact of land use on species' population abundances. Drawing on recent advances in expert elicitation to ensure data consistency, 200 experts were convened using a modified-Delphi process to estimate 'intactness scores': the remaining proportion of an 'intact' reference population of a species group in a particular land use, on a scale from 0 (no remaining individuals) to 1 (same abundance as the reference) and, in rare cases, to 2 (populations that thrive in human-modified landscapes). The resulting bii4africa dataset contains intactness scores representing terrestrial vertebrates (tetrapods: ±5,400 amphibians, reptiles, birds, mammals) and vascular plants (±45,000 forbs, graminoids, trees, shrubs) in sub-Saharan Africa across the region's major land uses (urban, cropland, rangeland, plantation, protected, etc.) and intensities (e.g., large-scale vs smallholder cropland). This dataset was co-produced as part of the Biodiversity Intactness Index for Africa Project. Additional uses include assessing ecosystem condition; rectifying geographic/taxonomic biases in global biodiversity indicators and maps; and informing the Red List of Ecosystems.
|
|
| 6. |
- Dallaire-Fortier, Clara
(författare)
-
A comprehensive historical and geolocalized database of mining activities in Canada
- 2024
-
Ingår i: Scientific Data. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- This paper introduces the MinCan database that presents mine-level estimates for the Canadian mining industry with a persistent annual coverage between 1950 and 2022. These estimates are based on archival maps and a selection of historical sources, which follows a hierarchy of criteria-based credibility and standardization. The information contained in MinCan covers 947 mines and provides information about their location (longitude and latitude in decimal), the company ownership, the principal commodities produced, and the years of operation (opening and closing dates). It is the first open access database to propose an exhaustive, free, and reliable compilation of the principal past and present mines producing in Canada. The geographic coordinates enable matching with other local, regional, and national databases, and allow for a wide range of research objectives to be met.
|
|
| 7. |
- Fridman, Jonas, et al.
(författare)
-
Harmonised statistics and maps of forest biomass and increment in Europe
- 2024
-
Ingår i: Scientific Data. - 2052-4463. ; 11
-
Tidskriftsartikel (refereegranskat)abstract
- Forest biomass is an essential resource in relation to the green transition and its assessment is key for the sustainable management of forest resources. Here, we present a forest biomass dataset for Europe based on the best available inventory and satellite data, with a higher level of harmonisation and spatial resolution than other existing data. This database provides statistics and maps of the forest area, biomass stock and their share available for wood supply in the year 2020, and statistics on gross and net volume increment in 2010–2020, for 38 European countries. The statistics of most countries are available at a sub-national scale and are derived from National Forest Inventory data, harmonised using common reference definitions and estimation methodology, and updated to a common year using a modelling approach. For those counties without harmonised statistics, data were derived from the State of Europe’s Forest 2020 Report at the national scale. The maps are coherent with the statistics and depict the spatial distribution of the forest variables at 100 m resolution.
|
|
| 8. |
- Gnacek, Michal, et al.
(författare)
-
AVDOS-VR : Affective Video Database with Physiological Signals and Continuous Ratings Collected Remotely in VR
- 2024
-
Ingår i: Scientific Data. - 2052-4463. ; 11
-
Tidskriftsartikel (refereegranskat)abstract
- Investigating emotions relies on pre-validated stimuli to evaluate induced responses through subjective self-ratings and physiological changes. The creation of precise affect models necessitates extensive datasets. While datasets related to pictures, words, and sounds are abundant, those associated with videos are comparatively scarce. To overcome this challenge, we present the first virtual reality (VR) database with continuous self-ratings and physiological measures, including facial EMG. Videos were rated online using a head-mounted VR device (HMD) with attached emteqPRO mask and a cinema VR environment in remote home and laboratory settings with minimal setup requirements. This led to an affective video database with continuous valence and arousal self-rating measures and physiological responses (PPG, facial-EMG (7x), IMU). The AVDOS-VR database includes data from 37 participants who watched 30 randomly ordered videos (10 positive, neutral, and negative). Each 30-second video was assessed with two-minute relaxation between categories. Validation results suggest that remote data collection is ecologically valid, providing an effective strategy for future affective study designs. All data can be accessed via: www.gnacek.com/affective-video-database-online-study.
|
|
| 9. |
- Halbritter, Aud H., et al.
(författare)
-
Plant trait and vegetation data along a 1314 m elevation gradient with fire history in Puna grasslands, Perú
- 2024
-
Ingår i: SCIENTIFIC DATA. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Alpine grassland vegetation supports globally important biodiversity and ecosystems that are increasingly threatened by climate warming and other environmental changes. Trait-based approaches can support understanding of vegetation responses to global change drivers and consequences for ecosystem functioning. In six sites along a 1314 m elevational gradient in Puna grasslands in the Peruvian Andes, we collected datasets on vascular plant composition, plant functional traits, biomass, ecosystem fluxes, and climate data over three years. The data were collected in the wet and dry season and from plots with different fire histories. We selected traits associated with plant resource use, growth, and life history strategies (leaf area, leaf dry/wet mass, leaf thickness, specific leaf area, leaf dry matter content, leaf C, N, P content, C and N isotopes). The trait dataset contains 3,665 plant records from 145 taxa, 54,036 trait measurements (increasing the trait data coverage of the regional flora by 420%) covering 14 traits and 121 plant taxa (ca. 40% of which have no previous publicly available trait data) across 33 families.
|
|
| 10. |
- Latz, Meike, et al.
(författare)
-
A comprehensive dataset on spatiotemporal variation of microbial plankton communities in the Baltic Sea
- 2024
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- The Baltic Sea is one of the largest brackish water environments on earth and is characterised by pronounced physicochemical gradients and seasonal dynamics. Although the Baltic Sea has a long history of microscopy-based plankton monitoring, DNA-based metabarcoding has so far mainly been limited to individual transect cruises or time-series of single stations. Here we report a dataset covering spatiotemporal variation in prokaryotic and eukaryotic microbial communities and physicochemical parameters. Within 13-months between January 2019 and February 2020, 341 water samples were collected at 22 stations during monthly cruises along the salinity gradient. Both salinity and seasonality are strongly reflected in the data. Since the dataset was generated with both metabarcoding and microscopy-based methods, it provides unique opportunities for both technical and ecological analyses, and is a valuable biodiversity reference for future studies, in the prospect of climate change.
|
|
