Search

Your search keyword '"Ravindran, Sriram"' showing total 319 results
Start Over Author "Ravindran, Sriram"
319 results on '"Ravindran, Sriram"'

1. SEMPART: Self-supervised Multi-resolution Partitioning of Image Semantics

Author

Ravindran, Sriram and Basu, Debraj

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning
Abstract

Accurately determining salient regions of an image is challenging when labeled data is scarce. DINO-based self-supervised approaches have recently leveraged meaningful image semantics captured by patch-wise features for locating foreground objects. Recent methods have also incorporated intuitive priors and demonstrated value in unsupervised methods for object partitioning. In this paper, we propose SEMPART, which jointly infers coarse and fine bi-partitions over an image's DINO-based semantic graph. Furthermore, SEMPART preserves fine boundary details using graph-driven regularization and successfully distills the coarse mask semantics into the fine mask. Our salient object detection and single object localization findings suggest that SEMPART produces high-quality masks rapidly without additional post-processing and benefits from co-optimizing the coarse and fine branches.

Published
2023

2. Multiple Attribute Fairness: Application to Fraud Detection

Author

Y, Meghanath Macha, Ravindran, Sriram, Pai, Deepak, Narang, Anish, and Srivastava, Vijay

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Computers and Society
Abstract

We propose a fairness measure relaxing the equality conditions in the popular equal odds fairness regime for classification. We design an iterative, model-agnostic, grid-based heuristic that calibrates the outcomes per sensitive attribute value to conform to the measure. The heuristic is designed to handle high arity attribute values and performs a per attribute sanitization of outcomes across different protected attribute values. We also extend our heuristic for multiple attributes. Highlighting our motivating application, fraud detection, we show that the proposed heuristic is able to achieve fairness across multiple values of a single protected attribute, multiple protected attributes. When compared to current fairness techniques, that focus on two groups, we achieve comparable performance across several public data sets., Comment: 5 pages, 5 figures, 1 table

Published
2022

9. Comparative Osteogenesis of Three Clinical Bone Graft Materials: An In Vivo Study.

Author

Gabriel, Anthony, Ravindran, Sriram, Cooper, Lyndon F., Gajendrareddy, Praveen, Chun-Chieh Huang, Miya Kang, and Thalji, Ghadeer

Subjects
TOOTH demineralization, COLLAGEN, BONE growth, XENOGRAFTS, BONES, ANIMAL experimentation, IMMUNOHISTOCHEMISTRY, COMPARATIVE studies, RATS, CYTOCHEMISTRY, DESCRIPTIVE statistics, RESEARCH funding, DATA analysis software, BONE regeneration, COMPUTED tomography, TRANSCRIPTION factors, BONE density, BONE grafting
Abstract

Purpose: To investigate bone regeneration among three different bone graft materials in a rat calvarum model. Materials and Methods: A total of 24 rats had two 5-mm defects placed per calvarial. Rats were divided into four groups: bovine xenograft (XG), demineralized bone matrix (DBM), mineralized bone graft (MBG), and collagen membrane control (CC). Within each group, samples were collected at two time points: 4 weeks (T4) and 8 weeks (T8). Bone regeneration was assessed by microcomputed tomography (micro-CT) imaging and was analyzed using MATLAB software. Additionally, the fixed samples were subsequently demineralized for immunohistochemistry and histomorphometry. Slides were mounted and stained with hematoxylin and eosin (H&E) stain as well as bone morphogenetic protein 2 (BMP-2) and runt-related transcription factor 2 (RUNX2) markers. The numbers of positive cells/area were calculated for each group and analyzed. Results: At 4 weeks, DBM showed low mineral density (7.7%) compared to the control (25.2%), but increased dramatically at 8 weeks (DBM, T8 = 27.6%; CC, T8 = 27.2%). Xenograft material showed an increase in mineral desnity between T4 and T8 (XG, T4 = 25.0%; XG, T8 = 32.3%). MBG remained consistent over the 8-week trial period (MBG, T4 = 30.4%; MBG, T8 = 30.4%). BMP-2 expression was present in cells adherent to all graft materials. RUNX2 expression was also observed in cells adherent to all graft materials, indicating that during the 4- to 8-week healing period, all materials supported osteogenesis. Conclusions: Compared to other materials, the DBM had high osteoinductive properties during the 4- to 8-week time period based on increased mineral content. All materials were associated with immunohistologic evidence of osteogenesis in the rat calvarial defect model. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

11. Immunomodulatory properties of naïve and inflammationinformed dental pulp stem cell derived extracellular vesicles.

Author

Umar, Sadiq, Debnath, Koushik, Leung, Kasey, Chun-Chieh Huang, Yu Lu, Gajendrareddy, Praveen, and Ravindran, Sriram

Subjects
DENTAL pulp, EXTRACELLULAR vesicles, MESENCHYMAL stem cells, STEM cells, THERAPEUTICS
Abstract

Mesenchymal stem cell derived extracellular vesicles (MSC EVs) are paracrine modulators of macrophage function. Scientific research has primarily focused on the immunomodulatory and regenerative properties MSC EVs derived from bone marrow. The dental pulp is also a source for MSCs, and their anatomical location and evolutionary function has primed them to be potent immunomodulators. In this study, we demonstrate that extracellular vesicles derived from dental pulp stem cells (DPSC EVs) have pronounced immunomodulatory effect on primary macrophages by regulating the NFkb pathway. Notably, the anti-inflammatory activity of DPSC-EVs is enhanced following exposure to an inflammatory stimulus (LPS). These inhibitory effects were also observed in vivo. Sequencing of the naïve and LPS preconditioned DPSC-EVs and comparison with our published results from marrow MSC EVs revealed that Naïve and LPS preconditioned DPSC-EVs are enriched with anti-inflammatory miRNAs, particularly miR-320a-3p, which appears to be unique to DPSC-EVs and regulates the NFκb pathway. Overall, our findings highlight the immunomodulatory properties of DPSC-EVs and provide vital clues that can stimulate future research into miRNA-based EV engineering as well as therapeutic approaches to inflammation control and disease treatment. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

23. CABG Patients Develop Global DNA Hypermethylation, That Negatively Affect the Mitochondrial Function and Promote Post-Surgical Cognitive Decline: A Proof of Concept in Small Cohort

Author

Boovarahan, Sri Rahavi, primary, Kale, Suresh Babu, additional, Prem, Priyanka N., additional, Ravindran, Sriram, additional, Arthanarisami, Akshayakeerthi, additional, Rengaraju, Jeyashri, additional, Ali, Nemat, additional, Ramalingam, Senthilkumar, additional, Mohany, Mohamed, additional, AlAsmari, Abdullah F., additional, Al-Rejaie, Salim S., additional, Waseem, Mohammad, additional, and Kurian, Gino A., additional

Published
2023
Full Text
View/download PDF

35. Linearizing Convolutional Neural Network improves P300 detection

Author

Ravindran, Sriram

Subjects
Computer science, Cognitive psychology, BCI, Classification, Convolutional Neural Network, Deep learning, P300 Speller
Abstract

P300 is an event-related potential evoked as a response to external stimuli. The P300-speller is a widely used BCI that has proven to be a reliable method in enabling people who cannot communicate via normal methods. Improving single-trial P300 classification helps increase communication bandwidth as it reduces the averaging process necessary to reduce noise. Deep learning approaches in this area have become more popular recently [2][3]. Certain architectural choices result in a better classifier. In this paper, we performed a study of EEGNet [2] as it performed well in our experiments. We report some improvements to the model. We also confirm that these suggestions improve other CNN based models by showing improved results with CNN-1 proposed in [4] on BCI Competition III Dataset II. We perform knowledge distillation of the resultant linear model into a logistic regression model and show that the model learns information similar but superior to LDA. We show that the logistic regression model obtained this way outperforms more complex models and is easy to interpret.

Published
2018

44. sj-docx-1-jbc-10.1177_08839115221095154 – Supplemental material for Galloylated proanthocyanidins in dentin matrix exhibit biocompatibility and induce differentiation in dental stem cells

Author

Kulakowski, Daniel, Phansalkar, Rasika M, Leme-Kraus, Ariene A, McAlpine, James B, Chen, Shao-Nong, Pauli, Guido F, Ravindran, Sriram, and Bedran-Russo, Ana K

Subjects
FOS: Materials engineering, 91299 Materials Engineering not elsewhere classified
Abstract

Supplemental material, sj-docx-1-jbc-10.1177_08839115221095154 for Galloylated proanthocyanidins in dentin matrix exhibit biocompatibility and induce differentiation in dental stem cells by Daniel Kulakowski, Rasika M Phansalkar, Ariene A Leme-Kraus, James B McAlpine, Shao-Nong Chen, Guido F Pauli, Sriram Ravindran and Ana K Bedran-Russo in Journal of Bioactive and Compatible Polymers

Published
2022
Full Text
View/download PDF

50. Uptake and Distribution of Administered Bone Marrow Mesenchymal Stem Cell Extracellular Vesicles in Retina

Author

Mathew, Biji, primary, Torres, Leianne A., additional, Gamboa Acha, Lorea, additional, Tran, Sophie, additional, Liu, Alice, additional, Patel, Raj, additional, Chennakesavalu, Mohansrinivas, additional, Aneesh, Anagha, additional, Huang, Chun-Chieh, additional, Feinstein, Douglas L., additional, Mehraeen, Shafigh, additional, Ravindran, Sriram, additional, and Roth, Steven, additional

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results