MLVICX: Multi-Level Variance-Covariance Exploration for Chest X-ray Self-Supervised Representation Learning.

Self-supervised learning (SSL) is potentially useful in reducing the need for manual annotation and making deep learning models accessible for medical image analysis tasks. By leveraging the representations learned from unlabeled data, self-supervised models perform well on tasks that require little to no fine-tuning. However, for medical images, like chest X-rays, characterized by complex anatomical structures and diverse clinical conditions, a need arises for representation learning techniques that encode fine-grained details while preserving the broader contextual information. In this context, we introduce MLVICX (Multi-Level Variance-Covariance Exploration for Chest X-ray Self-Supervised Representation Learning), an approach to capture rich representations in the form of embeddings from chest X-ray images. Central to our approach is a novel multi-level variance and covariance exploration strategy that effectively enables the model to detect diagnostically meaningful patterns while reducing redundancy. MLVICX promotes the retention of critical medical insights by adapting global and local contextual details and enhancing the variance and covariance of the learned embeddings. We demonstrate the performance of MLVICX in advancing self-supervised chest X-ray representation learning through comprehensive experiments. The performance enhancements we observe across various downstream tasks highlight the significance of the proposed approach in enhancing the utility of chest X-ray embeddings for precision medical diagnosis and comprehensive image analysis. For pertaining, we used the NIH-Chest X-ray dataset, while for downstream tasks, we utilized NIH-Chest X-ray, Vinbig-CXR, RSNA pneumonia, and SIIM-ACR Pneumothorax datasets. Overall, we observe up to 3% performance gain over SOTA SSL approaches in various downstream tasks. Additionally, to demonstrate the generalizability of the proposed method, we conducted additional experiments on fundus images and observed superior performance on multiple datasets. Codes are available at https://github.com/azad6629/mlvicx/ GitHub.

The influential responsibility of sirtuins in senescence and associated diseases: A review.

Aging is a process of time-associated depletion in the physiological functions, essential for the survival and reproducibility of living beings. Some age-related disorders can be successfully controlled with some biomedical techniques or pharmaceutical approaches. There are some precise remedies that demonstrate conspicuous promise in the preclinical and clinical setup of extending lifespan or enhancing health by altering natural senescence. The sirtuin family of proteins is one of the most favorable targets for antiaging strategies. Sirtuins were initially identified as transcription repressors in yeast, but today they are known to exist in bacteria and eukaryotes, as well as humans. The SIRT (1-7) family of proteins in humans is made up of seven members, each of which has either mono-ADP ribosyl transferase or deacetylase activity. Researchers suggest that sirtuins are essential for cell metabolism and play a major role in how cells react to various stimuli, such as oxidative or genotoxic stress. A healthy lifestyle, which includes exercise and a balanced diet, has been demonstrated to impact health span by adjusting the levels of sirtuins, suggesting the involvement of sirtuins in extending human longevity. The hunt for sirtuin activators is among the most extensive and comprehensive research subjects in the present scenario. Some optimism has been generated to investigate antiaging therapies by natural compounds, such as curcumin and others. This review article highlights the role of sirtuins in native senescence and their primordial roles in the progression of several life-threatening diseases. Further, it also provides recent information on the sirtuin activators and inhibitors and their therapeutic benefits.

The spatial distribution of intermediate fibroblasts and myeloid-derived cells dictate lymph node metastasis dynamics in oral cancer.

BACKGROUND: Oral cancer poses a significant health challenge due to limited treatment protocols and therapeutic targets. We aimed to investigate the invasive margins of gingivo-buccal oral squamous cell carcinoma (GB-OSCC) tumors in terms of the localization of genes and cell types within the margins at various distances that could lead to nodal metastasis. METHODS: We collected tumor tissues from 23 resected GB-OSCC samples for gene expression profiling using digital spatial transcriptomics. We monitored differential gene expression at varying distances between the tumor and its microenvironvent (TME), and performed a deconvulation study and immunohistochemistry to identify the cells and genes regulating the TME. RESULTS: We found that the tumor-stromal interface (a distance up to 200 µm between tumor and immune cells) is the most active region for disease progression in GB-OSCC. The most differentially expressed apex genes, such as FN1 and COL5A1, were located at the stromal ends of the margins, and together with enrichment of the extracellular matrix (ECM) and an immune-suppressed microenvironment, were associated with lymph node metastasis. Intermediate fibroblasts, myocytes, and neutrophils were enriched at the tumor ends, while cancer-associated fibroblasts (CAFs) were enriched at the stromal ends. The intermediate fibroblasts transformed into CAFs and relocated to the adjacent stromal ends where they participated in FN1-mediated ECM modulation. CONCLUSION: We have generated a functional organization of the tumor-stromal interface in GB-OSCC and identified spatially located genes that contribute to nodal metastasis and disease progression. Our dataset might now be mined to discover suitable molecular targets in oral cancer.

Environmental drivers behind the exceptional increase in cyanobacterial blooms in Okavango Delta, Botswana.

The Okavango Delta region in Botswana experienced exceptionally intense landscape-wide cyanobacterial harmful algal blooms (CyanoHABs) in 2020. In this study, the drivers behind CyanoHABs were determined from thirteen independent environmental variables, including vegetation indices, climate and meteorological parameters, and landscape variables. Annual Land Use Land Cover (LULC) maps were created from 2017 to 2020, with ∼89% accuracy to compute landscape variables such as LULC change. Generalized Additive Models (GAM) and Structural Equation Models (SEM) were used to determine the most important drivers behind the CyanoHABs. Normalized Difference Chlorophyll Index (NDCI) and Green Line Height (GLH) algorithms served as proxies for chlorophyll-a (green algae) and phycocyanin (cyanobacteria) concentrations. GAM models showed that seven out of the thirteen variables explained 89.9% of the variance for GLH. The models showcased that climate variables, including monthly precipitation (8.8%) and Palmer Severity Drought Index- PDSI (3.2%), along with landscape variables such as changes in Wetlands area (7.5%), and Normalized Difference Vegetation Index (NDVI) (5.4%) were the determining drivers behind the increased cyanobacterial activity within the Delta. Both PDSI and NDVI showed negative correlations with GLH, indicating that increased drought conditions could have led to large increases in toxic CyanoHAB activity within the region. This study provides new information about environmental drivers which can help monitor and predict regions at risk of future severe CyanoHABs outbreaks in the Okavango Delta, Botswana, and other similar data-scarce and ecologically sensitive areas in Africa. Plain Language Summary: The waters of the Okavango Delta in Northern Botswana experienced an exceptional increase in toxic cyanobacterial activity in recent years. Cyanobacterial blooms have been shown to affect local communities and wildlife in the past. To determine the drivers behind this increased bloom activity, we analyzed the effects of thirteen independent environmental variables using two different statistical models. Within this research, we focused on vegetation indices, meteorological, and landscape variables, as previous studies have shown their effect on cyanobacterial activity in other parts of the world. While driver determination for cyanobacteria has been done before, the environmental conditions most important for cyanobacterial growth can be specific to the geographic setting of a study site. The statistical analysis indicated that the increases in cyanobacterial bloom activity within the region were mainly driven by persistent drier conditions. To our knowledge, this is the first study to determine the driving factors behind cyanobacterial activity in this region of the world. Our findings will help to predict and monitor areas at risk of future severe cyanobacterial blooms in the Okavango Delta and other similar African ecosystems.

Correlation of glycosylated hemoglobin (HbA1c) with retinal nerve fiber layer (RNFL) thickness and central macular thickness (CMT) in the diabetic population in North India.

PURPOSE: The current study was aimed to find correlation of glycosylated hemoglobin with retinal nerve fiber layer thickness (RNFLT) and central macular thickness (CMT) in the diabetic population in North India. METHODS: This was a cross-sectional observational study of 300 diabetic patients divided equally in two groups with and without retinopathy, and 150 people were included as control. The study was conducted from October 2020 to August 2022. All patients underwent slitlamp fundoscopy with a +78 D lens, and spectral-domain (SD) optical coherence tomography was performed to measure the RNFLT and CMT, and the staging of retinopathy was done as per the ETDRS classification. Along with that, blood investigations were ordered, including fasting (FBS) and post-prandial (PPBS) blood sugar and glycosylated hemoglobin (HbA1c). Quantitative variables were compared using one-way analysis of variance, or Kruskal-Wallis test was applied for inter-group comparison, followed by a Student Newman Keuls Test. RESULTS: The mean age of the patients in the diabetic group with retinopathy was 52.62 ± 9.38 years. The overall male: female ratio was 3:2. The mean FBS in the diabetic group with retinopathy was 146.54 ± 45.40mg/dl; the PPBS and HbA1c in the same were 210.39 ± 63.71mg/dl and 7.85 ± 1.33%, respectively. RNFL thinning was found in all four quadrants in diabetics irrespective of the status of retinopathy (P-value = 0.000) with a significant weak negative (r<0.4) correlation of glycosylated hemoglobin values with RNFLT in the inferior (r value = -0.300, P-value = 0.000) and superior (r value = -0.236, P-value = 0.004) quadrants of right eyes and in inferior (r value = -0.176, p- value = 0.031), superior (r value = -0.222, P value = 0.006), and nasal quadrants (r value = -0.166, p- value = 0.043) of left eyes in diabetics with retinopathy. However, in diabetics without retinopathy, no correlation was found. On correlating HbA1c with CMT, a weak positive (r<0.3) association existed in both eyes in the diabetic group without retinopathy (r = 0.020 and 0.048 for OD and OS, respectively) and diabetics with retinopathy (r = 0.152 and 0.127 for OD and OS, respectively). However, the association was not found to be significant in either of the groups (P-value > 0.05). CONCLUSION: The study concluded that neurodegeneration occurs in diabetic retinopathy as evident with nerve fiber layer thinning, and it is negatively correlated with glycosylated hemoglobin (HbA1c).

Deep Efficient Data Association for Multi-Object Tracking: Augmented with SSIM-Based Ambiguity Elimination.

Recently, to address the multiple object tracking (MOT) problem, we harnessed the power of deep learning-based methods. The tracking-by-detection approach to multiple object tracking (MOT) involves two primary steps: object detection and data association. In the first step, objects of interest are detected in each frame of a video. The second step establishes the correspondence between these detected objects across different frames to track their trajectories. This paper proposes an efficient and unified data association method that utilizes a deep feature association network (deepFAN) to learn the associations. Additionally, the Structural Similarity Index Metric (SSIM) is employed to address uncertainties in the data association, complementing the deep feature association network. These combined association computations effectively link the current detections with the previous tracks, enhancing the overall tracking performance. To evaluate the efficiency of the proposed MOT framework, we conducted a comprehensive analysis of the popular MOT datasets, such as the MOT challenge and UA-DETRAC. The results showed that our technique performed substantially better than the current state-of-the-art methods in terms of standard MOT metrics.

Demonstration of photonics-based D-band integrated localization and communication.

The terahertz spectrum has the ability to provide high-speed communication and millimeter-level resolution. As a result, terahertz-integrated sensing and communication (ISAC) has been identified as a key enabler for 6G wireless networks. This work discusses a photonics-based D-band communication system for integrated high-resolution localization and high-speed wireless communication. Our empirical results show that a communication rate of 5 Gbps over a distance of 1.5 m and location identification of the target with millimeter-level (<4m m) range resolution can be conducted simultaneously using the same signal. We also show that the error due to the thickness of the beam splitter can be eliminated, while the quantization error and the random drift errors are the limiting factors of the resolution achieved. This experimental demonstration using D-band communication indicates that terahertz ISAC can be realized for 6G networks while considering the underlying system restrictions (e.g., bandwidth limit and lens diameter).

Melphalan-associated encephalopathy following autologous stem cell transplant: a transplanter's nightmare!

Melphalan-induced encephalopathy is a rare complication observed in patients undergoing autologous stem cell transplantation (ASCT) and is characterized by symptoms ranging from drowsiness to seizures. Previous reports have described similar cases, including a review of a large cohort of patients in whom melphalan-associated encephalopathy was identified in 2% of the patients undergoing ASCT. We describe the case of a 63-year-old male with Multiple Myeloma and underlying chronic kidney disease (CKD) who underwent ASCT with a reduced dose of melphalan due to renal dysfunction in complete remission following induction therapy and subsequent neurological deterioration, which necessitated an extensive evaluation of several neurological and infective etiologies. In this report, we highlight that melphalan-associated encephalopathy is a distinct entity complicating ASCT in patients with myeloma, especially in those with preexisting renal insufficiency, and consider its management.

Developments in the stereoselective synthesis of benzopyran, benzopyrone and flavonoid based natural product analogues using C-glycosides as an intrinsic chiral synthon.

Stereoselective synthesis is essential for propelling mainstream academia toward a relentless pursuit of novel and cutting-edge strategies for constructing molecules with unparalleled precision. Naturally derived benzopyrans, benzopyrones, and flavonoids are an essentially prominent group of oxa-heterocycles, highly significant targets in medicinal chemistry owing to their extensive abundance in biologically active natural products and pharmaceuticals. The molecular complexity and stereoselectivity induced by heterocycles embedded with C-glycosides have attracted considerable interest and emerged as a fascinating area of research for synthetic organic chemists. This present article emphasizes the existing growths in the strategies involving the diastereoselective synthesis of C-glycosylated benzopyrans, benzopyrones, and flavonoids using naturally acquired glycones as chiral synthons.

Uracil as a biomarker for spatial pyrimidine metabolism in the development of gingivobuccal oral squamous cell carcinoma.

No biomarker has yet been identified that allows accurate diagnosis and prognosis of oral cancers. In this study, we investigated the presence of key metabolites in oral cancer using proton nuclear magnetic resonance (NMR) spectroscopy to identify metabolic biomarkers of gingivobuccal oral squamous cell carcinoma (GB-OSCC). NMR spectroscopy revealed that uracil was expressed in 83.09% of tumor tissues and pyrimidine metabolism was active in GB-OSCC; these results correlated well with immunohistochemistry (IHC) and RNA sequencing data. Based on further gene and protein analyses, we proposed a pathway for the production of uracil in GB-OSCC tissues. Uridinetriphosphate (UTP) is hydrolyzed to uridine diphosphate (UDP) by CD39 in the tumor microenvironment (TME). We hypothesized that UDP enters the cell with the help of the UDP-specific P2Y6 receptor for further processing by ENTPD4/5 to produce uracil. As the ATP reserves diminish, the weakened immune cells in the TME utilize pyrimidine metabolism as fuel for antitumor activity, and the same mechanism is hijacked by the tumor cells to promote their survival. Correspondingly, the differential expression of ENTPD4 and ENTPD5 in immune and tumor cells, respectively, indicatedtheir involvement in disease progression. Furthermore, higher uracil levels were detected in patients with lymph node metastasis, indicating that metastatic potential is increased in the presence of uracil. The presence of uracil and/or expression patterns of intermediate molecules in purine and pyrimidine pathways, such asCD39, CD73, and P2Y6 receptors together with ENTPD4 and ENTPD5, hold promise as biomarker(s) for oral cancer diagnosis and prognosis.

An air- and moisture-stable ruthenium precatalyst for diverse reactivity.

Versatile, efficient and robust (pre)catalysts are pivotal in accelerating the discovery and optimization of chemical reactions, shaping diverse synthetic fields such as cross-coupling, C-H functionalization and polymer chemistry. Yet, their scarcity in certain domains has hindered the advancement and adoption of new applications. Here we present a highly reactive air- and moisture-stable ruthenium precatalyst [(tBuCN)5Ru(H2O)](BF4)2, featuring a key exchangeable water ligand. This versatile precatalyst drives an array of transformations, including late-stage C(sp2)-H arylation, primary/secondary alkylation, methylation, hydrogen/deuterium exchange, C(sp3)-H oxidation, alkene isomerization and oxidative cleavage, consistently outperforming conventionally used ruthenium (pre)catalysts. The generality and applicability of this precatalyst is exemplified through the potential for rapid screening and optimization of photocatalytic reactions with a suite of in situ generated ruthenium photocatalysts containing hitherto unknown complexes, and through the rapid discovery of reactivities previously unreported for ruthenium. The diverse applicability observed is suggestive of a generic platform for reaction simplification and accelerated synthetic discovery that will enable broader applicability and accessibility to state-of-the-art ruthenium catalysis.

Pediatric corneal transplantation: techniques, challenges, and outcomes.

Pediatric corneal transplant is a highly demanding and technically challenging procedure for the cornea surgeon in today's era. These cases pose unique challenges in clinical and surgical management. The indications of pediatric corneal transplant can be therapeutic, tectonic, optical, and cosmetic. Pediatric patients undergoing corneal transplants are at a high risk of graft infection, failure, rejection, dehiscence, and amblyopia due to young age, robust immune system, increased incidence of trauma, and compliance issues. The other factors contributing to graft failure can be allograft rejection, secondary glaucoma, corneal vascularization, multiple surgeries, vitreous prolapse, and lack of treatment compliance. A successful corneal transplant in children depends on meticulous preoperative evaluation, uneventful surgery, the expertise of a corneal surgeon, and regular and timely postoperative follow-up. Therapeutic and optical penetrating keratoplasty are the most commonly performed transplants in children. However, with the advancements in surgical technique and management protocol, the current focus has shifted toward lamellar keratoplasty. Lamellar keratoplasty offers early visual recovery and potentially fewer complications. Visual rehabilitation through corneal transplant in otherwise blind eyes can be a boon for the children. Recently, keratoprostheses have been promising in children with multiple graft failures. The current review gives insights into epidemiology, etiology, indications, clinical characteristics, investigations, management options, recent advances, and the future of pediatric corneal transplants. As surgical techniques continue to grow and comprehension of pediatric corneal transplants is improving, we can safeguard these eyes with the best possible anatomical and functional outcomes.

Computational ensemble expert system classification for the recognition of bruxism using physiological signals.

This study aimed to develop an automatic diagnostic scheme for bruxism, a sleep-related disorder characterized by teeth grinding and clenching. The aim was to improve on existing methods, which have been proven to be inefficient and challenging. We utilized a novel hybrid machine learning classifier, facilitated by the Weka tool, to diagnose bruxism from biological signals. The study processed and examined these biological signals by calculating the power spectral density. Data were categorized into normal or bruxism categories based on the EEG channel (C4-A1), and the sleeping phases were classified into wake (w) and rapid eye movement (REM) stages using the ECG channel (ECG1-ECG2). The classification resulted in a maximum specificity of 93% and an accuracy of 95% for the EEG-based diagnosis. The ECG-based classification yielded a supreme specificity of 87% and an accuracy of 96%. Furthermore, combining these phases using the EMG channel (EMG1-EMG2) achieved the highest specificity of 95% and accuracy of 98%. The ensemble Weka tool combined all three physiological signals EMG, ECG, and EEG, to classify the sleep stages and subjects. This integration increased the specificity and accuracy to 97% and 99%, respectively. This indicates that a more precise bruxism diagnosis can be obtained by including all three biological signals. The proposed method significantly improves bruxism diagnosis accuracy, potentially enhancing automatic home monitoring systems for this disorder. Future studies may expand this work by applying it to patients for practical use.

An unusual case of transverse myelitis in dengue fever: A case report from Nepal.

Key Clinical Message: Dengue fever can also have various neurological complications but involvement of the spinal cord is often unusual. This is a case where the patient had transverse myelitis as a complication of dengue fever. Abstract: Dengue fever can have various neurological complications but involvement of the spinal cord is often unusual. We report a case of a 49-year-old female, a known case of dengue fever, who presented with urine retention, inability to stand and walk with tingling sensation of bilateral lower limbs. Her vibration and joint position sensation was reduced below T2 level along with altered reflexes but MRI could not explain the examination findings. She was diagnosed clinically as transverse myelitis (TM) in the background of dengue fever. She showed drastic improvement with treatment of steroids. As TM as a complication in a patient with dengue fever is rare, and due to the paucity of similar case reports in Nepal, this case report is of value for the scientific community.

Low-dose atropine 0.01% for the treatment of childhood myopia: a pan-India multicentric retrospective study.

OBJECTIVE: The objective of this study was to assess the efficacy of low-dose atropine 0.01% in controlling myopia progression among Indian children over a 2-year period. METHODS: This retrospective study, conducted across 20 centres in India, monitored the progression of myopia over 2 years after initiating treatment with 0.01% atropine eye drops. This included children between 6 and 14 years with baseline myopia ranging from -0.5 D to -6 D, astigmatism≤-1.5 D, anisometropia ≤ -1 D and documented myopia progression of ≥0.5 D in the year prior to starting atropine. Subjects with any other ocular pathologies were excluded. RESULTS: A total of 732 children were included in the data analysis. The mean age of the subjects was 9.3±2.7 years. The mean myopia progression at baseline (1 year before starting atropine) was -0.75±0.31 D. The rate of myopia progression was higher in younger subjects and those with higher baseline myopic error. After initiating atropine, myopia progression significantly decreased to -0.27±0.14 D at the end of the first year and -0.24±0.15 D at the end of the second year (p<0.001). Younger children (p<0.001) and higher baseline myopia (p<0.001) was associated with greater myopia progression and poor treatment response (p<0.001 for both). CONCLUSION: Low-dose atropine (0.01%) effectively reduces myopia progression over 2 years in Indian children.

Preclinical pediatric brain tumor models for immunotherapy: Hurdles and a way forward.

Brain tumors are the most common solid tumor in children and the leading cause of cancer-related deaths. Over the last few years, improvements have been made in the diagnosis and treatment of children with Central Nervous System tumors. Unfortunately, for many patients with high-grade tumors, the overall prognosis remains poor. Lower survival rates are partly attributed to the lack of efficacious therapies. The advent and success of immune checkpoint inhibitors (ICIs) in adults have sparked interest in investigating the utility of these therapies alone or in combination with other drug treatments in pediatric patients. However, to achieve improved clinical outcomes, the establishment and selection of relevant and robust preclinical pediatric high-grade brain tumor models is imperative. Here, we review the information that influenced our model selection as we embarked on an international collaborative study to test ICIs in combination with epigenetic modifying agents to enhance adaptive immunity to treat pediatric brain tumors. We also share challenges that we faced and potential solutions.

Bulk Schottky Junctions-Based Flexible Triboelectric Nanogenerators to Power Backscatter Communications in Green 6G Networks.

This work introduces a novel method to construct Schottky junctions to boost the output performance of triboelectric nanogenerators (TENGs). Perovskite barium zirconium titanate (BZT) core/metal silver shell nanoparticles are synthesized to be embedded into electrospun polyvinylidene fluoride-co-hexafluoropropylene (PVDF-HFP) nanofibers before they are used as tribo-negative layers. The output power of TENGs with composite fiber mat exhibited >600% increase compared to that with neat polymer fiber mat. The best TENG achieved 1339 V in open-circuit voltage, 40 µA in short-circuit current and 47.9 W m-2 in power density. The Schottky junctions increased charge carrier density in tribo-layers, ensuring a high charge transfer rate while keeping the content of conductive fillers low, thus avoiding charge loss and improving performance. These TENGs are utilized to power radio frequency identification (RFID) tags for backscatter communication (BackCom) systems, enabling ultra-massive connectivity in the 6G wireless networks and reducing information communications technology systems' carbon footprint. Specifically, TENGs are used to provide an additional energy source to the passive tags. Results show that TENGs can boost power for BackCom and increase the communication range by 386%. This timely contribution offers a novel route for sustainable 6G applications by exploiting the expanded communication range of BackCom tags.

Severity of error in hierarchical datasets.

Classification tasks today, especially for the medical domain, use datasets which are often hierarchical. These tasks are approached using methods that consider the class taxonomy for predicting a label. The classifiers are gradually becoming increasingly accurate over the complex datasets. While increasing accuracy is a good way to judge a model, in high-risk applications, it needs to be ensured that even if the model makes a mistake, it does not bear a severe consequence. This work explores the concept of severity of an error and extends it to the medical domain. Further, it aims to point out that accuracy or AUROC alone are not sufficient metrics to decide the performance of a model in a setting where a misclassification will incur a severe cost. Various approaches to reduce severity for classification models are compared and evaluated in this work, which indicate that while many of them might be suited for a traditional image classification setting, there is a need for techniques tailored toward tasks and settings of medical domain to push artificial intelligence in healthcare to a deployable state.

Large Scale Time-Series Representation Learning via Simultaneous Low-and High-Frequency Feature Bootstrapping.

Learning representations from unlabeled time series data is a challenging problem. Most existing self-supervised and unsupervised approaches in the time-series domain fall short in capturing low-and high-frequency features at the same time. As a result, the generalization ability of the learned representations remains limited. Furthermore, some of these methods employ large-scale models like transformers or rely on computationally expensive techniques such as contrastive learning. To tackle these problems, we propose a noncontrastive self-supervised learning (SSL) approach that efficiently captures low-and high-frequency features in a cost-effective manner. The proposed framework comprises a Siamese configuration of a deep neural network with two weight-sharing branches which are followed by low-and high-frequency feature extraction modules. The two branches of the proposed network allow bootstrapping of the latent representation by taking two different augmented views of raw time series data as input. The augmented views are created by applying random transformations sampled from a single set of augmentations. The low-and high-frequency feature extraction modules of the proposed network contain a combination of multilayer perceptron (MLP) and temporal convolutional network (TCN) heads, respectively, which capture the temporal dependencies from the raw input data at various scales due to the varying receptive fields. To demonstrate the robustness of our model, we performed extensive experiments and ablation studies on five real-world time-series datasets. Our method achieves state-of-art performance on all the considered datasets.