Wearable, Machine Washable, Breathable Polyethylenimine/Sodium Alginate Layer-by-Layer-Coated Cotton-Based Multifunctional Triboelectric Nanogenerators.

Cotton-based textiles are ubiquitous in daily life and are prime candidates for application in wearable triboelectric nanogenerators. However, pristine cotton is vulnerable to bacterial attack, lacks antioxidant and ultraviolet (UV)-protective abilities, and shows lower triboelectric charge generation against tribonegative materials because it is present in the neutral region of the triboelectric series. To overcome such drawbacks, herein, a facile layer-by-layer method is proposed, involving the deposition of alternate layers of polyethylenimine (PEI) and sodium alginate (SA) on cotton. Such modified fabric remains breathable and flexible, retains its comfort properties, and simultaneously shows multifunctionalities and improved triboelectric output, which are retained even after 50 home laundering cycles. Also, the modified fabric becomes more tribopositive than nylon, silk, and wool. A triboelectric nanogenerator consisting of modified cotton and polyester fabric is proposed that shows a maximum power density of 338 mW/m2. An open-circuit voltage of ∼97.3 V and a short-circuit current of ∼4.59 µA are obtained under 20 N force and 1 Hz tapping frequency. Further, the modified cotton exhibits excellent antibacterial, antioxidant, and UV-protective properties because of the incorporation of PEI, and its moisture management properties are retained due to the presence of sodium alginate in the layer. This study provides a simple yet effective approach to obtaining durable multifunctionalities and improved triboelectric performance in cotton substrates.

Attention De-sparsification Matters: Inducing diversity in digital pathology representation learning.

We propose DiRL, a Diversity-inducing Representation Learning technique for histopathology imaging. Self-supervised learning (SSL) techniques, such as contrastive and non-contrastive approaches, have been shown to learn rich and effective representations of digitized tissue samples with limited pathologist supervision. Our analysis of vanilla SSL-pretrained models' attention distribution reveals an insightful observation: sparsity in attention, i.e, models tends to localize most of their attention to some prominent patterns in the image. Although attention sparsity can be beneficial in natural images due to these prominent patterns being the object of interest itself, this can be sub-optimal in digital pathology; this is because, unlike natural images, digital pathology scans are not object-centric, but rather a complex phenotype of various spatially intermixed biological components. Inadequate diversification of attention in these complex images could result in crucial information loss. To address this, we leverage cell segmentation to densely extract multiple histopathology-specific representations, and then propose a prior-guided dense pretext task, designed to match the multiple corresponding representations between the views. Through this, the model learns to attend to various components more closely and evenly, thus inducing adequate diversification in attention for capturing context-rich representations. Through quantitative and qualitative analysis on multiple tasks across cancer types, we demonstrate the efficacy of our method and observe that the attention is more globally distributed.

Toyota Smarthome Untrimmed: Real-World Untrimmed Videos for Activity Detection.

Designing activity detection systems that can be successfully deployed in daily-living environments requires datasets that pose the challenges typical of real-world scenarios. In this paper, we introduce a new untrimmed daily-living dataset that features several real-world challenges: Toyota Smarthome Untrimmed (TSU). TSU contains a wide variety of activities performed in a spontaneous manner. The dataset contains dense annotations including elementary, composite activities and activities involving interactions with objects. We provide an analysis of the real-world challenges featured by our dataset, highlighting the open issues for detection algorithms. We show that current state-of-the-art methods fail to achieve satisfactory performance on the TSU dataset. Therefore, we propose a new baseline method for activity detection to tackle the novel challenges provided by our dataset. This method leverages one modality (i.e. optic flow) to generate the attention weights to guide another modality (i.e RGB) to better detect the activity boundaries. This is particularly beneficial to detect activities characterized by high temporal variance. We show that the method we propose outperforms state-of-the-art methods on TSU and on another popular challenging dataset, Charades.

Density-dependent selection in Drosophila: evolution of egg size and hatching time.

Many different laboratory studies of adaptation to larval crowding in Drosophila spp. have all yielded the evolution of preadult competitive ability, even though the ecological context in which crowding was experienced varied across studies. However, the evolution of competitive ability was achieved through different suites of traits in studies wherein crowding was imposed in slightly different ways. Earlier studies showed the evolution of increased competitive ability via increased larval feeding rate and tolerance to nitrogenous waste, at the cost of food to biomass conversion efficiency. However, more recent studies, with crowding imposed at relatively low food levels, showed the evolution of competitive ability via decreased larval development time and body size, and an increase in the time efficiency of conversion of food to biomass, with no change in larval feeding rate or waste tolerance. Taken together, these studies have led to a more nuanced understanding of how the specific details of larval numbers, food amounts etc. can affect which traits evolve to confer increased competitive ability. Here, we report results from a study in which egg size and hatching time were assayed on three sets of populations adapted to larval crowding experienced in slightly different ways, as well as their low density ancestral control populations. Egg size and hatching time are traits that may provide larvae with initial advantages under crowding through increased starting larval size and a temporal head-start, respectively. In each set of populations adapted to some form of larval crowding, the evolution of longer and wider eggs was seen, compared to controls, thus making egg size the first consistent correlate of the evolution of increased larval competitive ability across Drosophila populations experiencing crowding in slightly different ways. Among the crowding-adapted populations, those crowded at the lowest overall eggs/food density, but the highest density of larvae in the feeding band, showed the largest eggs, on an average. All three sets of crowding-adapted populations showed shorter average egg hatching time than controls, but the difference was significant only in the case of populations experiencing the highest feeding band density. Our results underscore the importance of considering factors other than just eggs/food density when studying the evolution of competitive ability, as also the advantages of having multiple selection regimes within one experimental set up, allowing for a more nuanced understanding of the subtlety with which adaptive evolutionary trajectories can vary across even fairly similar selection regimes.

VPN++: Rethinking Video-Pose Embeddings for Understanding Activities of Daily Living.

Many attempts have been made towards combining RGB and 3D poses for the recognition of Activities of Daily Living (ADL). ADL may look very similar and often necessitate to model fine-grained details to distinguish them. Because the recent 3D ConvNets are too rigid to capture the subtle visual patterns across an action, this research direction is dominated by methods combining RGB and 3D Poses. But the cost of computing 3D poses from RGB stream is high in the absence of appropriate sensors. This limits the usage of aforementioned approaches in real-world applications requiring low latency. Then, how to best take advantage of 3D Poses for recognizing ADL? To this end, we propose an extension of a pose driven attention mechanism: Video-Pose Network (VPN), exploring two distinct directions. One is to transfer the Pose knowledge into RGB through a feature-level distillation and the other towards mimicking pose driven attention through an attention-level distillation. Finally, these two approaches are integrated into a single model, we call VPN++. It is worth noting that VPN++ exploits the pose embeddings at training via distillation but not at inference. We show that VPN++ is not only effective but also provides a high speed up and high resilience to noisy Poses. VPN++, with or without 3D Poses, outperforms the representative baselines on 4 public datasets. Code is available at https://github.com/srijandas07/vpnplusplus.

Calcified chronic subdural hematoma: illustrative case.

BACKGROUND: Calcified chronic subdural hematomas (CCSDHs) are rare variants of chronic subdural hematomas (CSDHs) accounting to only 0.3-2.7% of CSDHs. Although the majority of the patients with CSDHs recover from surgery, there still is some doubt about its being applied to CCSDHs. OBSERVATIONS: In this case report, the authors present a case of a 75-year-old male presenting with deterioration of motor function in his left limbs over the course of 18 months and acute neurological deterioration in the form of altered sensorium for 7 days. The patient experienced an episode of aspiration in the preoperative period that led to deterioration of pulmonary function in the postoperative period. A chest radiograph showed diffuse patches suggesting pulmonary compromise. Computed tomography and magnetic resonance imaging (MRI) documented a large subdural collection at the right frontal and parietal hemisphere with calcification, which was successfully and completely removed by surgery. LESSONS: The chances of a subdural hematoma progressing to calcification is extremely rare. The presentation of this case was such that surgical intervention was the only option left for the patient. The presence of lacunar infarcts in the thalamus on MRI can also be attributed to the calcified hematoma.

In silico analysis of new flavonoids from Pongamia pinnata with a therapeutic potential for age-related macular degeneration.

Age-related macular degeneration (AMD) leads to progressive degeneration of the macula which ultimately results in the complete loss of central vision. The present study aims to identify the new therapeutic agents for curing AMD. In the present study we have isolated, and compared the activity of natural flavonoids (Karanjin, Karanjachromene, Pongachromene, Pongapin) from plant species Pongamia pinnata (L.) Pierre (Family: Fabaceae) with known flavonol, Quercetin, and a drug Pazopanib through in silico approaches. Chemical structures of isolated flavonoids passed the ADME and PASS analysis, showed drug-like properties without violation of Lipinski parameters. Molecular docking studies were also performed for all isolated flavonoids with the receptors responsible for AMD viz. P2X7, PPAR, RAGE, and TLR3. Docking scores of the flavonoids with the receptors were found to be comparable to that of Quercetin, and Pazopanib (drugs already known for AMD treatment). Among all the flavonoids, Karanjachromene [P2X7 (- 31.39)] and Pongachromene [PPAR (- 65.13), RAGE (- 43.42)] showed a very good binding affinity with receptors predicting them to be the new potent chemical entities for the treatment of AMD.