| 1. |
- Bengtsson Bernander, Karl, et al.
(författare)
-
Classification of rotation-invariant biomedical images using equivariant neural networks
- 2024
-
Ingår i: Scientific Reports. - : Springer Nature. - 2045-2322. ; 14:1
-
Tidskriftsartikel (refereegranskat)abstract
- Transmission electron microscopy (TEM) is an imaging technique used to visualize and analyze nano-sized structures and objects such as virus particles. Light microscopy can be used to diagnose diseases or characterize e.g. blood cells. Since samples under microscopes exhibit certain symmetries, such as global rotation invariance, equivariant neural networks are presumed to be useful. In this study, a baseline convolutional neural network is constructed in the form of the commonly used VGG16 classifier. Thereafter, it is modified to be equivariant to the p4 symmetry group of rotations of multiples of 90 degrees using group convolutions. This yields a number of benefits on a TEM virus dataset, including higher top validation set accuracy by on average 7.6% and faster convergence during training by on average 23.1% of that of the baseline. Similarly, when training and testing on images of blood cells, the convergence time for the equivariant neural network is 7.9% of that of the baseline. From this it is concluded that augmentation strategies for rotation can be skipped. Furthermore, when modelling the accuracy versus amount of TEM virus training data with a power law, the equivariant network has a slope of - 0.43 compared to - 0.26 of the baseline. Thus the equivariant network learns faster than the baseline when more training data is added. This study extends previous research on equivariant neural networks applied to images which exhibit symmetries to isometric transformations.
|
|
| 2. |
|
|
| 3. |
- Bernander, Karl B., et al.
(författare)
-
Improving the stochastic watershed
- 2013
-
Ingår i: Pattern Recognition Letters. - : Elsevier BV. - 0167-8655 .- 1872-7344. ; 34:9, s. 993-1000
-
Tidskriftsartikel (refereegranskat)abstract
- The stochastic watershed is an unsupervised segmentation tool recently proposed by Angulo and Jeulin. By repeated application of the seeded watershed with randomly placed markers, a probability density function for object boundaries is created. In a second step, the algorithm then generates a meaningful segmentation of the image using this probability density function. The method performs best when the image contains regions of similar size, since it tends to break up larger regions and merge smaller ones. We propose two simple modifications that greatly improve the properties of the stochastic watershed: (1) add noise to the input image at every iteration, and (2) distribute the markers using a randomly placed grid. The noise strength is a new parameter to be set, but the output of the algorithm is not very sensitive to this value. In return, the output becomes less sensitive to the two parameters of the standard algorithm. The improved algorithm does not break up larger regions, effectively making the algorithm useful for a larger class of segmentation problems.
|
|
| 4. |
- Bitencourt de Oliveira, Felipe, 1982, et al.
(författare)
-
Assessing Metal Use and Scarcity Impacts of Vehicle Gliders
- 2024
-
Ingår i: Circular Economy and Sustainability. - 2730-597X .- 2730-5988. ; In Press
-
Tidskriftsartikel (refereegranskat)abstract
- This study assesses the metal composition of two vehicle gliders, configured with different equipment levels and evaluates the risk of short and long-term metal scarcity. Entropy analysis is also used for insights on secondary metal recovery strategies. Fifty-five metals are evaluated, with gold, copper, bismuth, lead, molybdenum, and certain rare-earth metals (REMs) subject to the largest supply risks. Differences in equipment levels significantly impact the short-term supply risk for specific metals. Entertainment and communications equipment contain significant amounts of REMs, whereas mirrors and electrical infrastructure contain considerable shares of gold, silver and copper. Some metals are concentrated in a few components while some are dispersed across thousands, impacting recycling opportunities. The broad metal demand of the gliders underscores the automotive industry's role in supply risks for its own manufacturing needs and other societal domains. This emphasizes the significance of comprehensively evaluating metal requirements beyond powertrains for informed resource management.
|
|
| 5. |
- Lexén, Jenny, et al.
(författare)
-
Assessing exposure of semi-volatile organic compounds (SVOCs) in car cabins : Current understanding and future challenges in developing a standardized methodology
- 2021
-
Ingår i: Environment International. - : Elsevier Ltd. - 0160-4120 .- 1873-6750. ; 157
-
Tidskriftsartikel (refereegranskat)abstract
- Semi-volatile organic compounds (SVOCs) can be found in air, dust and on surfaces in car cabins, leading to exposure to humans via dust ingestion, inhalation, and dermal contact. This review aims at describing current understanding concerning sampling, levels, and human exposure of SVOCs from car cabin environments. To date, several different methods are used to sample SVOCs in car cabin air and dust and there are no standard operating procedures for sampling SVOCs in cars detailed in the literature. The meta-analysis of SVOCs in car cabin air and dust shows that brominated flame retardants (BFRs) and organophosphate flame retardants (OPFRs) have been most frequently studied, primarily focusing on concentrations in dust. In dust, detected concentrations span over three to seven orders of magnitude, with highest median concentrations for OPFRs, followed by BFRs and, thereafter, polychlorinated biphenyls (PCBs). In air, the variation is smaller, spanning over one to three orders of magnitude, with phthalates and siloxanes having the highest median concentrations, followed by OPFRs, fluorotelomer alcohols (FTOHs) and BFRs. Assessments of human exposures to SVOCs in cars have, so far, mainly focused on external exposure, most often only studying one exposure route, primarily via dust ingestion. In order to perform relevant and complete assessments of human exposure to SVOCs in cars, we suggest broadening the scope to which SVOCs should be studied, promoting more comprehensive external exposure assessments that consider exposure via all relevant exposure routes and making comparisons of external and internal exposure, in order to understand the importance of in-car exposure as a source of SVOC exposure. We also suggest a new sampling approach that includes sampling of SVOCs in both car cabin air and dust, aiming to reduce variability in data due to differences in sampling techniques and protocols. © 2021 The Authors
|
|
| 6. |
- Lexén, Jenny, et al.
(författare)
-
Concentrations of potentially endocrine disrupting chemicals in car cabin air and dust : effect of temperature and ventilation
- 2024
-
Ingår i: Science of the Total Environment. - : Elsevier. - 0048-9697 .- 1879-1026. ; 947
-
Tidskriftsartikel (refereegranskat)abstract
- Materials in car cabins contain performance-enhancing semi-volatile organic compounds (SVOCs). As these SVOCs are not chemically bound to the materials, they can emit from the materials at slow rates to the surrounding, causing human exposure. This study aimed at increasing the understanding on abundance of SVOCs in car cabins by studying 18 potential endocrine disrupting chemicals in car cabin air (gas phase and airborne particles) and dust. We also studied how levels of these chemicals varied by temperature inside the car cabin along with ventilation settings, relevant to human exposure. A positive correlation was observed between temperature and SVOC concentration in both the gas and the particle phase, where average gas phase levels at 80 °C were a factor of 18–16,000 higher than average levels at 25 °C, while average particle phase levels were a factor of 4.6–40,000 higher for the studied substances. This study also showed that levels were below the limit of detection for several SVOCs during realistic driving conditions, i.e., with the ventilation activated. To limit human exposure to SVOCs in car cabins, it is recommended to ventilate a warm car before entering and have the ventilation on during driving, as both temperature and ventilation have a significant impact on SVOC levels.
|
|
| 7. |
- Lindås, Ann-Christin, et al.
(författare)
-
A unique cell division machinery in the Archaea
- 2008
-
Ingår i: Proceedings of the National Academy of Sciences of the United States of America. - : Proceedings of the National Academy of Sciences. - 0027-8424 .- 1091-6490. ; 105:48, s. 18942-18946
-
Tidskriftsartikel (refereegranskat)abstract
- In contrast to the cell division machineries of bacteria, euryarchaea, and eukaryotes, no division components have been identified in the second main archaeal phylum, Crenarchaeota. Here, we demonstrate that a three-gene operon, cdv, in the crenarchaeon Sulfolobus acidocaldarius, forms part of a unique cell division machinery. The operon is induced at the onset of genome segregation and division, and the Cdv proteins then polymerize between segregating nucleoids and persist throughout cell division, forming a successively smaller structure during constriction. The cdv operon is dramatically down-regulated after UV irradiation, indicating division inhibition in response to DNA damage, reminiscent of eukaryotic checkpoint systems. The cdv genes exhibit a complementary phylogenetic range relative to FtsZ-based archaeal division systems such that, in most archaeal lineages, either one or the other system is present. Two of the Cdv proteins, CdvB and CdvC, display homology to components of the eukaryotic ESCRT-III sorting complex involved in budding of luminal vesicles and HIV-1 virion release, suggesting mechanistic similarities and a common evolutionary origin.
|
|
| 8. |
- Wei, Dixin, 1993, et al.
(författare)
-
PM2.5 and ultrafine particles in passenger car cabins in Sweden and northern China—the influence of filter age and pre-ionization
- 2020
-
Ingår i: Environmental Science and Pollution Research. - : Springer Science and Business Media LLC. - 0944-1344 .- 1614-7499. ; 27:24, s. 30815-30830
-
Tidskriftsartikel (refereegranskat)abstract
- The main aim of the study was to evaluate the influence of filter status (new and aged), pre-ionization, on the particle filtration in modern passenger cars. Measurements of in-cabin and outside PM2.5 (dp < 2.5 μm) concentration and UFP (ultrafine particle, dp < 100 nm) counts, to calculate I/O (indoor to outdoor) ratios, were performed. They were done at two locations, to study the influence of different outside conditions on the HVAC (heating, ventilation, and air-conditioning) system. The measurements were performed in two new cars, with similar HVAC systems and settings, using a new and an aged synthetic filter. Furthermore, an ionization unit was installed upstream of the filter in both cars. This enabled the study of filter status, with and without ionization, under common driving conditions. The results show that the HVAC system performances were very similar at the two locations, with average I/O ratios of 0.35-0.40 without ionization and 0.15-0.20 with ionization applied, although the outside conditions were considerably different. Furthermore, the aged filter clearly worsened the filtration ability. Considering the corresponding average PM2.5 I/O ratios in one location as an example, the average for the new filter was 0.20 and 0.60 for the aged filter. The corresponding UFP I/O ratios were 0.24 and 0.57. Other findings are that the aged filter with ionization reached a performance close to the new filter (without ionization), and that increased ventilation airflow and decreased recirculation degree, as expected, led to an increase of I/O ratio for both particle sizes.
|
|
