Hybrid manifold smoothing and label propagation technique for Kannada handwritten character recognition.

Handwritten character recognition is one of the classical problems in the field of image classification. Supervised learning techniques using deep learning models are highly effective in their application to handwritten character recognition. However, they require a large dataset of labeled samples to achieve good accuracies. Recent supervised learning techniques for Kannada handwritten character recognition have state of the art accuracy and perform well over a large range of input variations. In this work, a framework is proposed for the Kannada language that incorporates techniques from semi-supervised learning. The framework uses features extracted from a convolutional neural network backbone and uses regularization to improve the trained features and label propagation to classify previously unseen characters. The episodic learning framework is used to validate the framework. Twenty-four classes are used for pre-training, 12 classes are used for testing and 11 classes are used for validation. Fine-tuning is tested using one example per unseen class and five examples per unseen class. Through experimentation the components of the network are implemented in Python using the Pytorch library. It is shown that the accuracy obtained 99.13% make this framework competitive with the currently available supervised learning counterparts, despite the large reduction in the number of labeled samples available for the novel classes.

Discovery of novel Li SSE and anode coatings using interpretable machine learning and high-throughput multi-property screening.

All-solid-state batteries with Li metal anode can address the safety issues surrounding traditional Li-ion batteries as well as the demand for higher energy densities. However, the development of solid electrolytes and protective anode coatings possessing high ionic conductivity and good stability with Li metal has proven to be a challenge. Here, we present our informatics approach to explore the Li compound space for promising electrolytes and anode coatings using high-throughput multi-property screening and interpretable machine learning. To do this, we generate a database of battery-related materials properties by computing [Formula: see text] migration barriers and stability windows for over 15,000 Li-containing compounds from Materials Project. We screen through the database for candidates with good thermodynamic and electrochemical stabilities, and low [Formula: see text] migration barriers, identifying promising new candidates such as [Formula: see text]N, [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text], among others. We train machine learning models, using ensemble methods, to predict migration barriers and oxidation and reduction potentials of these compounds by engineering input features that ensure accuracy and interpretability. Using only a small number of features, our gradient boosting regression models achieve [Formula: see text] values of 0.95 and 0.92 on the oxidation and reduction potential prediction tasks, respectively, and 0.86 on the migration barrier prediction task. Finally, we use Shapley additive explanations and permutation feature importance analyses to interpret our machine learning predictions and identify materials properties with the largest impact on predictions in our models. We show that our approach has the potential to enable rapid discovery and design of novel solid electrolytes and anode coatings.

The axis of local cardiac endogenous Klotho-TGF-ß1-Wnt signaling mediates cardiac fibrosis in human.

BACKGROUND: Cardiovascular fibrosis is a major contributor to cardiovascular disease, the primary cause of death in patients with chronic kidney disease (CKD). We previously reported expression of endogenous Klotho in human arteries, and that CKD is a state of Klotho deficiency, resulting in vascular calcification, but myocardial expression of Klotho is poorly understood. This study aimed to further clarify endogenous Klotho's functional roles in cardiac fibrosis in patients with underlying CKD. METHODS AND RESULTS: Human atrial appendage specimens were collected during cardiac surgery from individuals with or without CKD. Cardiac fibrosis was quantified using trichrome staining. For endogenous Klotho functional studies, primary human cardiomyocytes (HCMs) were treated with uremic serum from CKD patients or recombinant human TGF-ß1. The effects of endogenous Klotho in HCMs were studied using Klotho-siRNA and Klotho-plasmid transfection. Both gene and protein expression of endogenous Klotho are found in human heart, but decreased Klotho expression is clearly associated with the degree of cardiac fibrosis in CKD patients. Moreover, we show that endogenous Klotho is expressed by HCMs and cardiac fibroblasts (HCFs) but that HCM expression is suppressed by uremic serum or TGF-ß1. Klotho knockdown or overexpression aggravates or mitigates TGF-ß1-induced fibrosis and canonical Wnt signaling in HCMs, respectively. Furthermore, co-culture of HCMs with HCFs increases TGF-ß1-induced fibrogenic proteins in HCFs, but overexpression of endogenous Klotho in HCMs mitigates this effect, suggesting functional crosstalk between HCMs and HCFs. CONCLUSIONS: Our data from analysis of human hearts as well as functional in vitro studies strongly suggests that the loss of cardiac endogenous Klotho in CKD patients, specifically in cardiomyocytes, facilitates intensified TGF-ß1 signaling which enables more vigorous cardiac fibrosis through upregulated Wnt signaling. Upregulation of endogenous Klotho inhibits pathogenic Wnt/ß-catenin signaling and may offer a novel strategy for prevention and treatment of cardiac fibrosis in CKD patients.

Preclinical in vivo evaluation of NPe6-mediated photodynamic therapy on normal vasculature.

BACKGROUND AND OBJECTIVE: Current treatments of port-wine stain birthmarks typically involve use of a pulsed dye laser (PDL) combined with cooling of the skin. Currently, PDL therapy protocols result in varied success, as some patients experience complete blanching, while others do not. Over the past decade, we have studied the use of photodynamic therapy (PDT) as either a replacement or adjuvant treatment option to photocoagulate both small and large vasculature. The objective of the current study was to evaluate a PDT protocol that involves use of an alternate intravascular photosensitizer mono-L-aspartylchlorin-e6 (NPe6) activated by an array of low-cost light emitting diodes. STUDY DESIGN/MATERIALS AND METHODS: To monitor the microvasculature, a dorsal window chamber model was installed on 22 adult male mice. The light source consisted of a custom-built LED array that emitted 10 W at a center wavelength of 664 nm (FWHM = 20 nm). The light source was positioned at a fixed distance from the window chamber to achieve a fixed irradiance of 127 mW/cm(2). A retroorbital injection of NPe6 (5 mg/kg) was performed to deliver the drug into the bloodstream. Laser irradiation was initiated immediately after injection. To monitor blood-flow dynamics in response to PDT, we used laser speckle imaging. We employed a dose-response experimental design to evaluate the efficacy of NPe6-mediated PDT. RESULTS: We observed three general hemodynamic responses to PDT: (1) At low radiant exposures, we did not observe any persistent vascular shutdown; (2) at intermediate radiant exposures, we observed an acute decrease in blood flow followed by gradual restoration of blood flow over the 7-day monitoring period; and (3) at high radiant exposures, we observed acute vascular shutdown that persisted during the entire 7-day monitoring period. Dose-response analysis enabled identification of 85 J/cm(2) as a characteristic radiant exposure required to achieve persistent vascular shutdown at Day 7 following PDT. CONCLUSION: The experimental data suggest that NPe6-mediated PDT can achieve persistent vascular shutdown of normal microvasculature.

Evaluation of glutathione in oral squamous cell carcinoma.

PURPOSE: To estimate the RBC level of glutathione among oral squamous cell carcinoma cases and normal healthy individual. The study compares the pre and postoperative blood levels of glutathione in patients with oral squamous cell carcinoma. MATERIALS AND METHOD: In the present study a total of 45, age matched subjects were recruited. The 45 subjects were split into 15 subjects in control, 30 clinically and histopathologically diagnosed cases of oral squamous cell carcinoma who had not received any prior treatment. RESULTS: A statistically significant difference of mean value of glutathione levels in RBC of the control group was found to be higher than that of oral squamous cell carcinoma patient group. A noticeable rise in glutathione concentration in RBC of oral squamous cell carcinoma towards normal was seen postoperatively, however these variations were not statistically significant. CONCLUSION: Pre and postoperatively a marked variation in levels of glutathione was observed. In addition, it was noticed that the levels tended to shift closer to control/normal values postoperatively. The value of glutathione as a prognostic marker could be ratified if the study was carried out on a larger scale over a sustained period of time.