Mitochondria are an indispensable organelle for energy production and regulation of cellular meta... more Mitochondria are an indispensable organelle for energy production and regulation of cellular metabolism. The structural and functional alterations to mitochondria instigate pathological conditions of cancer, and aging-associated and neurodegenerative disorders. The normal functioning of mitochondria is maintained by quality control mechanisms involving dynamic fission, fusion, biogenesis and mitophagy. Under conditions of mitophagy and neurodegenerative diseases, mitochondria are exposed to different acidic environments and high levels of reactive oxygen species (ROS). Therefore stable molecular tools and methods are required to monitor the pathways linked to mitochondrial dysfunction and disease conditions. Herein, we report a far-red fluorescent probe (Mito-TG) with excellent biocompatibility, biostability, photostability, chemical stability and turn on emission for selective targeting of the mitochondrial matrix in different live cells. The probe was successfully employed for mon...
2017 IEEE International Conference on Image Processing (ICIP), 2017
In this work we propose to regularize the encoding and decoding weights of an autoencoder using l... more In this work we propose to regularize the encoding and decoding weights of an autoencoder using low-rank penalty in the form of nuclear norm. Such a formulation models redundancy in the network. We show that our proposed method yields better classification accuracy (on an average) and denoising results than other stochastic and deterministic regularization techniques used in deep autoencoders. Our method is also considerably faster compared to these techniques. The experiments have been carried out on benchmark deep learning datasets.
Microstrip Filters are the essential part of the microwave system and play important role in many... more Microstrip Filters are the essential part of the microwave system and play important role in many communication applications especially wireless and mobile communications. Microstrip line is one of the most popular type of planar transmission lines and is easily miniaturized and integrated with both passive and active microwave devices. These are getting popular due to their compact size, light weight, low cost and ease of fabrication. A novel microstrip planar filters for wireless communication systems has been proposed in this paper. The proposed is composed of a low pass filter, a bandpass filter, a high pass filter, and bandstop filter via wireless communication. Scientists have developed a variety of approaches over the years, which can be utilized to achieve one or more of these design objectives. Most communication system contains an RF front end which performs signal processing with RF filters. Planar based filters are popular and relatively practical to design. Band pass fi...
Advances in Intelligent Systems and Computing, 2018
With the onset of the industrial revolution, the environment is going through severe pollution le... more With the onset of the industrial revolution, the environment is going through severe pollution leading to biological imbalance. The intensity of air pollution in the world has increased at such an alarming rate that it is the need of the hour to determine the changes in the pollution pattern. Air quality dispersion modeling can be done through preferred and recommended models, the most efficient being Eulerian grid-based model. The objective of the paper is to formulate the concentrations of air pollutants using Eulerian model. Various existing methods of prediction work on the basis of models result in satisfactory outcomes but with some certain loopholes. This paper involves methods of predicting pollutant concentration and air quality using machine learning. The data of different sites of Delhi are collected, and the pollutant contributing maximum to the pollution is elucidated using machine learning based methods. Further solutions can be identified to reduce these pollutants.
ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
This paper presents a quantitative approach to demonstrate the robustness of neural networks for ... more This paper presents a quantitative approach to demonstrate the robustness of neural networks for tabular data. These data form the backbone of the data structures found in most industrial applications. We analyse the effect of various widely used techniques we encounter in neural network practice, such as regularization of weights, addition of noise to the data, and positivity constraints. This analysis is performed by using three state-of-the-art techniques, which provide mathematical proofs of robustness in terms of Lipschitz constant for feed-forward networks. The experiments are carried out on two prediction tasks and one classification task. Our work brings insights into building robust neural network architectures for safety critical systems that require certification or approval from a competent authority.
In this work, we propose a novel neural network architecture, called Adaptive Convolutional Neura... more In this work, we propose a novel neural network architecture, called Adaptive Convolutional Neural Network (ACNN), which can be viewed as an intermediate solution between a standard convolutional network and a fully connected one. A constrained training strategy is developed to learn the parameters of such a network. The proposed algorithm allows us to control the Lipschitz constant of our ACNN to secure its robustness to adversarial noise. The resulting learning approach is evaluated for signal denoising based on a database of music recordings. Both qualitative and quantitative results show that the designed network is successful in removing Gaussian noise with unknown variance.
This work addresses the problem of reconstructing biomedical signals from their lower dimensional... more This work addresses the problem of reconstructing biomedical signals from their lower dimensional projections. Traditionally Compressed Sensing (CS) based techniques have been employed for this task. These are transductive inversion processes; the problem with these approaches is that the inversion is time-consuming and hence not suitable for real-time applications. With the recent advent of deep learning, Stacked Sparse Denoising Autoencoder (SSDAE) has been used for learning inversion in an inductive setup. The training period for inductive learning is large but is very fast during application - capable of real-time speed. This work proposes a new approach for inductive learning of the inversion process. It is based on Coupled Analysis Dictionary Learning. Results on Biomedical signal reconstruction show that our proposed approach is very fast and yields result far better than CS and SSDAE.
The paper proposes a methodology for the evaluation of disassembly features of an automotive syst... more The paper proposes a methodology for the evaluation of disassembly features of an automotive system and constructs an overall metric called the Disassembly Index. For this, various attributes related to design, cost, etc. are considered. The methodology is illustrated using the case study of an automobile clutch system. It is envisaged that this will help in redesigning the system(s) with an overall objective of reducing time, effort and expenditure involved in the automobile servicing process.
In this paper we propose a joint optimization technique for coupled autoencoder which learns the ... more In this paper we propose a joint optimization technique for coupled autoencoder which learns the autoencoder weights and coupling map (between source and target) simultaneously. The technique is applicable to any transfer learning problem. In this work, we propose a new formulation that recasts deblurring as a transfer learning problem; it is solved using the proposed coupled autoencoder. The proposed technique can operate on-the-fly; since it does not require solving any costly inverse problem. Experiments have been carried out on state-of-the-art techniques; our method yields better quality images in shorter operating times.
A persulfide/hydrogen sulfide generation strategy through artificial substrates for 3-mercaptopyr... more A persulfide/hydrogen sulfide generation strategy through artificial substrates for 3-mercaptopyruvate sulfurtransferase (3-MST) is reported, which enhances cellular persulfides, attenuates reactive oxygen species (ROS), and alleviates inflammation.
The SARS-COV-2 pandemic has put pressure on Intensive Care Units, and made the identification of ... more The SARS-COV-2 pandemic has put pressure on Intensive Care Units, and made the identification of early predictors of disease severity a priority. We collected clinical, biological, chest CT scan data, and radiology reports from 1,003 coronavirus-infected patients from two French hospitals. Among 58 variables measured at admission, 11 clinical and 3 radiological variables were associated with severity. Next, using 506,341 chest CT images, we trained and evaluated deep learning models to segment the scans and reproduce radiologists’ annotations. We also built CT image-based deep learning models that predicted severity better than models based on the radiologists’ reports. Finally, we showed that adding CT scan information—either through radiologist lesion quantification or through deep learning—to clinical and biological data, improves prediction of severity. These findings show that CT scans contain novel and unique prognostic information, which we included in a 6-variable ScanCov se...
This work proposes a new formulation for supervised stacked autoencoder. We argue that features f... more This work proposes a new formulation for supervised stacked autoencoder. We argue that features from the same class should be similar to each other and hence linearly dependent. This means that, when stacked as columns, the feature matrix for each class will be rank deficient (low-rank). We impose this constraint into the stacked autoencoder formulation in the form of nuclear norm penalties on class-wise feature matrices at each level. The nuclear norm penalty is the convex surrogate of rank, and promotes a low-rank solution as desired by our proposal. Owing to the nuclear norm penalties, our formulation is non-smooth; hence cannot be solved using gradient descent based techniques like backpropagation directly. Moreover we learn the stacked autoencoder in one go, without the usual pre-training followed by fine-tuning regime. Both the ends (non-smooth cost function and single stage training for all the layers simultaneously) are met by employing the variable splitting followed by augmented Lagrangian method of alternating directions. Two sets of experiments have been carried out. The first set is on a variety of benchmark datasets. Our method excels over other deep learning models compared against—class sparse stacked autoencoder, deep belief network and discriminative deep belief network. The second experiment is on the brain computer classification problem; we find that our method outperforms prior deep learning based solutions utilized for this task.
The paper presents a new universal biquadratic filter using single universal voltage conveyor (UV... more The paper presents a new universal biquadratic filter using single universal voltage conveyor (UVC), two resistors and two capacitors. The offered structure has three inputs and one output and can realise all the five basic biquad-ratic filters: high-pass (HP), low-pass (LP), band-reject (BR), band-pass (BP) and all-pass (AP) from the same circuit topology. The proposed universal filter also provides following advantageous features, not available simultaneously in any UVC based universal biquadratic filter so far: (i) low active and passive sensitivities, (ii) independent control of natural frequency (ω 0) and bandwidth (BW) and (iii) no requirement of any component matching condition and inversion of input signal(s) (as needed in most of the earlier reported structures). The workability of proposed structure has been presented by SPICE (Version 16.5) simulation using 0.18 µm TSMC technology.
Generation of hydrogen sulfide (H2S) is challenging and few methods are capable of localized deli... more Generation of hydrogen sulfide (H2S) is challenging and few methods are capable of localized delivery of this gas. Here, a boron dipyrromethene-based carbamothioate (BDP-H2S) that is uncaged by visible light of 470 nm to generate carbonyl sulfide (COS), which is rapidly hydrolyzed to H2S in the presence of carbonic anhydrase, a widely prevalent enzyme, is reported.
Mitochondria are an indispensable organelle for energy production and regulation of cellular meta... more Mitochondria are an indispensable organelle for energy production and regulation of cellular metabolism. The structural and functional alterations to mitochondria instigate pathological conditions of cancer, and aging-associated and neurodegenerative disorders. The normal functioning of mitochondria is maintained by quality control mechanisms involving dynamic fission, fusion, biogenesis and mitophagy. Under conditions of mitophagy and neurodegenerative diseases, mitochondria are exposed to different acidic environments and high levels of reactive oxygen species (ROS). Therefore stable molecular tools and methods are required to monitor the pathways linked to mitochondrial dysfunction and disease conditions. Herein, we report a far-red fluorescent probe (Mito-TG) with excellent biocompatibility, biostability, photostability, chemical stability and turn on emission for selective targeting of the mitochondrial matrix in different live cells. The probe was successfully employed for mon...
2017 IEEE International Conference on Image Processing (ICIP), 2017
In this work we propose to regularize the encoding and decoding weights of an autoencoder using l... more In this work we propose to regularize the encoding and decoding weights of an autoencoder using low-rank penalty in the form of nuclear norm. Such a formulation models redundancy in the network. We show that our proposed method yields better classification accuracy (on an average) and denoising results than other stochastic and deterministic regularization techniques used in deep autoencoders. Our method is also considerably faster compared to these techniques. The experiments have been carried out on benchmark deep learning datasets.
Microstrip Filters are the essential part of the microwave system and play important role in many... more Microstrip Filters are the essential part of the microwave system and play important role in many communication applications especially wireless and mobile communications. Microstrip line is one of the most popular type of planar transmission lines and is easily miniaturized and integrated with both passive and active microwave devices. These are getting popular due to their compact size, light weight, low cost and ease of fabrication. A novel microstrip planar filters for wireless communication systems has been proposed in this paper. The proposed is composed of a low pass filter, a bandpass filter, a high pass filter, and bandstop filter via wireless communication. Scientists have developed a variety of approaches over the years, which can be utilized to achieve one or more of these design objectives. Most communication system contains an RF front end which performs signal processing with RF filters. Planar based filters are popular and relatively practical to design. Band pass fi...
Advances in Intelligent Systems and Computing, 2018
With the onset of the industrial revolution, the environment is going through severe pollution le... more With the onset of the industrial revolution, the environment is going through severe pollution leading to biological imbalance. The intensity of air pollution in the world has increased at such an alarming rate that it is the need of the hour to determine the changes in the pollution pattern. Air quality dispersion modeling can be done through preferred and recommended models, the most efficient being Eulerian grid-based model. The objective of the paper is to formulate the concentrations of air pollutants using Eulerian model. Various existing methods of prediction work on the basis of models result in satisfactory outcomes but with some certain loopholes. This paper involves methods of predicting pollutant concentration and air quality using machine learning. The data of different sites of Delhi are collected, and the pollutant contributing maximum to the pollution is elucidated using machine learning based methods. Further solutions can be identified to reduce these pollutants.
ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
This paper presents a quantitative approach to demonstrate the robustness of neural networks for ... more This paper presents a quantitative approach to demonstrate the robustness of neural networks for tabular data. These data form the backbone of the data structures found in most industrial applications. We analyse the effect of various widely used techniques we encounter in neural network practice, such as regularization of weights, addition of noise to the data, and positivity constraints. This analysis is performed by using three state-of-the-art techniques, which provide mathematical proofs of robustness in terms of Lipschitz constant for feed-forward networks. The experiments are carried out on two prediction tasks and one classification task. Our work brings insights into building robust neural network architectures for safety critical systems that require certification or approval from a competent authority.
In this work, we propose a novel neural network architecture, called Adaptive Convolutional Neura... more In this work, we propose a novel neural network architecture, called Adaptive Convolutional Neural Network (ACNN), which can be viewed as an intermediate solution between a standard convolutional network and a fully connected one. A constrained training strategy is developed to learn the parameters of such a network. The proposed algorithm allows us to control the Lipschitz constant of our ACNN to secure its robustness to adversarial noise. The resulting learning approach is evaluated for signal denoising based on a database of music recordings. Both qualitative and quantitative results show that the designed network is successful in removing Gaussian noise with unknown variance.
This work addresses the problem of reconstructing biomedical signals from their lower dimensional... more This work addresses the problem of reconstructing biomedical signals from their lower dimensional projections. Traditionally Compressed Sensing (CS) based techniques have been employed for this task. These are transductive inversion processes; the problem with these approaches is that the inversion is time-consuming and hence not suitable for real-time applications. With the recent advent of deep learning, Stacked Sparse Denoising Autoencoder (SSDAE) has been used for learning inversion in an inductive setup. The training period for inductive learning is large but is very fast during application - capable of real-time speed. This work proposes a new approach for inductive learning of the inversion process. It is based on Coupled Analysis Dictionary Learning. Results on Biomedical signal reconstruction show that our proposed approach is very fast and yields result far better than CS and SSDAE.
The paper proposes a methodology for the evaluation of disassembly features of an automotive syst... more The paper proposes a methodology for the evaluation of disassembly features of an automotive system and constructs an overall metric called the Disassembly Index. For this, various attributes related to design, cost, etc. are considered. The methodology is illustrated using the case study of an automobile clutch system. It is envisaged that this will help in redesigning the system(s) with an overall objective of reducing time, effort and expenditure involved in the automobile servicing process.
In this paper we propose a joint optimization technique for coupled autoencoder which learns the ... more In this paper we propose a joint optimization technique for coupled autoencoder which learns the autoencoder weights and coupling map (between source and target) simultaneously. The technique is applicable to any transfer learning problem. In this work, we propose a new formulation that recasts deblurring as a transfer learning problem; it is solved using the proposed coupled autoencoder. The proposed technique can operate on-the-fly; since it does not require solving any costly inverse problem. Experiments have been carried out on state-of-the-art techniques; our method yields better quality images in shorter operating times.
A persulfide/hydrogen sulfide generation strategy through artificial substrates for 3-mercaptopyr... more A persulfide/hydrogen sulfide generation strategy through artificial substrates for 3-mercaptopyruvate sulfurtransferase (3-MST) is reported, which enhances cellular persulfides, attenuates reactive oxygen species (ROS), and alleviates inflammation.
The SARS-COV-2 pandemic has put pressure on Intensive Care Units, and made the identification of ... more The SARS-COV-2 pandemic has put pressure on Intensive Care Units, and made the identification of early predictors of disease severity a priority. We collected clinical, biological, chest CT scan data, and radiology reports from 1,003 coronavirus-infected patients from two French hospitals. Among 58 variables measured at admission, 11 clinical and 3 radiological variables were associated with severity. Next, using 506,341 chest CT images, we trained and evaluated deep learning models to segment the scans and reproduce radiologists’ annotations. We also built CT image-based deep learning models that predicted severity better than models based on the radiologists’ reports. Finally, we showed that adding CT scan information—either through radiologist lesion quantification or through deep learning—to clinical and biological data, improves prediction of severity. These findings show that CT scans contain novel and unique prognostic information, which we included in a 6-variable ScanCov se...
This work proposes a new formulation for supervised stacked autoencoder. We argue that features f... more This work proposes a new formulation for supervised stacked autoencoder. We argue that features from the same class should be similar to each other and hence linearly dependent. This means that, when stacked as columns, the feature matrix for each class will be rank deficient (low-rank). We impose this constraint into the stacked autoencoder formulation in the form of nuclear norm penalties on class-wise feature matrices at each level. The nuclear norm penalty is the convex surrogate of rank, and promotes a low-rank solution as desired by our proposal. Owing to the nuclear norm penalties, our formulation is non-smooth; hence cannot be solved using gradient descent based techniques like backpropagation directly. Moreover we learn the stacked autoencoder in one go, without the usual pre-training followed by fine-tuning regime. Both the ends (non-smooth cost function and single stage training for all the layers simultaneously) are met by employing the variable splitting followed by augmented Lagrangian method of alternating directions. Two sets of experiments have been carried out. The first set is on a variety of benchmark datasets. Our method excels over other deep learning models compared against—class sparse stacked autoencoder, deep belief network and discriminative deep belief network. The second experiment is on the brain computer classification problem; we find that our method outperforms prior deep learning based solutions utilized for this task.
The paper presents a new universal biquadratic filter using single universal voltage conveyor (UV... more The paper presents a new universal biquadratic filter using single universal voltage conveyor (UVC), two resistors and two capacitors. The offered structure has three inputs and one output and can realise all the five basic biquad-ratic filters: high-pass (HP), low-pass (LP), band-reject (BR), band-pass (BP) and all-pass (AP) from the same circuit topology. The proposed universal filter also provides following advantageous features, not available simultaneously in any UVC based universal biquadratic filter so far: (i) low active and passive sensitivities, (ii) independent control of natural frequency (ω 0) and bandwidth (BW) and (iii) no requirement of any component matching condition and inversion of input signal(s) (as needed in most of the earlier reported structures). The workability of proposed structure has been presented by SPICE (Version 16.5) simulation using 0.18 µm TSMC technology.
Generation of hydrogen sulfide (H2S) is challenging and few methods are capable of localized deli... more Generation of hydrogen sulfide (H2S) is challenging and few methods are capable of localized delivery of this gas. Here, a boron dipyrromethene-based carbamothioate (BDP-H2S) that is uncaged by visible light of 470 nm to generate carbonyl sulfide (COS), which is rapidly hydrolyzed to H2S in the presence of carbonic anhydrase, a widely prevalent enzyme, is reported.
Uploads