•Medical image segmentation typically faces limited datasets.•Dataset limitations are broadly grouped into scarce and weak annotations.•Scarce annotations can be addressed proactively via ...cost-effective annotation or by leveraging external labeled or unlabeled datasets.•Scarce annotations can be addressed reactively via various forms of conditional random fields as post processing.•Weak annotations may manifest as sparse, noisy, or only image-level labels.•Sparse and noisy annotations can leveraged via selective and noise-resilient loss functions, respectively.•Image-level labels can leveraged via various forms of class activation maps.•Recommended solutions based on a cost-gain trade-off are provided.
Display omitted
The medical imaging literature has witnessed remarkable progress in high-performing segmentation models based on convolutional neural networks. Despite the new performance highs, the recent advanced segmentation models still require large, representative, and high quality annotated datasets. However, rarely do we have a perfect training dataset, particularly in the field of medical imaging, where data and annotations are both expensive to acquire. Recently, a large body of research has studied the problem of medical image segmentation with imperfect datasets, tackling two major dataset limitations: scarce annotations where only limited annotated data is available for training, and weak annotations where the training data has only sparse annotations, noisy annotations, or image-level annotations. In this article, we provide a detailed review of the solutions above, summarizing both the technical novelties and empirical results. We further compare the benefits and requirements of the surveyed methodologies and provide our recommended solutions. We hope this survey article increases the community awareness of the techniques that are available to handle imperfect medical image segmentation datasets.
The continuous developments of urban and industrial environments have increased the demand for intelligent video surveillance. Deep learning has achieved remarkable performance for anomaly detection ...in surveillance videos. Previous approaches achieve anomaly detection with a single-pretext task (image reconstruction or prediction) and detect anomalies by larger reconstruction error or poor prediction. However, they cannot fully exploit the discriminative semantics and temporal context information. Moreover, tackling anomaly detection with a single pretext task is suboptimal due to the nonalignment between the pretext task and anomaly detection. In this article, we propose a temporal-aware contrastive network (TAC-Net) to address the abovementioned problems of anomaly detection for intelligence video surveillance. TAC-Net is an unsupervised method that utilizes deep contrastive self-supervised learning to capture the high-level semantic features and tackles anomaly detection with multiple self-supervised tasks. During inference phase, the multiple task losses and contrastive similarity are utilized to calculate the anomaly score. Experimental results show that our method is superior to state-of-the-art approaches on three benchmarks, which demonstrates the validity and advancement of TAC-Net.
Recent studies suggest that moderate red wine consumption may confer several health benefits, including protection against heart disease, certain cancers and multiple age-related neurological ...diseases such as Alzheimer's disease. These health benefits are assumed to come from a compound from grape skin called resveratrol, which has been proposed to be a pro-longevity agent. Whether resveratrol accounts for all the health benefits of grape-derived nutrients and the molecular and cellular mechanisms underlying the beneficial effects of such nutrients are not well understood. Here we investigated the effect of supplementing grape skin extract (GSE) left from red wine-production process to the daily food intake of a Drosophila melanogaster model of Parkinson's disease (PD) associated with PTEN-induced kinase 1 (PINK1) loss-of-function. Consumption of GSE resulted in rescue of mitochondrial morphological defects, improvement of indirect flight muscle function and health-span, and prolonged lifespan of the PINK1 mutant flies. Further biochemical and genetic studies revealed a link between activation of mitophagy and the beneficial effects of GSE. Our results indicate that GSE can promote autophagy activation, preserve mitochondria function, and protect against PD pathogenesis, and that the beneficial effect of GSE in mitophagy activation is not accounted for by resveratrol alone.
•GSE rescued muscle degeneration and extended lifespan of the fly PINK1 PD model.•GSE exerted its protective effects through activation of mitophagy.•Resveratrol may only partially account for the beneficial effects of GSE.
A new sediment P release risk index (SPRRI) for “in-situ” phosphorus (P) release risk in lake sediment, is developed based on diffusive gradients in thin films (DGT) technique, DGT induced flux in ...sediments (DIFS) model and sediment properties. SPRRI includes three sub-indexes, which contain (1) the labile P pool size, (2) resupply constant (r) and desorption rate (Dspt rate) for P transfer and (3) the molar ratio between iron (Fe) in sequential extraction for sediment P by bicarbonate-dithionite (BD) and aluminum (Al) by NaOH (at 25 °C), i.e. BD(Fe)/AlNaOH25 in sediment solid. The first sub-index considers P release from (i) sediment with NH4Cl-P+BD-P pool, i.e. the loosely sorbed P (NH4Cl-P) plus iron associated P (BD-P), or (ii) sediment with NH4Cl-P pool, respectively. The second and third sub-indexes reflect kinetic P desorption and resupply ability of solid phase, and the effect of P sequestration by Al hydroxide on P release, in turn. The inner relationship between SPRRI and sub-indexes, and their effects on P release risk are elucidated. SPRRI can be used to evaluate sediment P reactivity by five release risk ranks. For Lake Dianchi (China), P transfer dynamics, labile P pool, resupply ability and Al-P in sediment, and “external P-loading” control and affect P release risk in different regions, which is reflected by the spatial distribution map for SPRRI. The present SPRRI can be applied for lakes with (1) pH range varying from moderate acidity to weak alkalinity in waterbody and (2) NH4Cl-P or NH4Cl-P+BD-P pool in sediment solid.
Display omitted
•A new sediment P release risk index (SPRRI) for lake sediment has been designed.•DIFS model and sediment properties have been combined to derive SPRRI.•SPRRI considers not only P fraction and Al, but also “in-situ” dynamic P transfer.•P release risk for NH4Cl-P or NH4Cl-P+BD-P sediment pool can be evaluated.•SPRRI can be applied for pH from moderate acidity to weak alkalinity in water.
A new assessment method for P release risk in lake sediment has been developed based on DIFS model and sediment properties.
Calcium (Ca2+) homeostasis is essential for neuronal function and survival. Altered Ca2+ homeostasis has been consistently observed in neurological diseases. How Ca2+ homeostasis is achieved in ...various cellular compartments of disease-relevant cell types is not well understood. Here we show in Drosophila Parkinson’s disease (PD) models that Ca2+ transport from the endoplasmic reticulum (ER) to mitochondria through the ER–mitochondria contact site (ERMCS) critically regulates mitochondrial Ca2+ (mito-Ca2+) homeostasis in dopaminergic (DA) neurons, and that the PD-associated PINK1 protein modulates this process. In PINK1 mutant DA neurons, the ERMCS is strengthened and mito-Ca2+ level is elevated, resulting in mitochondrial enlargement and neuronal death. Miro, a well-characterized component of the mitochondrial trafficking machinery, mediates the effects of PINK1 on mito-Ca2+ and mitochondrial morphology, apparently in a transport-independent manner. Miro overexpression mimics PINK1 loss-of-function effect, whereas inhibition of Miro or components of the ERMCS, or pharmacological modulation of ERMCS function, rescued PINK1 mutant phenotypes. Mito-Ca2+ homeostasis is also altered in the LRRK2-G2019S model of PD and the PAR-1/MARK model of neurodegeneration, and genetic or pharmacological restoration of mito-Ca2+ level is beneficial in these models. Our results highlight the importance of mito-Ca2+ homeostasis maintained by Miro and the ERMCS to mitochondrial physiology and neuronal integrity. Targeting this mito-Ca2+ homeostasis pathway holds promise for a therapeutic strategy for neurodegenerative diseases.
Item co-occurrence is an important pattern in recommendation. Due to the difference in correlation, the matching degrees between the target and historical items vary. The higher the matching degree, ...the greater probability they co-occur. Recently, the recommendation performance has been greatly improved by leveraging item relations. As an important bond imposed by relations, these connected items should have a strong correlation in the calculation of certain measures. This kind of correlation can be the biased knowledge that benefits parameter training. Specifically, we focus on tuples containing the target item and latest relational items that have relations such as complement or substitute with the target item in user's behavior sequence. Such close relations mean the matching degrees between relational items and historical items should be highly affected by that of the target item and historical items. For example, given a relational item having relation complement with the target item, if the target item has high matching degrees with some items in user's behavior sequence, this complementary item should behave similarly for the co-occurrence of complementary items. Under guidance of the above thought, in this work, we propose a target-relation regulated mechanism which converts the biased knowledge of high correlation of matching degrees into a regulation. It integrates the target item and relational items in history as a whole to characterize the matching score between the target item and historical items. Experiments conducted on three real-world datasets demonstrate that our model can significantly outperform a set of state-of-the-art models.
Abstract The neurotoxicity of aluminum (Al) − the most abundant metal element on earth − has been known for years. However, the mechanism of Al-induced neurodegeneration and its relationship to ...Alzheimer's disease are still controversial. In particular, in vivo functional data are lacking. In a Drosophila model with chronic dietary Al overloading, general neurodegeneration and several behavioral changes were observed. Al-induced neurodegeneration is independent of β-amyloid or tau-associated toxicity, suggesting they act in different molecular pathways. Interestingly, Drosophila frataxin ( dfh ), which causes Friedreich's ataxia if mutated in humans, displayed an interacting effect with Al, suggesting Friedreich's ataxia patients might be more susceptible to Al toxicity. Al-treated flies accumulated large amount of iron and reactive oxygen species (ROS), and exhibited elevated SOD2 activity. Genetic and pharmacological efforts to reduce ROS or chelate excess Fe significantly mitigated Al toxicity. Our results indicate that Al toxicity is mediated through ROS production and iron accumulation and suggest a remedial route to reduce toxicity due to Al exposure.
There is a lack of the understanding of the transfer of phosphate (PO4-P), ammonium (NH4-N) and nitrate (NO3-N) at the millimeter scale in sediments related to internal loadings in the semi-arid and ...cold Lake Hulun in China. In this paper, a passive sampling technique-diffusion gradients in thin films (DGT) has been used for the biogeochemical behaviors and dynamic transfer of PO4-P, NH4-N and NO3-N in sediments in this lake. The two-sided DGT probe (AMP-TH&ZrO-Chelex) and ZrO DGT piston are designed for sediment/water interface. Based on change characters of DGT concentration profiles (nutrients, Fe and Mn) related to redox status and chemical properties, N and P mobilizations are derived. The reductive dissolution of Mn(IV) or Fe(III) (oxyhydr)oxides accounts for subsequent P release is corresponded with the oxic or anoxic status in sediment layers; Fe(III)-reduction is the main reason for P remobilization. The breakdown of algae biomass deposited on the uppermost sediment and phosphatase enzymes activity enhanced by NH4-N also lead to P remobilization. In sub-anoxic or anoxic layer, denitrification and dissimilatory nitrate reduction to ammonium (DNRA) induce exchangeable ammonium (NH4ex) enrichment and NH4-N release, while, anammox depletes NH4-N mobility. Aerobic nitrification results in NO3-N remobilization. Organic matter favors NH4-N release. DGT induced fluxes in sediments (DIFS) model is used to derive the labile P pool (51.5–465.5 cm3 g−1), resupply constant (0.186–0.734), response time (54–11,070 s) and Dspt rate (32–382,517 nmol cm−3 d−1) for describing dynamic P transfers with two opposite types of “slow” and “fast” resupplies. The diffusive fluxes across sediment/water interface in ranges of 67.9–498.7 μg m−2 d−1 (PO4-P), −1416–106.8 (NH4-N) and –2147–301.1 μg m−2 d−1 (NO3-N) and DIFS parameters can be used for release risks in lake regions and the sediment remediation strategies are recommended.
Display omitted
•AMP-TH&ZrO-Chelex DGT probe and ZrO piston are used for nutrients in lake sediment.•P release is mainly due to Fe reduction and then Mn affected by redox status.•DNRA, anammox, nitrification and organic matter control NH4-N and NO3-N mobility.•Labile P pool, resupply constant and desorption rate are evaluated by DIFS model.•Lake management strategies are proposed based on diffusive flux and DIFS parameters.
Nitrogen (N), in the form of ammonia or nitrate, is a key limiting nutrient in many aquatic systems. Under certain environmental conditions it can be released from sediments into overlying water, ...which may have significant impact on water quality and result in continuous eutrophication. However, few studies have examined the long-term (nearly two months) coupling effect of environmental parameters on N dynamics at the sediment–water interface. This is particularly pertinent to improve the understanding of lake eutrophication processes. This study examines the coupling effects of pH and dissolved oxygen (DO) on N release at the sediment–water interface for the shallow Erhai Lake in China, and analyzes recent changes in environmental conditions and water quality to predict the risk of nitrogen release from sediment in the near future. Experimental results indicated that under anaerobic condition (DO<1 mg/L) and lower pH (pH = 6), ammonium was easily released into overlying water, potentially triggering algal blooms. Conversely aerobic conditions (DO = 8–10 mg/L) and higher pH (pH = 10) promoted nitrate release from sediment. The study also discusses possible mechanisms about the nitrogen dynamics at the sediment–water interface. Considering the overall effects of ammonium and nitrate on the trophic status of the water column, the recommended environmental condition in overlying water should be pH of around 8 under aerobic conditions. Based on the study findings, the nitrogen balance at the water–sediment interface was evaluated for different environmental conditions. Analysis of environmental conditions and water quality during 1992–2010 shows that present environmental conditions are not conducive to the release of nutrients from sediment, thereby protecting the water quality from serious endogenous pollution. However, the risk of nitrogen release from sediment sources might increase if environmental conditions change.
•pH, DO, and N dynamics at sediment–water interface are key to lake eutrophication.•Dissolved oxygen (DO) and pH affect nitrogen (N) release at sediment–water interface.•High-pH aerobic conditions promoted nitrate release from sediment.•Low-pH anaerobic conditions release ammonium, potentially triggering algal blooms.•The risk of N release from sediment can increase if environmental conditions change.
Translation of mRNAs is tightly regulated and constantly surveyed for errors. Aberrant translation can trigger co-translational protein and RNA quality control processes, impairments of which cause ...neurodegeneration by still poorly understood mechanism(s). Here we show that quality control of translation of mitochondrial outer membrane (MOM)-localized mRNA intersects with the turnover of damaged mitochondria, both orchestrated by the mitochondrial kinase PINK1. Mitochondrial damage causes stalled translation of complex-I 30 kDa subunit (C-I30) mRNA on MOM, triggering the recruitment of co-translational quality control factors Pelo, ABCE1, and NOT4 to the ribosome/mRNA-ribonucleoprotein complex. Damage-induced ubiquitination of ABCE1 by NOT4 generates poly-ubiquitin signals that attract autophagy receptors to MOM to initiate mitophagy. In the Drosophila PINK1 model, these factors act synergistically to restore mitophagy and neuromuscular tissue integrity. Thus ribosome-associated co-translational quality control generates an early signal to trigger mitophagy. Our results have broad therapeutic implications for the understanding and treatment of neurodegenerative diseases.
Display omitted
•Enhanced translation of OXPHOS-related mRNAs on mitochondrial surface under stress•Severe mitochondrial damage induces translational stalling on mitochondrial surface•Stalled ribosomes recruit co-translational quality control machinery•Remodeling of stalled ribosome/mRNP generates signals that trigger mitophagy
Removal of damaged mitochondria is essential for maintaining cellular vitality, but the earliest signal that initiates the mitophagy process is not well defined. Wu et al. show that mitochondrial damage causes stalled translation of OXPHOS-related mRNAs on the mitochondrial surface. Co-translational quality control of stalled ribosomes generates ubiquitin-containing signals that trigger mitophagy.