Classifying and identifying surface defects is essential during the production and use of aluminum profiles. Recently, the dual-convolutional neural network(CNN) model fusion framework has shown promising performance for defects classification and recognition. Spurred by this trend, this paper proposes an improved dual-CNN model fusion framework to classify and identify defects in aluminum profiles. Compared with traditional dual-CNN model fusion frameworks, the proposed architecture involves an improved fusion layer, fusion strategy, and classifier block. Specifically, the suggested method extracts the feature map of the aluminum profile RGB image from the pre-trained VGG16 model's pool5 layer and the feature map of the maximum pooling layer of the suggested A4 network, which is added after the Alexnet model. then, weighted bilinear interpolation unsamples the feature maps extracted from the maximum pooling layer of the A4 part. The network layer and upsampling schemes ensure equal feature map dimensions ensuring feature map merging utilizing an improved wavelet transform. Finally, global average pooling is employed in the classifier block instead of dense layers to reduce the model's parameters and avoid overfitting. The fused feature map is then input into the classifier block for classification. The experimental setup involves data augmentation and transfer learning to prevent overfitting due to the small-sized data sets exploited, while the K cross-validation method is employed to evaluate the model's performance during the training process. The experimental results demonstrate that the proposed dual-CNN model fusion framework attains a classification accuracy higher than current techniques, and specifically 4.3% higher than Alexnet, 2.5% for VGG16, 2.9% for Inception v3, 2.2% for VGG19, 3.6% for Resnet50, 3% for Resnet101, and 0.7% and 1.2% than the conventional dual-CNN fusion framework 1 and 2, respectively, proving the effectiveness of the proposed strategy.
Citation: Xiaochen Liu, Weidong He, Yinghui Zhang, Shixuan Yao, Ze Cui. Effect of dual-convolutional neural network model fusion for Aluminum profile surface defects classification and recognition[J]. Mathematical Biosciences and Engineering, 2022, 19(1): 997-1025. doi: 10.3934/mbe.2022046
| [1] | Theresa Vertigan, Kriya Dunlap, Arleigh Reynolds, Lawrence Duffy . Effects of Methylmercury exposure in 3T3-L1 Adipocytes. AIMS Environmental Science, 2017, 4(1): 94-111. doi: 10.3934/environsci.2017.1.94 |
| [2] | Gytautas Ignatavičius, Murat H. Unsal, Peter E. Busher, Stanisław Wołkowicz, Jonas Satkūnas, Giedrė Šulijienė, Vaidotas Valskys . Geochemistry of mercury in soils and water sediments. AIMS Environmental Science, 2022, 9(3): 277-297. doi: 10.3934/environsci.2022019 |
| [3] | Alma Sobrino-Figueroa, Sergio H. Álvarez Hernandez, Carlos Álvarez Silva C . Evaluation of the freshwater copepod Acanthocyclops americanus (Marsh, 1983) (Cyclopidae) response to Cd, Cr, Cu, Hg, Mn, Ni and Pb. AIMS Environmental Science, 2020, 7(6): 449-463. doi: 10.3934/environsci.2020029 |
| [4] | Kayla L. Penta, Diego Altomare, Devon L. Shirley, Jennifer F. Nyland . Female immune system is protected from effects of prenatal exposure to mercury. AIMS Environmental Science, 2015, 2(3): 448-463. doi: 10.3934/environsci.2015.3.448 |
| [5] | Martina Grifoni, Francesca Pedron, Gianniantonio Petruzzelli, Irene Rosellini, Meri Barbafieri, Elisabetta Franchi, Roberto Bagatin . Assessment of repeated harvests on mercury and arsenic phytoextraction in a multi-contaminated industrial soil. AIMS Environmental Science, 2017, 4(2): 187-205. doi: 10.3934/environsci.2017.2.187 |
| [6] | Margarida Casadevall, Conxi Rodríguez-Prieto, Jordi Torres . The importance of the age when evaluating mercury pollution in fishes: the case of Diplodus sargus (Pisces, Sparidae) in the NW Mediterranean. AIMS Environmental Science, 2017, 4(1): 17-26. doi: 10.3934/environsci.2017.1.17 |
| [7] | Romilda Z. Boerleider, Nel Roeleveld, Paul T.J. Scheepers . Human biological monitoring of mercury for exposure assessment. AIMS Environmental Science, 2017, 4(2): 251-276. doi: 10.3934/environsci.2017.2.251 |
| [8] | Nurul Akhma Zakaria, A.A. Kutty, M.A. Mahazar, Marina Zainal Abidin . Arsenic acute toxicity assessment on select freshwater organism species in Malaysia. AIMS Environmental Science, 2016, 3(4): 804-814. doi: 10.3934/environsci.2016.4.804 |
| [9] | Cornelius Tsamo, Tita Mangi Germaine, Adjia Henriette Zangue . Reuse of wastewaters from slaughterhouse and palm oil mill: Influence on the growth performance of catfish (Clarias gariepinus). AIMS Environmental Science, 2023, 10(6): 743-763. doi: 10.3934/environsci.2023041 |
| [10] | Junaidi, Hafrijal Syandri, Azrita, Abdullah Munzir . Floating cage aquaculture production in Indonesia: Assessment of opportunities and challenges in Lake Maninjau. AIMS Environmental Science, 2022, 9(1): 1-15. doi: 10.3934/environsci.2022001 |
Classifying and identifying surface defects is essential during the production and use of aluminum profiles. Recently, the dual-convolutional neural network(CNN) model fusion framework has shown promising performance for defects classification and recognition. Spurred by this trend, this paper proposes an improved dual-CNN model fusion framework to classify and identify defects in aluminum profiles. Compared with traditional dual-CNN model fusion frameworks, the proposed architecture involves an improved fusion layer, fusion strategy, and classifier block. Specifically, the suggested method extracts the feature map of the aluminum profile RGB image from the pre-trained VGG16 model's pool5 layer and the feature map of the maximum pooling layer of the suggested A4 network, which is added after the Alexnet model. then, weighted bilinear interpolation unsamples the feature maps extracted from the maximum pooling layer of the A4 part. The network layer and upsampling schemes ensure equal feature map dimensions ensuring feature map merging utilizing an improved wavelet transform. Finally, global average pooling is employed in the classifier block instead of dense layers to reduce the model's parameters and avoid overfitting. The fused feature map is then input into the classifier block for classification. The experimental setup involves data augmentation and transfer learning to prevent overfitting due to the small-sized data sets exploited, while the K cross-validation method is employed to evaluate the model's performance during the training process. The experimental results demonstrate that the proposed dual-CNN model fusion framework attains a classification accuracy higher than current techniques, and specifically 4.3% higher than Alexnet, 2.5% for VGG16, 2.9% for Inception v3, 2.2% for VGG19, 3.6% for Resnet50, 3% for Resnet101, and 0.7% and 1.2% than the conventional dual-CNN fusion framework 1 and 2, respectively, proving the effectiveness of the proposed strategy.
Mercury (Hg) is considered a devastating environmental pollutant, mainly after the environmental disaster at Minamata (Japan) and several other poisoning accidents due to the use of Hg pesticides in agriculture [1]. Hg exists as elemental form, inorganic (iHg) and organic Hg (primarily methylmercury, MeHg). In the environment, Hg is released from natural and anthropogenic sources [2]. iHg enters the air from mining ore deposits, burning coal and waste, or from manufacturing plants. It enters the water or soil from natural deposits, disposal of wastes, and the use of Hg containing fungicides. iHg is maintained in the upper sedimentary layers of water-beds and is methylated and thus transformed to the highly toxic species MeHg by sulphate-reducing bacteria.
Numerous toxicological studies have gained increasing interest in order to understand the impact of Hg on aquatic communities, which are particularly vulnerable. In this regard, fish have been used to assess pollution because they represent the most diverse group of vertebrates [3] and have genetic relatedness to the higher vertebrates including mammals. It is widely known that fish are a great source of Hg in our food and their accumulation could represent a serious risk for human beings [4]. Although fish have always been perceived as a healthy and nutritive food [5], the Environmental Protection Agency (EPA) has raised public concern as it claimed that the levels of Hg in certain fish species make them unsuitable or be restricted for children and pregnant women consumption [6]. Some data about the Hg concentration in common fish species are shown in Table 1. This data are worrying us if we consider that the projections show an increase in the demand for seafood products to year 2030.
| Specie | Mercury content (ppm) | Safety |
| Anchovies | 0.017 ± 0.015 | Eco-good |
| Atlantic cod | 0.095 ± 0.080 | Eco-bad |
| Bass (saltwater, black, striped, rockfish) | 0.167 ± 0.194 | Eco-bad |
| Bass Chilean | 0.354 ± 0.199 | Eco-bad |
| Carp | 0.140 ± 0.099 | Eco-bad |
| Catfish | 0.049 ± 0.084 | Eco-good |
| Croaker Atlantic (Atlantic) | 0.069 ± 0.049 | Eco-good |
| Croaker White (Pacific) | 0.287 ± 0.069 | Eco-bad |
| Grouper (all species) | 0.448 ± 0.287 | Eco-bad |
| Mullet | 0.05 ± 0.078 | Eco-bad |
| Salmon, wild (Alaska) | 0.014 ± 0.041 | Eco-good |
| Sardines, Pacific (US) | 0.016 ± 0.007 | Eco-good |
| Shark | 0.979 ± 0.626 | Eco-bad |
| Swordfish | 0.976 ± 0.510 | Eco-bad |
| Tilapia | 0.013 ± 0.023 | Eco-good |
| Trout, rainbow (farmed, freshwater) | 0.072 ± 0.143 | Eco-good |
| Tuna species | 0.415 ± 0.308 | Eco-bad |
DownLoad: CSVIn this regard, aquaculture is one of the most important food manufacturing industries in the coming decades trying to compensate the human consumption demand. Moreover, farmers have to know and control the impact of the environmental contaminants in the species produced for humans [7]. In this specific field, very little is known about the effects of Hg exposure in fish, and the comparison of any results is very difficult and sometimes contradictory because different investigators have used a variety of administration methods as well as concentrations of Hg. Moreover, the development of prominent trends in toxicity testing based on in vitro tests and especially in vitro mechanistic assays has gain considerable attention in recent years.
In the aquatic environment, Hg speciation, uptake, bioavailability, and toxicity are dependent upon environmental parameters including hydrophobicity, pH, salinity, hardness (Ca2+ and Mg2+ concentration) and interaction of metals with biotic and abiotic ligands [8]. In teleost fish species, bioavailability of Hg depend not only on total chemical concentration in the environment but also on how readily the fish can absorb these different Hg forms at the gill, across the skin, and within the digestive tract and on how chemical speciation affects distribution throughout the organism [9]. Thus, the metal absorption and distribution occurs as follows: first, Hg crosses the epithelium; second, it incorporates into blood including binding to plasma proteins, transport via the systemic circulation or freely dissolved to various tissues; and finally, it is transported from blood into tissues. Gills, digestive system, and, to a lesser extent, the skin, are the major sites of metal uptake in fish [8].
Hg uptake can be passive or energy-dependent, depending on the Hg species [10]. Concretely, in plasma, MeHg binds reversibly to cysteine amino acid and therefore to sulphur-containing molecules such as glutathione (GSH). The cysteine-bound form is of particular interest because it is transported by an L-neutral amino acid transporter system into the cells of sensitive tissues such as brain [11]. In the gastrointestinal tract, ingested MeHg is efficiently absorbed and its distribution to the blood is complete within approximately 30 h, and the blood level accounts for about 7% of the ingested dose. The brain is the primary target site for MeHg and approximately 10% is retained in the brain with the remainder transported to the liver and kidney where it is excreted through bile and urine [12]. In rainbow trout (Oncorhynchus mykiss), however, 90% of whole-blood MeHg is bound to the beta-chain of hemoglobin in red blood cells [13]. In addition, MeHg readily binds to metallothioneins (MTs) and metalloproteins with cysteine residues displacing Zn2+ [10,14]. The primary mechanism of MeHg as well as its specificity has yet to be identified. MeHg-cysteine conjugates have shown increased cellular efflux, presumably due to the generation and involvement of glutathione.
Regarding its bioaccumulation, concentrations of MeHg are magnified within the food chain, reaching concentrations in fish 10, 000-to 100, 000-fold greater than in the surrounding water [2]. The primary target tissues for Hg are the central nervous system (CNS) [15] and the kidney, triggering loss of appetite, brain lesions, cataracts, abnormal motor coordination, and behavioural changes, alterations that lead to the fish to have impaired growth, reproduction, and development.
Several studies have shown that Hg produce an imbalance between the reactive oxygen species (ROS) production and its clearance by the antioxidant system in the known oxidative stress response. Thus, in fish, it has been described the production of ROS after Hg exposure in vivo [16,17,18,19,20,21] or in vitro [22,23,24,25,26]. Indeed, Hg reacts with the thiol groups of GSH, which can induce GSH depletion and oxidative stress [19,20,21,27,28]. Thus, some studies have found alterations in the antioxidant system caused by Hg exposure including glutathione reductase (GR) and glutathione peroxidase (GPx) activities in zebra seabream (Diplodus cervinus) [29], modifications in superoxide dismutase (SOD), catalase (CAT), glutathione S-transferase (GST), and GPx activities in trahira species (Hoplias malabaricus) [30]. Very recently, MeHg exposure increased the ROS levels and decreased the antioxidant potential of gilthead seabream (Sparus aurata) serum while increased the SOD, CAT and GR activities in the liver [18]. In addition, significant alterations in the expression of the antioxidant enzyme genes sod, cat, gst, gpx, and gr have been observed in the freshwater fish Yamú (Brycon amazonicus) after Hg exposure leading to oxidation of lipids and proteins [31]. In a set of studies conducted on wild golden grey mullet (Liza aurata) it has been demonstrated the depletion on reduced GSH, GPx, SOD and lipid peroxidation (LPO) and increment on total GSH, GST and CAT activities towards other metabolites indicating the potential of using metabolomics to determine and probe the Hg exposure and oxidative stress relations [19,20,21]. In vitro studies including fish cell lines [SAF-1, derived from gilthead seabream, and DLB-1, derived from European sea bass (Dicentrarchus labrax)] or primary fish cell cultures (leucocytes or erythrocytes) showed an increase in ROS levels after Hg exposure; however, an up-or down-regulation in the expression of some antioxidant genes (sod, cat, gr or prx) was observed [24,25,26,27,32,33,34].
MTs also play a protective role in response to Hg exposure. The mRNA expression of two mt genes was noted in the liver of common carp (Cyprinus carpio) from Hg contaminated river [35]. However, no significant correlations between total Hg content and MT levels were described in different fish tissues from a Hg contaminated area [36]. Thus, in in vitro studies, mt gene expression was enhanced after Hg exposure in SAF-1 [26] and DLB-1 [24] cell lines as well as in gilthead seabream or European sea bass peripheral blood leucocytes (PBLs) [33] and erythrocytes [34]; however, a strong down-regulation was observed in leucocytes derived from the head-kidney, the main hematopoietic tissue, from the same species [25,32]. Indeed, Hg bind easily to MTs [37,38] but an excess of metal could provoke a MT dysfunction leading to an increase of ROS levels as we have observed in some fish or fish cell lines before.
Furthermore, the ROS imbalance leads to cell death by apoptosis. Hg forms induce apoptosis by inhibiting mitochondrial function [39] and releasing cytochrome C from the mitochondria to the cytosol [40]. Moreover, p38 mitogen-activated protein kinase is activated by mercury resulting in apoptosis [41]. In addition, mitochondrial membrane permeability is regulated through a family of anti-apoptotic (Bcl-2, Bcl-xL, etc.) and pro-apoptotic (Bad, Bax, Bak, Bid, etc.) proto-oncogenes [42]. This cell death mechanism has been demonstrated in fish exposed to Hg [43,44,45] as well as in in vitro fish systems including cell lines [24,26] and primary cultures of head-kidney leucocytes [25,32], PBLs [33] and erythrocytes [34].
Although the immunotoxicological effects of mercurial compounds have been well studied in mammals [46,47] far less is known concerning the effects in fish to date [7,48]. Moreover, due to the complexity and multifaceted of the immune system, recent articles show the difficulty to assess the immunotoxicological effects in fish and the challenge to select appropriate exposure and effect parameters out of the many immune parameters which could be measured [49]. Because of the immunotoxicological studies of Hg in fish have focused on almost exclusively on immunosuppressive effects, some aspects have been ignored. For instance, a range of scientific studies have attempted to investigate the role of Hg in mammalian autoimmunity [50] or hypersensibility [51,52].
MeHg exposure by dietary intake [53], injection [54] or waterborne [18] can trigger alterations of the fish immune system. Upon immune mediators, little is known about their regulation of cytokines in fish. The freshwater fish snakehead (Channa punctatus) exposed to 0.3 mg/L HgCl2showed an up-regulation of pro-inflammatory cytokines such as tumour necrosis factor-α (tnfα) and interleukin-6 (il6) after 7 days of exposure [1]. In vitro models have shown a down-regulation of pro-inflammatory il1b gene expression after 2 or 24 h of MeHg exposure in gilthead seabream head-kidney leucocytes, while alterations were not observed in European sea bass leucocytes [25,32]. On the other hand, effects on soluble humoral factors of fish have been widely studied after Hg exposure. For example, lysozyme activity was enhanced in goldfish (Carassius auratus) [55] or rainbow trout [56] after exposure to different HgCl2 concentrations but decreased in plaice (Pleuronectes platessa) [57]. Other humoral activities such as serum complement or peroxidase activities or total IgM levels were increased or impaired after MeHg exposure in diverse fish species [18,54,56]. Interestingly, waterborne MeHg significantly increased the immune responses in the gilthead seabream skin mucus including the microbicidal activity [18]. Upon the cellular innate immune response, MeHg increased the phagocytosis in gilthead seabream head-kidney leucocytes exposed in vivo by waterborne [18] and in vitro [25] whilst in European sea bass leucocytes the phagocyte functions (phagocytosis and respiratory burst) by in vitro exposure to Hg was reduced [22,32]. In addition, in the European sea bass leucocytes, in vitro treatment with HgCl2 induced apoptosis in head-kidney macrophages as well as reduced the ROS production and the benefits of macrophage-activating factors (MAF) [22,58]. Thus, the presence of factors such as the serum levels of corticosteroids and catecholamines that do not operate in vitro could explain the differences observed between in vitro and in vivo studies [59]. However, the mechanisms by which metals alter the phagocyte functions are still poorly understood. Strikingly, few studies have dealt with the leucocyte death by Hg. Thus, an increase in the transcription of genes related to apoptosis was shown after MeHg exposure to gilthead seabream or European sea bass blood leucocytes (PBLs) [33]. However, Hg exposure promoted both apoptosis and necrosis cell death in gilthead seabream or European sea bass head-kidney leucocytes [25,32].
Although exposure to some metals often disturbs normal metabolic processes in fish, including irritation of respiratory epithelium, changes in ventilation frequency, or inefficient oxygen delivery to tissues [60] considerably less is known regarding the effect of Hg on the respiratory system. Previous studies have shown that an exposure to dissolved Hg disrupts gill epithelium, potentially affecting gas exchange and permeability of cell membranes to cations [61,62]. Such disruptions may result in compensatory changes in ventilation frequency, increased energy demands, or altered gas exchange efficiency, possibly resulting in the increase in metabolic rate [63]. However, gill damage and a subsequent increase of metabolic rate did not occur in the mosquitofish (Gambusia holbrooki) fed with HgCl2 probably due to the Hg accumulation via intestinal absorption from dietary sources [64]. More recently, HgCl2 via food or diet decreases the plasticity of the cardiorespiratory responses reducing the survival chances of Yamú and trahira under hypoxic conditions frequently observed in theirs wild [65].
Upon histopathology studies, the gills is the organ to better study due to their function in the respiration process and since this organ is continuously and directly exposed to the external environment [62,66,67,68,69,70]. Moreover, a report shows an increase of the chloride cells (CCs) in the European sea bass gills after exposure to Hg [71], which is in concordance with another study using mosquitofish [62]. The CAT activity remained unchanged while GPx and GR activities showed a significant decrease in trahira gills after exposure to HgCl2 [65]. In wild grey mullet from Hg-contaminated areas, gill GPx and SOD activities were depleted while GST and CAT activities were increased indicating a massive GSH oxidation [20,21]. In in vitro conditions, gill cell suspensions exposed to HgCl2 showed high rate of DNA breaks (single and double stranded) measured as the comet assay in common carp and rainbow trout [72,73].
Similarly, Hg accumulation in the heart is thought to contribute to cardiomyopathy. The mechanism by which Hg produces toxic effects on the cardiovascular system is not fully elucidated, but this mechanism is believed to involve an increase in oxidative stress [74]. In a recent study, trahira specimens exposed to HgCl2 showed an anticipated bradycardia and lower heart rate probably due to damage in cardiac myocytes [65], which is in agreement with other studies in trahira [75,76] or in other tropical species [77].
Numerous studies suggest that the inhibitory effects of Hg on fish reproduction occur at multiple sites within the hypothalamic–pituitary–gonadal (HPG) axis [78]. Moreover, in most cases, studies have been carrying out in in vivo conditions due to the missing information of hormone release in in vitro assays. Thus, recent studies have reported a significantly lesser transcripts of gonadotropin-releasing hormone (GnRH) (gnrh2 and gnrh3) in the brain of both male and female zebrafish (Danio rerio) after Hg exposure, suggesting that Hg could modulate hypothalamic production of GnRHs in fish and consequently disrupted production of gonadotropin hormones such as follicle stimulating hormone (FSH) and luteinizing hormone (LH) [81,82]. Furthermore, an alteration in expression of genes commonly associated with endocrine disruption and a decrease in the production of vitellogenin as a result of dietary MeHg exposure in fathead minnows (Pimephales promelas) was observed [83]. Similarly, by using cDNA microarray analysis in the cutthroat trout (Salmo clarkii), expression levels of gnrh1 and gnrh2 were down-regulated in the liver from those specimens containing high Hg levels compared to those with low Hg concentrations [84]. Moreover, down regulation of genes involved in early stages of the spermatogenesis (fshβ, fshr, lhr or lhβ) or genes involved in the synthesis of steroid hormones (cyp17 and cyp11b) were observed in male zebrafish testis after exposure to 30 µg Hg/L [81] which results are concomitant with other studies in catfish (Clarias batrachus) [85]. Likewise, as in male, an impaired RNA transcription of genes involved in promoting follicular growth (lhβ and lhr) was observed in the ovary of zebrafish [81].
On the other hand, severe histological lesions in the gonad have been observed in fish after Hg exposure in vivo. In testis, exposure to HgCl2 caused the thickening of the tubule walls and disorderly arranged spermatozoa in zebrafish [81] as well as in medaka (Oryzias latipes) after MeHg exposure [86], probably resulting in inhibition of spermatogenesis. Moreover, a disorganization of seminiferous lobules, proliferation of interstitial tissue, congestion of blood vessels, reduction of germ cells and sperm aggregation was induced after HgCl2 injection in the tropical fish Gymnotus carapo [87]. In ovary, degenerative changes such as atresia (follicular degeneration) were found after HgCl2 exposure in zebrafish [81] in agreement with other studies [88,89,90].
In recent years, research of reproductive effects such as spawning success, spawning behaviour, fertilization success and fecundity after Hg has highlighted their importance in fish [91]. A significant reduction of the spawning success and an increase time to spawn was observed in female fathead minnows after MeHg feeding [82]. Most recent studies have shown the maternal transfer as a significant route of exposure of MeHg diet for larval and juvenile fish. The cellular mechanisms by which MeHg moves through the body is readily complexed with cysteine S-conjugates. This structure mimics that of methionine and can therefore be transferred across cell membranes to developing oocytes via methionine transporters [82]. It was originally thought that MeHg partitioned from stores in female tissues into developing oocytes [93]. However, recent research has shown that the diet of the maternal adult during oogenesis, rather than adult body burden is the principal source of Hg in eggs [94]. A study [93] employed the use of stable MeHg isotopes to investigate the sources of Hg transferred to eggs. Adult sheepshead minnows (Cyprinodon variegatus) were exposed to MeHg-spiked diets. The diets administered during the pre-oogenesis stage contained different MeHg isotopes than the diet administered during oogenesis, allowing us to characterize the proportion of Hg in eggs derived from maternal body burden versus maternal diet during oogenesis. The results indicate that a constant percentage of maternal body burdens were transferred to eggs across all treatments; however, the majority of total Hg found in eggs was from recent maternal dietary exposure. In another study, males of fathead minnow fed with MeHg had significantly higher gonad concentrations of MeHg than females, which may be attributed to losses of contaminants due to maternal transfer of dietary MeHg to eggs [95] and is in concordance with other studies [93,96]. Moreover, resulting embryos from the MeHg low-diet treatment displayed altered embryonic movement patterns (hyperactivity) and decreased time to hatch while embryos from the MeHg high-diet treatment had delayed hatching and increased mortality compared with the other treatments [95].
Notwithstanding that Hg neurotoxicity has been well reported in both human and mammalian models [15,97,98,99], information regarding its threat to fish nervous system and underlying mechanisms is still scarce. Both HgCl2 and MeHg are able to easily cross the blood-brain-barrier (BBB), reaching the fish brain where it exerts toxicity [100,101,102]. Moreover, both forms of Hg share the same toxic chemical entity [103] and, thus, neurotoxicity may depend mainly on the external bioavailability. Only a few neurotoxicological endpoints have been employed to evaluate the biological effects of Hg in fish, both in laboratory experiments and under field exposures. For instance, high concentrations of MeHg were accumulated in the medaka brain after exposure to graded sublethal concentrations [86] or in European sea bass brain [104]. Changes in oxidative stress profiles of the Atlantic salmon (Salmo salar) [105] and European sea bass brain [101] or alterations in the protein expression in the marine medaka brain [106] have been reported after HgCl2exposure. Adult zebrafish exposed to dietary HgCl2showed induced mt production at gene level [107]; however, no changes in the expression of genes involved in antioxidant defences, metal chelation, active efflux of organic compounds, mitochondrial metabolism, DNA repair, and apoptosis were observed [108]. Moreover, histopathological examinations of fish brain was performed, revealing a widespread neuronal degradation [105] or for the first time a deficit in the number of the cells of the white seabream (Diplodus sargus) brain as an effect of Hg deposition [102]. Curiously, as in humans, the long-term effects of Hg were also disclosed by numerous alterations on motor and mood/anxiety-like behaviour after 28 days of recovery [102]. Concerning in vitro studies, only one report has dealt with Hg toxicity on the cell line DLB-1, derived from European sea bass brain. In particular, MeHg reduced the viability of the DLB-1 cells, failed to increase ROS, reduced the cat gene expression and increased the mta transcription. Furthermore, DLB-1 cells exposed to Hg elicited a rapid cell death by apoptosis [24].
Although an effective barrier to iHg, the intestinal wall is permeable to Hg, due to the high lipid solubility of the compound [109]. Thus, the gastrointestinal tract represents a major route of entry for a wide variety of toxicants present in the diet or in the water that the fish inhabit [110,111,112]; nonetheless there is little information relating to the protective mechanisms adopted by the intestine epithelial surfaces of the fish against Hg uptake [113]. Light microscopy based investigations have demonstrated that there are alterations in the gut of snakehead and Stinging catfish (Heteropneustes fossilis) following Hg intoxication [110,112] but surprisingly no histopathological changes were described in the arctic charr (Salvelinus alpinus) following exposure to dietary HgCl2 and MeHg [113]. In contrast, perturbations including the notable presence of vacuoles within the cytoplasm along with various inclusions, myeloid bodies and modifications to the endoplasmic reticulum and mitochondria on the intestinal epithelium of European sea bass exposed to Hg have been shown [71]. Moreover, the presence of MeHg in the intestine epithelial cells of trahira and, at a minor extent, in the extracellular matrix represents the main tissue targets [30]. Although epithelial cells from intestinal mucosa represent a biological barrier that selects the entrance of essential nutrients as well as contaminants, it should be remarked the fact that MeHg absorption can occur by passive diffusion through neutral amino acids carrier proteins [114] and accumulate in the epithelial cells or toward the connective tissue and transported via bloodstream to other target organs.
Unlike mammals, Hg can also be depurated by the kidney, liver, and, possibly, the gills of fish [116]. Small amounts of MeHg were detected in the urine of treated juvenile white sturgeon (Acipenser transmontanus) and high concentrations of MeHg were also found in the kidneys and gills [116]. Thus, the high concentration of Hg in the kidneys and gills may reflect a transient state before MeHg is eliminated from these organs. Moreover, liver registered the highest elimination percentages (up to 64% in the liver, 20% in the brain, and 3% in the muscle) in European sea bass exposed to MeHg during 28 days [104]. Surprisingly, it was verified that the concentration of MeHg in gilthead seabream liver was higher, approximately two-fold [18], than the amount detected in the muscle as it happened in other fish species such as goldfish [118], zebrafish [108] or European sea bass [117]. Furthermore, Hg can cause liver damage as demonstrated by some studies in gilthead seabream [18], the arctic charr [113] or in trahira [27].
The kidney of teleost receives a large portion of the cardiac output because of their extensive portal system. The role of kidney on Hg elimination depends on the mercurial form, preferably iHg form by urine [119]. Previous studies showed that after a 4-week exposure to dietary MeHg, the kidney of sturgeon species showed prominent degeneration of the renal tubules [120]. Similar changes were reported in guppy (Poecilia reticulata) [121] and Indian catfish (Clarias batrachus) [122] exposed to waterborne MeHg. The accumulation of MeHg in the renal tubes of trahira kidney has also demonstrated [30].
Due to the affinity to conjugate reversibly to cysteine amino acid, Hg is transported by the blood to the rest of the organs [123]. However, there is little available information in the literature related to hematological responses in fish chronically exposed to Hg. Thereby, some reports have focused on the effect of Hg on fish blood cells, namely erythrocytes. Because of it is thought to compete with iron for binding to haemoglobin, which can result in impaired hemoglobin formation, Hg exposure resulted in anemia in two fish species (Channa gachua and Pleuronectes platessa) [124,125]. However, dietary MeHg increased significantly the values of haematocrit after exposure in trahira [54] suggesting an increase in the blood oxygen capacity, in contrast to the not significantly changes observed in Nile tilapia (Oreochromis niloticus) exposed to sublethal concentrations of Hg [126]. [127] observed a decrease in membrane fluidity, change of internal viscosity, and internal protein conformation and hemolysis in erythrocytes of common carp subjected to Hg. Moreover, a differential sensitivity of fish species towards the induction of erythrocyte micronuclei (MN) and other nuclear abnormalities has been reported after intraperitoneal injection of Hg [128,129] or in olden grey mullet (Liza aurata) along an environmental Hg contamination gradient [130]. On the other hand, in vitro toxicological tests using human erythrocytes are gaining traction as alternatives to in vitro tests, however few studies are available in fish. For example, gilthead seabream or European sea bass erythrocytes exposed to Hg exhibited cytotoxicity and alteration in the gene expression profile of genes involved in oxidative stress, cellular protection and apoptosis death [34].
Although Hg is ubiquitous in the environment, it is considered one of the most toxic elements or substances on the planet that continues to be dumped into our waterways and soil, spilled into our atmosphere, and consumed in our food and water. Research indicates that Hg exposure can induce a variety of adverse effects in fish at physiologic, histologic, bio-chemical, enzymatic, and genetic levels. Certain fish species, however, appear to show more sensitivity to Hg toxicity than others. Hence, Hg-induced toxicological pathology in fish is influenced by such factors as species, age, environmental conditions, exposure time, and exposure concentration. The exact causes of fish death are multiple and depend mainly on time-concentration combinations. In-depth toxicodynamics and toxicokinetics studies are necessary to establish an exact cause-effect relation. The scientific data discussed in this review provide a basis for understanding the potential impact, as well as for advancing our knowledge of the ecotoxicology and risk assessment of Hg.
This work was partly supported by Fundación Séneca (Grupo de Excelencia de la Región de Murcia 19883/GERM/15).
All authors declare no conflict of interest in this paper.
| [1] |
Z. W. Liu, L. X. Li, J. Yi, S. K. Li, Z. H. Wang, G. Wang, Influence of heat treatment conditions on bending characteristics of 6063 aluminum alloy sheets, T. Nonferr. Metal. Soc., 27 (2017), 1498–1506. doi: 10.1016/s1003-6326(17)60170-5. doi: 10.1016/s1003-6326(17)60170-5
|
| [2] |
S. Bingol, A. Bozaci, Experimental and Numerical Study on the Strength of Aluminum Extrusion Welding, Materials (Basel), 8 (2015), 4389-4399. doi: 10.3390/ma8074389. doi: 10.3390/ma8074389
|
| [3] |
L. Donati, L. Tomesani, The effect of die design on the production and seam weld quality of extruded aluminum profiles, J. Mater. Process. Technol., 164-165 (2005), 1025–1031. doi: 10.1016/j.jmatprotec.2005.02.156. doi: 10.1016/j.jmatprotec.2005.02.156
|
| [4] | C. T. Mgonja, A review on effects of hazards in foundries to workers and environment, IJISET: Int. J. Innov. Sci. Eng. Technol., 4 (2017), 326–334. |
| [5] |
J. Ahmed, B. Gao, W. l. Woo, Sparse low-rank tensor decomposition for metal defect detection using thermographic imaging diagnostics, IEEE T. Ind. Inform., 17 (2020), 1810–1820. doi: 10.1109/TⅡ.2020.2994227. doi: 10.1109/TⅡ.2020.2994227
|
| [6] |
Q. Luo, B. Gao, W. l. Woo, Y. Yang, Temporal and spatial deep learning network for infrared thermal defect detection, NDT & E. Int., 108 (2019), 102164. doi: 10.1016/j.ndteint.2019.102164. doi: 10.1016/j.ndteint.2019.102164
|
| [7] |
B. Z. Hu, B. Gao, W. l. Woo, L. F. Ruan, J. K. Jin, A Lightweight Spatial and Temporal Multi-Feature Fusion Network for Defect Detection, IEEE T. Image Process., 30 (2020), 472–486. doi: 10.1109/TIP.2020.3036770. doi: 10.1109/TIP.2020.3036770
|
| [8] |
J. Ahmed, B. Gao, W. l. Woo, Y. Zhu, Ensemble Joint Sparse Low-Rank Matrix Decomposition for Thermography Diagnosis System, IEEE T. Ind. Electronics, 68 (2020), 2648–2658. doi: 10.1109/TIE.2020.2975484. doi: 10.1109/TIE.2020.2975484
|
| [9] |
J. Sun, C. Li, X. J. Wu, V. Palade, W. Fang, An effective method of weld defect detection and classification based on machine vision, IEEE T. Ind. Inform., 15 (2019), 6322–6333. doi: 10.1109/TⅡ.2019.2896357. doi: 10.1109/TⅡ.2019.2896357
|
| [10] |
Z. F. Zhang, G. R. Wen, S. B. Chen, Weld image deep learning-based on-line defects detection using convolutional neural networks for Al alloy in robotic arc welding, J. Manuf. Process., 45 (2019), 208–216. Doi: 10.1016/j.jmapro.2019.06.023. doi: 10.1016/j.jmapro.2019.06.023
|
| [11] |
Y. Q. Bao, K. C. Song, J. Liu, Y. Y. Wang, Y. H. Yan, H. Yu, et al., Triplet-Graph Reasoning Network for Few-shot Metal Generic Surface Defect Segmentation, IEEE Trans. Instrum. Meas., 70 (2021). doi: 10.1109/TIM.2021.3083561. doi: 10.1109/TIM.2021.3083561
|
| [12] |
S. Fekri-Ershad, F. Tajeripour, Multi-resolution and noise-resistant surface defect detection approach using new version of local binary patterns, Appl. Artif. Intell., 31 (2017), 395–410. doi: 10.1080/08839514.2017.1378012. doi: 10.1080/08839514.2017.1378012
|
| [13] |
P. Y. Jong, C. S. Woosang, K. Gyogwon, S. K. Min, L. Chungki, J. L. Sang, Automated defect inspection system for metal surfaces based on deep learning and data augmentation, J. Manuf. Syst., 55 (2020), 317–324. doi: 10.1016/j.jmsy.2020.03.009. doi: 10.1016/j.jmsy.2020.03.009
|
| [14] |
K. Ihor, M. Pavlo, B. Janette, B. Jakub, Steel surface defect classification using deep residual neural network, Metals, 10 (2020), 846. doi: 10.3390/met10060846. doi: 10.3390/met10060846
|
| [15] |
S. H. Guan, M. Lei, H. Lu, A steel surface defect recognition algorithm based on improved deep learning network model using feature visualization and quality evaluation, IEEE Access, 8 (2020), 49885–49895. doi: 10.1109/ACCESS.2020.2979755. doi: 10.1109/ACCESS.2020.2979755
|
| [16] |
B. Zhang, M. M. Liu, Y. Z. Tian, G. Wu, X. H. Yang, S. Y. Shi, et al., Defect inspection system of nuclear fuel pellet end faces based on machine vision, J. Nucl. Sci. Technol., 57 (2020), 617–623. doi: 10.1080/00223131.2019.1708827. doi: 10.1080/00223131.2019.1708827
|
| [17] |
Z. H. Liu, H. B. Shi, X. F. Zhou, Aluminum Profile Type Recognition Based on Texture Features, Appl. Mech. Mater., 556–562 (2014), 2846–2851. doi: 10.4028/www.scientific.net/AMM.556-562.2846. doi: 10.4028/www.scientific.net/AMM.556-562.2846
|
| [18] |
A. Chondronasios, I. Popov, I, Jordanov., Feature selection for surface defect classification of extruded aluminum profiles, Int. J. Adv. Manuf. Technol., 83 (2015), 33–41. doi: 10.1007/s00170-015-7514-3. doi: 10.1007/s00170-015-7514-3
|
| [19] | A. Krizhevsky, I. Sutskever, G. E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM, 60 (2017), 84–90. |
| [20] |
Q. H. Li, D. Liu, Aluminum Plate Surface Defects Classification Based on the BP Neural Network, Appl. Mech. Mater., 734 (2015), 543–547. doi: 10.4028/www.scientific.net/AMM.734.543. doi: 10.4028/www.scientific.net/AMM.734.543
|
| [21] |
R. F. Wei, Y. B. Bi, Research on Recognition Technology of Aluminum Profile Surface Defects Based on Deep Learning, Materials (Basel), 12 (2019), 1681. doi: 10.3390/ma12101681. doi: 10.3390/ma12101681
|
| [22] |
F. M. Neuhauser, G. Bachmann, P. Hora, Surface defect classification and detection on extruded aluminum profiles using convolutional neural networks, Int. J. Mater. Form., 13 (2019), 591–603. doi: 10.1007/s12289-019-01496-1. doi: 10.1007/s12289-019-01496-1
|
| [23] |
D. F. Zhang, K. C. Song, J. Xu, Y. He, Y. H. Yan, Unified detection method of aluminium profile surface defects: Common and rare defect categories, Opt. Lasers Eng., 126 (2020), 105936. doi: 10.1016/j.optlaseng.2019.105936. doi: 10.1016/j.optlaseng.2019.105936
|
| [24] |
R. X. Chen, D. Y. Cai, X. L. Hu, Z. Zhan, S. Wang, Defect Detection Method of Aluminum Profile Surface Using Deep Self-Attention Mechanism under Hybrid Noise Conditions, IEEE Trans. Instrum. Meas., (2021). doi: 10.1109/TIM.2021.3109723. doi: 10.1109/TIM.2021.3109723
|
| [25] |
J. Liu, K. C. Song, M. Z. Feng, Y. H. Yan, Z. B. Tu, L. Liu, Semi-supervised anomaly detection with dual prototypes autoencoder for industrial surface inspection, Opt. Lasers Eng., 136 (2021), 106324. doi: 10.1016/j.optlaseng.2020.106324. doi: 10.1016/j.optlaseng.2020.106324
|
| [26] |
C. M. Duan, T. C. Zhang, Two-Stream Convolutional Neural Network Based on Gradient Image for Aluminum Profile Surface Defects Classification and Recognition, IEEE Access, 8 (2020), 172152-172165. doi: 10.1109/ACCESS.2020.3025165. doi: 10.1109/ACCESS.2020.3025165
|
| [27] |
Y. L. Yu, F. X. Liu, A Two-Stream Deep Fusion Framework for High-Resolution Aerial Scene Classification, Comput. Intell. Neurosci., 2018 (2018), 8639367. doi: 10.1155/2018/8639367. doi: 10.1155/2018/8639367
|
| [28] |
C. Khraief, F. Benzarti, H. Amiri, Elderly fall detection based on multi-stream deep convolutional networks, Multimed. Tools Appl., 79 (2020), 19537–19560. doi: 10.1007/s11042-020-08812-x. doi: 10.1007/s11042-020-08812-x
|
| [29] |
W. Ye, J. Cheng, F. Yang, Y. Xu, Two-Stream Convolutional Network for Improving Activity Recognition Using Convolutional Long Short-Term Memory Networks, IEEE Access, 7 (2019), 67772–67780. doi: 10.1109/ACCESS.2019.2918808. doi: 10.1109/ACCESS.2019.2918808
|
| [30] |
Q. S. Yan, D. Gong, Y. N. Zhang, Two-Stream Convolutional Networks for Blind Image Quality Assessment, IEEE Trans. Image Process., 28 (2019), 2200–2211. doi: 10.1109/TIP.2018.2883741. doi: 10.1109/TIP.2018.2883741
|
| [31] |
T. Zhang, H. Zhang, R. Wang, Y. D. Wu, A new JPEG image steganalysis technique combining rich model features and convolutional neural networks, Math. Biosci. Eng., 16 (2019), 4069–4081. doi: 10.3934/mbe.2019201. doi: 10.3934/mbe.2019201
|
| [32] |
M. Uno, X. H. Han, Y. W. Chen, Comprehensive Study of Multiple CNNs Fusion for Fine-Grained Dog Breed Categorization, 2018 IEEE Int. Sym. Multim. (ISM), (2018), 198–203. doi: 10.1109/ISM.2018.000-7. doi: 10.1109/ISM.2018.000-7
|
| [33] | T. Akilan, Q. J. Wu, H. Zhang, Effect of fusing features from multiple DCNN architectures in image classification, IET Image Process., 12 (2018), 1102–1110. |
| [34] | D. J. Li, H. T. Guo, B. M. Zhang, C. Zhao, D. H. Yu, Double vision full convolution network for object extraction in remote sensing imagery, J. Image Graph., 25 (2020), 0535–0545. |
| [35] | M. Lin, Q. Chen, S. Yan, Network In Network, arXiv preprint arXiv: 1312. 4400(2013). |
| [36] | K. M. He, X. Zhang, S. Q. Ren, J. Sun, Deep residual learning for image recognition, Proc. IEEE confer. Computer vis. Pattern recognit., (2016), 770–778. |
| [37] | C. Szegedy, W. Liu, Y. Q. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., Going deeper with convolutions, Proc. IEEE confer. Computer vis. Pattern recognit., (2015), 1–9. |
| [38] | K. Simonyan, A. Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition, arXiv preprint arXiv: 1409. 1556 (2014). |
| [39] | Y. Lecun, Y. Bengio, Convolutional Networks for Images, Speech, and Time-Series, The Handbook of Brain Theory & Neural Networks, 3361 (10), 1995. |
| [40] |
V. Suarez-Paniagua, I. Segura-Bedmar, Evaluation of pooling operations in convolutional architectures for drug-drug interaction extraction, BMC Bioinformatics, 19 (2018), 209. doi: 10.1186/s12859-018-2195-1. doi: 10.1186/s12859-018-2195-1
|
| [41] |
X. L. Zhang, J. F. Xu, J. Yang, L. Chen, H. B. Zhou, X. J. Liu, et al., Understanding the learning mechanism of convolutional neural networks in spectral analysis, Anal Chim Acta, 1119 (2020), 41–51. doi: 10.1016/j.aca.2020.03.055. doi: 10.1016/j.aca.2020.03.055
|
| [42] |
S. W. Kwon, I. J. Choi, J. Y. Kang, W. I. Jang, G. H. Lee, M. C. Lee, Ultrasonographic Thyroid Nodule Classification Using a Deep Convolutional Neural Network with Surgical Pathology, J. Digit. Imaging, 33 (2020), 1202–1208. doi: 10.1007/s10278-020-00362-w. doi: 10.1007/s10278-020-00362-w
|
| [43] |
G. E. Dahl, T. N. Sainath, G. E. Hinton, Improving deep neural networks for LVCSR using rectified linear units and dropout, 2013 IEEE Int. Conf. Acoustics, IEEE, 2013. doi: 10.1109/ICASSP.2013.6639346. doi: 10.1109/ICASSP.2013.6639346
|
| [44] | N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, R. Salakhutdinov, Dropout: A Simple Way to Prevent Neural Networks from Overfitting, J. Mach. Learn. Res., 15 (2014), 1929–1958. |
| [45] | S. Ioffe, C. Szegedy, Batch normalization: Accelerating deep network training by reducing internal covariate shift, Int. Conf. Mach. Learn., PMLR, (2015), pp. 448–456. |
| [46] | V. Nair, G. E. Hinton, Rectified linear units improve restricted boltzmann machines, lcml, 2010. |
| [47] |
P. Li, X. Liu, Bilinear interpolation method for quantum images based on quantum Fourier transform, Int. J. Quantum Inf., 16 (2018), 1850031. doi: 10.1142/S0219749918500314. doi: 10.1142/S0219749918500314
|
| [48] | D. Y. Han, Comparison of commonly used image interpolation methods, Proc. 2nd Int. Conf. Comput. Sci. Electron. Eng. (ICCSEE 2013), 10 (2013). |
| [49] |
X. Wang, X. Jia, W. Zhou, et al., Correction for color artifacts using the RGB intersection and the weighted bilinear interpolation, Appl. Opt., 58 (2019), 8083–8091. doi: 10.1364/AO.58.008083. doi: 10.1364/AO.58.008083
|
| [50] |
J. F. Dou, Q. Qin, Z. M. Tu, Image fusion based on wavelet transform with genetic algorithms and human visual system, Multimed. Tools Appl., 78 (2018), 12491–12517. doi: 10.1007/s11042-018-6756-0. doi: 10.1007/s11042-018-6756-0
|
| [51] |
H. M. Lu, L. F. Zhang, S. Serikawa, Maximum local energy: An effective approach for multisensor image fusion in beyond wavelet transform domain, Comput. Math. Appl. 64 (2012), 996–1003. doi: 10.1016/j.camwa.2012.03.017. doi: 10.1016/j.camwa.2012.03.017
|
| [52] | B. Zhang, Study on image fusion based on different fusion rules of wavelet transform, 2010 3rd Int. Conf. Adv. Comput. Theo. Eng. (ICACTE), Vol. 3. IEEE, 2010. doi: 10.1109/ICACTE.2010.5579586. |
| [53] | S. L. Liu, Z. J. Song, M. N. Wang, WaveFuse: A Unified Deep Framework for Image Fusion with Discrete Wavelet Transform, arXiv preprint arXiv: 2007. 14110(2020). |
| [54] |
D. Kusumoto, M. Lachmann, T. Kunihiro, S. Yuasa, Y. Kishino, M. Kimura, et al., Automated Deep Learning-Based System to Identify Endothelial Cells Derived from Induced Pluripotent Stem Cells, Stem Cell Rep., 10 (2018), 1687–1695. doi: 10.1016/j.stemcr.2018.04.007. doi: 10.1016/j.stemcr.2018.04.007
|
| [55] |
Su. P, Guo. S, Roys. S, F. Maier, H. Bhat, J. Zhuo, et al., Transcranial MR Imaging-Guided Focused Ultrasound Interventions Using Deep Learning Synthesized CT, AJNR Am. J. Neuroradiol., 41 (2020), 1841–1848. doi: 10.3174/ajnr.A6758. doi: 10.3174/ajnr.A6758
|
| [56] |
S. J. Pan, Q. Yang, A Survey on Transfer Learning, IEEE Trans. Knowl. Data Eng., 22 (2010), 1345–1359. doi: 10.1109/TKDE.2009.191. doi: 10.1109/TKDE.2009.191
|
| [57] |
S. Medghalchi, C. F. Kusche, E. Karimi, U. Kerzel, S. K. Kerzel, et al., Damage Analysis in Dual-Phase Steel Using Deep Learning: Transfer from Uniaxial to Biaxial Straining Conditions by Image Data Augmentation, JOM, 72 (2020), 4420–4430. doi: 10.1007/s11837-020-04404-0. doi: 10.1007/s11837-020-04404-0
|
| [58] |
X. R. Yu, X. M. Wu, C. B. Luo, P. Ren, Deep learning in remote sensing scene classification: a data augmentation enhanced convolutional neural network framework, GISci. Remote Sens., 54 (2017), 741–758. doi: 10.1080/15481603.2017.1323377. doi: 10.1080/15481603.2017.1323377
|
| [59] |
A. Taheri-Garavand, H. Ahmadi, M. Omid, S. S. Mohtasebi, K. Mollazade, G. M. Carlomagno, et al., An intelligent approach for cooling radiator fault diagnosis based on infrared thermal image processing technique, Appl. Therm. Eng., 87 (2015), 434–443. doi: 10.1016/j.applthermaleng.2015.05.038. doi: 10.1016/j.applthermaleng.2015.05.038
|
| [60] | M. Drozdzal, E. Vorontsov, G. Chartrand, S. Kadoury, C. Pal, The Importance of Skip Connections in Biomedical Image Segmentation, Deep learning and data labeling for medical applications, Springer, Cham, 2016. 179–187. doi: 10.1007/978-3-319-46976-8_19. |
| [61] |
Y-Lan. Boureau, Bach. F, Y. LeCun, Ponce. J, Learning mid-level features for recognition, 2010 IEEE Computer Society Conf. Comput. Vis. Pattern Recognit., IEEE, (2010), 2559–2566. doi: 10.1109/CVPR.2010.5539963. doi: 10.1109/CVPR.2010.5539963
|
| 1. | I. G. Konkina, S. P. Ivanov, Yu. I. Murinov, Binuclear Mercury(I) Complex with D-Gluconic Acid, 2019, 64, 0036-0236, 201, 10.1134/S0036023619020116 | |
| 2. | Said Majdood Raihan, Mohammad Moniruzzaman, Youngjin Park, Seunghan Lee, Sungchul C. Bai, Evaluation of Dietary Organic and Inorganic Mercury Threshold Levels on Induced Mercury Toxicity in a Marine Fish Model, 2020, 10, 2076-2615, 405, 10.3390/ani10030405 | |
| 3. | Christopher C. Koenig, Felicia C. Coleman, Christopher R. Malinowski, Atlantic Goliath Grouper of Florida: To Fish or Not to Fish, 2020, 45, 0363-2415, 20, 10.1002/fsh.10349 | |
| 4. | Oliver N. Shipley, Cheng-Shiuan Lee, Nicholas S. Fisher, James K. Sternlicht, Sami Kattan, Erica R. Staaterman, Neil Hammerschlag, Austin J. Gallagher, Metal concentrations in coastal sharks from The Bahamas with a focus on the Caribbean Reef shark, 2021, 11, 2045-2322, 10.1038/s41598-020-79973-w | |
| 5. | Agnieszka Jędruch, Magdalena Bełdowska, Mercury forms in the benthic food web of a temperate coastal lagoon (southern Baltic Sea), 2020, 153, 0025326X, 110968, 10.1016/j.marpolbul.2020.110968 | |
| 6. | Patricia Morcillo, María A. Esteban, Alberto Cuesta, Metal detoxification in the marine teleost fish Sparus aurata L. and Dicentrarchus labrax L., 2018, 133, 0025326X, 835, 10.1016/j.marpolbul.2018.06.043 | |
| 7. | Dongmei Xie, Qiliang Chen, Shiling Gong, Jingjing An, Yingwen Li, Xiaolong Lian, Zhihao Liu, Yanjun Shen, John P. Giesy, Exposure of zebrafish to environmentally relevant concentrations of mercury during early life stages impairs subsequent reproduction in adults but can be recovered in offspring, 2020, 229, 0166445X, 105655, 10.1016/j.aquatox.2020.105655 | |
| 8. | Patrícia Pereira, Malgorzata Korbas, Vitória Pereira, Tiziana Cappello, Maria Maisano, João Canário, Armando Almeida, Mário Pacheco, A multidimensional concept for mercury neuronal and sensory toxicity in fish - From toxicokinetics and biochemistry to morphometry and behavior, 2019, 1863, 03044165, 129298, 10.1016/j.bbagen.2019.01.020 | |
| 9. | Ji-Won Jang, Seunghyung Lee, Bong-Joo Lee, Sang-Woo Hur, Maeng-Hyun Son, Kang-Woong Kim, Kyoung-Duck Kim, Hyon-Sob Han, A comparative study of effects of dietary mercuric chloride and methylmercury chloride on growth performance, tissue accumulation, stress and immune responses, and plasma measurements in Korean rockfish, Sebastes schlegeli, 2020, 260, 00456535, 127611, 10.1016/j.chemosphere.2020.127611 | |
| 10. | Zeinab M. El‐Bouhy, Rasha M. Reda, Heba H. Mahboub, Fify N. Gomaa, Chelation of mercury intoxication and testing different protective aspects of Lactococcus lactis probiotic in African catfish , 2021, 1355-557X, 10.1111/are.15227 | |
| 11. | Sergio Fernández-Trujillo, Jhon J. López-Perea, María Jiménez-Moreno, Rosa C. Rodríguez Martín-Doimeadios, Rafael Mateo, Metals and metalloids in freshwater fish from the floodplain of Tablas de Daimiel National Park, Spain, 2021, 208, 01476513, 111602, 10.1016/j.ecoenv.2020.111602 | |
| 12. | Bojian Chen, Shiyuan Dong, Mercury Contamination in Fish and Its Effects on the Health of Pregnant Women and Their Fetuses, and Guidance for Fish Consumption—A Narrative Review, 2022, 19, 1660-4601, 15929, 10.3390/ijerph192315929 | |
| 13. | Amanda L. Jeanson, Dietrich Gotzek, Kosal Mam, Luke Hecht, Patricia Charvet, Simon Eckerström-Liedholm, Steven J. Cooke, Thomas Pool, Vittoria Elliott, Yan Torres, 2022, 9780128220412, 343, 10.1016/B978-0-12-819166-8.00170-5 | |
| 14. | Bonsignore Maria, Messina Concetta Maria, Bellante Antonio, Manuguerra Simona, Arena Rosaria, Santulli Andrea, Maricchiolo Giulia, Del Core Marianna, Sprovieri Mario, Chemical and biochemical responses to sub−lethal doses of mercury and cadmium in gilthead seabream (Sparus aurata), 2022, 307, 00456535, 135822, 10.1016/j.chemosphere.2022.135822 | |
| 15. | Guoxiu Shang, Xiaogang Wang, Long Zhu, Shan Liu, Hongze Li, Zhe Wang, Biao Wang, Zhengxian Zhang, Heavy Metal Pollution in Xinfengjiang River Sediment and the Response of Fish Species Abundance to Heavy Metal Concentrations, 2022, 19, 1660-4601, 11087, 10.3390/ijerph191711087 | |
| 16. | Suman Bhusan Chakraborty, Non-Essential Heavy Metals as Endocrine Disruptors: Evaluating Impact on Reproduction in Teleosts, 2021, 74, 0373-5893, 417, 10.1007/s12595-021-00399-x | |
| 17. | Ulrike Kammann, Marc-Oliver Aust, Maike Siegmund, Nicole Schmidt, Katharina Straumer, Thomas Lang, Deep impact? Is mercury in dab (Limanda limanda) a marker for dumped munition? Results from munition dump site Kolberger Heide (Baltic Sea), 2021, 193, 0167-6369, 10.1007/s10661-021-09564-3 | |
| 18. | E.M. Nalley, C.M. Pirkle, M.C. Schmidbauer, C.J. Lewis, R.S. Dacks, M.D. Thompson, M.D. Sudnovsky, J.L. Whitney, M.J. Donahue, Trophic and spatial patterns of contaminants in fishes from the Republic of the Marshall Islands in the equatorial Pacific, 2023, 314, 00456535, 137593, 10.1016/j.chemosphere.2022.137593 | |
| 19. | Eduardo Franco-Fuentes, Nicolas Moity, Jorge Ramírez-González, Solange Andrade-Vera, Arturo Hardisson, Soraya Paz, Carmen Rubio, Verónica Martín, Ángel J. Gutiérrez, Mercury in fish tissues from the Galapagos marine reserve: Toxic risk and health implications, 2023, 115, 08891575, 104969, 10.1016/j.jfca.2022.104969 | |
| 20. | Hsiang-Chieh Chuang, Huai-Ting Huang, Novi-Rosmala Dewi, Hsi-Hua Hsiao, Bo-Ying Chen, Zhen-Hao Liao, Meng-Chou Lee, Po-Tsang Lee, Yu-Sheng Wu, Yu-Ju Lin, Fan-Hua Nan, Effect of Methylmercury Exposure on Bioaccumulation and Nonspecific Immune Respsonses in Hybrid Grouper Epinephelus fuscoguttatus × Epinephelus lanceolatus, 2022, 12, 2076-2615, 147, 10.3390/ani12020147 | |
| 21. | Annette F. Muttray, Derek C.G. Muir, Gerald R. Tetreault, Mark E. McMaster, James P. Sherry, Spatial trends and temporal declines in tissue metals/metalloids in the context of wild fish health at the St. Clair River Area of Concern, 2021, 47, 03801330, 900, 10.1016/j.jglr.2021.02.007 | |
| 22. | Shubhajit Saha, Kishore Dhara, Azubuike V. Chukwuka, Prasenjit Pal, Nimai Chandra Saha, Caterina Faggio, Sub-lethal acute effects of environmental concentrations of inorganic mercury on hematological and biochemical parameters in walking catfish, Clarias batrachus, 2023, 264, 15320456, 109511, 10.1016/j.cbpc.2022.109511 | |
| 23. | Je-Won Yoo, Hyeon-Jeong Bae, Min Jeong Jeon, Tae-Yong Jeong, Young-Mi Lee, Metabolomic analysis of combined exposure to microplastics and methylmercury in the brackish water flea Diaphanosoma celebensis, 2022, 0269-4042, 10.1007/s10653-022-01435-1 | |
| 24. | Jérémy Lemaire, François Brischoux, Oliver Marquis, Rosanna Mangione, Stéphane Caut, Maud Brault-Favrou, Carine Churlaud, Paco Bustamante, Relationships between stable isotopes and trace element concentrations in the crocodilian community of French Guiana, 2022, 837, 00489697, 155846, 10.1016/j.scitotenv.2022.155846 | |
| 25. | Olof Regnell, Sylvie V. M. Tesson, Nikolay Oskolkov, Michelle Nerentorp, Mercury–Selenium Accumulation Patterns in Muscle Tissue of Two Freshwater Fish Species, Eurasian Perch (Perca fluviatilis) and Vendace (Coregonus albula), 2022, 233, 0049-6979, 10.1007/s11270-022-05709-3 | |
| 26. | Ulrike Kammann, Jan-Dag Pohlmann, Fatima Wariaghli, Hajar Bourassi, Klara Regelsberger, Ahmed Yahyaoui, Reinhold Hanel, Heavy metal contamination in European conger (Conger conger, Linnaeus 1758) along the coastline of Morocco, 2022, 34, 2190-4707, 10.1186/s12302-022-00694-0 | |
| 27. | Kamila Pokorska-Niewiada, Agata Witczak, Mikołaj Protasowicki, Jacek Cybulski, Estimation of Target Hazard Quotients and Potential Health Risks for Toxic Metals and Other Trace Elements by Consumption of Female Fish Gonads and Testicles, 2022, 19, 1660-4601, 2762, 10.3390/ijerph19052762 | |
| 28. | Jérémy Lemaire, Paco Bustamante, Rosanna Mangione, Olivier Marquis, Carine Churlaud, Maud Brault-Favrou, Charline Parenteau, François Brischoux, Lead, mercury, and selenium alter physiological functions in wild caimans (Caiman crocodilus), 2021, 286, 02697491, 117549, 10.1016/j.envpol.2021.117549 | |
| 29. | Carmen G. Montaña, Elford Liverpool, Donald C. Taphorn, Christopher M. Schalk, The cost of gold: Mercury contamination of fishes in a Neotropical river food web, 2021, 19, 1982-0224, 10.1590/1982-0224-2020-0155 | |
| 30. | Katrina K. Knott, Emma Coleman, Jacob A. Cianci–Gaskill, Rebecca O’Hearn, Darby Niswonger, John D. Brockman, Alba Argerich, Rebecca North, Jeff Wenzel, Mercury, selenium, and fatty acids in the axial muscle of largemouth bass: evaluating the influence of seasonal and sexual changes in fish condition and reproductive status, 2022, 31, 0963-9292, 761, 10.1007/s10646-022-02544-4 | |
| 31. | Antonio Belmonte, Pilar Muñoz, Juan Santos-Echeandía, Diego Romero, Tissue Distribution of Mercury and Its Relationship with Selenium in Atlantic Bluefin Tuna (Thunnus thynnus L.), 2021, 18, 1660-4601, 13376, 10.3390/ijerph182413376 | |
| 32. | Ulrike Kammann, Pedro Nogueira, Maike Siegmund, Nicole Schmidt, Stefan Schmolke, Torben Kirchgeorg, Matthias Hasenbein, Klaus Wysujack, Temporal trends of mercury levels in fish (dab, Limanda limanda) and sediment from the German Bight (North Sea) in the period 1995–2020, 2023, 195, 0167-6369, 10.1007/s10661-022-10655-y | |
| 33. | Fernando Morgado, Ruy M. A. L. Santos, Daniela Sampaio, Luiz Drude de Lacerda, Amadeu M. V. M. Soares, Hugo C. Vieira, Sizenando Abreu, Chronological Trends and Mercury Bioaccumulation in an Aquatic Semiarid Ecosystem under a Global Climate Change Scenario in the Northeastern Coast of Brazil, 2021, 11, 2076-2615, 2402, 10.3390/ani11082402 | |
| 34. | Shriya Garg, Mangesh Gauns, 2023, 9780323959193, 195, 10.1016/B978-0-323-95919-3.00011-2 | |
| 35. | Joanna Łuczyńska, Marek Jan Łuczyński, Joanna Nowosad, Monika Kowalska-Góralska, Magdalena Senze, Total Mercury and Fatty Acids in Selected Fish Species on the Polish Market: A Risk to Human Health, 2022, 19, 1660-4601, 10092, 10.3390/ijerph191610092 | |
| 36. | Giovanni Denaro, Luciano Curcio, Alessandro Borri, Laura D'Orsi, Andrea De Gaetano, A dynamic integrated model for mercury bioaccumulation in marine organisms, 2023, 75, 15749541, 102056, 10.1016/j.ecoinf.2023.102056 | |
| 37. | Vicki S. Blazer, Heather L. Walsh, Adam J. Sperry, Brenna Raines, James J. Willacker, Collin A. Eagles-Smith, A multi-level assessment of biological effects associated with mercury concentrations in smallmouth bass, Micropterus dolomieu, 2023, 02697491, 121688, 10.1016/j.envpol.2023.121688 | |
| 38. | Bambang Yulianto, Agoes Soegianto, Moch Affandi, Carolyn Melissa Payus, The impact of various periods of mercury exposure on the osmoregulatory and blood gas parameters of tilapia (Oreochromis niloticus), 2023, 9, 24056650, 100244, 10.1016/j.emcon.2023.100244 | |
| 39. | Giun-Yi Hung, Yu-Chin Pan, Jiun-Lin Horng, Li-Yih Lin, Sublethal effects of methylmercury on lateral line sensory and ion-regulatory functions in zebrafish embryos, 2023, 271, 15320456, 109700, 10.1016/j.cbpc.2023.109700 | |
| 40. | Jaehyuk Lee, Seyeong Lee, Jihee Kim, Zahid Hanif, Seunghee Han, Sukwon Hong, Myung‐Han Yoon, Solution‐based Sulfur‐Polymer Coating on Nanofibrillar Films for Immobilization of Aqueous Mercury Ions, 2018, 39, 1229-5949, 84, 10.1002/bkcs.11350 | |
| 41. | Jérémy Lemaire, Using Crocodylians for monitoring mercury in the tropics, 2023, 32, 0963-9292, 977, 10.1007/s10646-023-02703-1 | |
| 42. | David C. P. King, Michael J. Watts, Elliott M. Hamilton, Robert J. G. Mortimer, Mike Coffey, Odipo Osano, Maureene Auma Ondayo, Marcello Di Bonito, Field method for preservation of total mercury in waters, including those associated with artisanal scale gold mining, 2024, 1759-9660, 10.1039/D3AY02216A | |
| 43. | David C. Evers, Joshua T. Ackerman, Staffan Åkerblom, Dominique Bally, Nil Basu, Kevin Bishop, Nathalie Bodin, Hans Fredrik Veiteberg Braaten, Mark E. H. Burton, Paco Bustamante, Celia Chen, John Chételat, Linroy Christian, Rune Dietz, Paul Drevnick, Collin Eagles-Smith, Luis E. Fernandez, Neil Hammerschlag, Mireille Harmelin-Vivien, Agustin Harte, Eva M. Krümmel, José Lailson Brito, Gabriela Medina, Cesar Augusto Barrios Rodriguez, Iain Stenhouse, Elsie Sunderland, Akinori Takeuchi, Tim Tear, Claudia Vega, Simon Wilson, Pianpian Wu, Global mercury concentrations in biota: their use as a basis for a global biomonitoring framework, 2024, 0963-9292, 10.1007/s10646-024-02747-x | |
| 44. | Daniel Esteban Romero-Suárez, Liseth Pérez-Flórez , Adolfo Consuegra-Solórzano, Jhon Vidal-Durango , Jorge Buelvas-Soto, José Marrugo-Negrete, Mercurio total (Hg-T) en ictiofauna de mayor consumo en San Marcos - Sucre, Colombia, 2024, 27, 1909-0544, e2488, 10.21897/rmvz.2488 | |
| 45. | Heri Budi SANTOSO, Rizmi YUNITA , KRISDIANTO KRISDIANTO, Assessing the Health of South Kalimantan Coastal Swamp Wetlands using Measurements of Heavy Metals in Commercial Fish Species, 2024, 15, 2067533X, 1095, 10.36868/IJCS.2024.02.23 | |
| 46. | Viacheslav V. Krylov, Irina L. Golovanova, Andrey A. Filippov, Elena A. Osipova, Ekaterina A. Kulivatskaya, Effects of mercury and magnetic fields on the activity of proteinases and glycosidases in the intestine of common carp Cyprinus carpio, 2024, 196, 0167-6369, 10.1007/s10661-024-13274-x | |
| 47. | Marcin Pigłowski, Alberto Nogales, Maria Śmiechowska, Hazards in Products from Northern Mediterranean Countries Reported in the Rapid Alert System for Food and Feed (RASFF) in 1997–2021 in the Context of Sustainability, 2025, 17, 2071-1050, 889, 10.3390/su17030889 | |
| 48. | Donald T.A. Tapfuma, Desmond Mwembe, Yogeshkumar Naik, Non-lethal method for the assessment of bioavailable metals in aquatic ecosystems surrounding ASGM activity, 2025, 138, 14747065, 103874, 10.1016/j.pce.2025.103874 | |
| 49. | Guy Sisma-Ventura, Yael Segal, Yaron Gertner, Maya Mar Mori, Maria Abu Hadra, Eli Biton, Aviv Shachnai, Barak Herut, Long-term (1979–2024) trends and remobilization process of mercury pollution, the case study of Haifa Bay, Southeast Mediterranean Sea, 2025, 490, 03043894, 137760, 10.1016/j.jhazmat.2025.137760 | |
| 50. | Clarisse Seguin, Alice Marant, Séverine Palacios-Paris, Isabelle Bonnard, Jean-Luc Loizeau, Elise David, Damien Rioult, Claudia Cosio, Unveiling the hidden threat: Molecular, cellular and behavioral effects of dietborne inorganic mercury and methylmercury in Dreissena polymorpha, 2025, 376, 00456535, 144306, 10.1016/j.chemosphere.2025.144306 |
Figures(19) / Tables(4)
Xiaochen Liu, Weidong He, Yinghui Zhang, Shixuan Yao, Ze Cui. Effect of dual-convolutional neural network model fusion for Aluminum profile surface defects classification and recognition[J]. Mathematical Biosciences and Engineering, 2022, 19(1): 997-1025. doi: 10.3934/mbe.2022046
| Specie | Mercury content (ppm) | Safety |
| Anchovies | 0.017 ± 0.015 | Eco-good |
| Atlantic cod | 0.095 ± 0.080 | Eco-bad |
| Bass (saltwater, black, striped, rockfish) | 0.167 ± 0.194 | Eco-bad |
| Bass Chilean | 0.354 ± 0.199 | Eco-bad |
| Carp | 0.140 ± 0.099 | Eco-bad |
| Catfish | 0.049 ± 0.084 | Eco-good |
| Croaker Atlantic (Atlantic) | 0.069 ± 0.049 | Eco-good |
| Croaker White (Pacific) | 0.287 ± 0.069 | Eco-bad |
| Grouper (all species) | 0.448 ± 0.287 | Eco-bad |
| Mullet | 0.05 ± 0.078 | Eco-bad |
| Salmon, wild (Alaska) | 0.014 ± 0.041 | Eco-good |
| Sardines, Pacific (US) | 0.016 ± 0.007 | Eco-good |
| Shark | 0.979 ± 0.626 | Eco-bad |
| Swordfish | 0.976 ± 0.510 | Eco-bad |
| Tilapia | 0.013 ± 0.023 | Eco-good |
| Trout, rainbow (farmed, freshwater) | 0.072 ± 0.143 | Eco-good |
| Tuna species | 0.415 ± 0.308 | Eco-bad |
DownLoad: CSV| Specie | Mercury content (ppm) | Safety |
| Anchovies | 0.017 ± 0.015 | Eco-good |
| Atlantic cod | 0.095 ± 0.080 | Eco-bad |
| Bass (saltwater, black, striped, rockfish) | 0.167 ± 0.194 | Eco-bad |
| Bass Chilean | 0.354 ± 0.199 | Eco-bad |
| Carp | 0.140 ± 0.099 | Eco-bad |
| Catfish | 0.049 ± 0.084 | Eco-good |
| Croaker Atlantic (Atlantic) | 0.069 ± 0.049 | Eco-good |
| Croaker White (Pacific) | 0.287 ± 0.069 | Eco-bad |
| Grouper (all species) | 0.448 ± 0.287 | Eco-bad |
| Mullet | 0.05 ± 0.078 | Eco-bad |
| Salmon, wild (Alaska) | 0.014 ± 0.041 | Eco-good |
| Sardines, Pacific (US) | 0.016 ± 0.007 | Eco-good |
| Shark | 0.979 ± 0.626 | Eco-bad |
| Swordfish | 0.976 ± 0.510 | Eco-bad |
| Tilapia | 0.013 ± 0.023 | Eco-good |
| Trout, rainbow (farmed, freshwater) | 0.072 ± 0.143 | Eco-good |
| Tuna species | 0.415 ± 0.308 | Eco-bad |
