Showing papers by "Delhi Technological University published in 2021"

Artificial intelligence to deep learning: machine intelligence approach for drug discovery.

Abstract: Drug designing and development is an important area of research for pharmaceutical companies and chemical scientists. However, low efficacy, off-target delivery, time consumption, and high cost impose a hurdle and challenges that impact drug design and discovery. Further, complex and big data from genomics, proteomics, microarray data, and clinical trials also impose an obstacle in the drug discovery pipeline. Artificial intelligence and machine learning technology play a crucial role in drug discovery and development. In other words, artificial neural networks and deep learning algorithms have modernized the area. Machine learning and deep learning algorithms have been implemented in several drug discovery processes such as peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modeling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Evidence from the past strengthens the implementation of artificial intelligence and deep learning in this field. Moreover, novel data mining, curation, and management techniques provided critical support to recently developed modeling algorithms. In summary, artificial intelligence and deep learning advancements provide an excellent opportunity for rational drug design and discovery process, which will eventually impact mankind. The primary concern associated with drug design and development is time consumption and production cost. Further, inefficiency, inaccurate target delivery, and inappropriate dosage are other hurdles that inhibit the process of drug delivery and development. With advancements in technology, computer-aided drug design integrating artificial intelligence algorithms can eliminate the challenges and hurdles of traditional drug design and development. Artificial intelligence is referred to as superset comprising machine learning, whereas machine learning comprises supervised learning, unsupervised learning, and reinforcement learning. Further, deep learning, a subset of machine learning, has been extensively implemented in drug design and development. The artificial neural network, deep neural network, support vector machines, classification and regression, generative adversarial networks, symbolic learning, and meta-learning are examples of the algorithms applied to the drug design and discovery process. Artificial intelligence has been applied to different areas of drug design and development process, such as from peptide synthesis to molecule design, virtual screening to molecular docking, quantitative structure-activity relationship to drug repositioning, protein misfolding to protein-protein interactions, and molecular pathway identification to polypharmacology. Artificial intelligence principles have been applied to the classification of active and inactive, monitoring drug release, pre-clinical and clinical development, primary and secondary drug screening, biomarker development, pharmaceutical manufacturing, bioactivity identification and physiochemical properties, prediction of toxicity, and identification of mode of action.

Managing supply chains for sustainable operations in the era of industry 4.0 and circular economy: Analysis of barriers

Abstract: Organizations are struggling to leverage emerging opportunities for maintaining sustainability in the global markets due to many barriers in the era of Industry 4.0 and circular economy. The main aim of this study is to analyze these barriers to improve the sustainability of a supply chain. Our study identifies the major criteria for sustainable operations and barriers that need to be overcome to achieve the objectives of sustainability through literature review and experts’ opinions. An integrated approach comprising Analytic Hierarchy Process (AHP) and Elimination and Choice Expressing Reality (ELECTRE) is used to analyze these barriers and ensure the sustainable supply chain operations. Resource circularity, increasing profits from green products, and designing processes for resource and energy efficiency have been found to be as major sustainability criteria. There are many barriers to the implementation of Industry 4.0. These barriers include but are not limited to, a lack of a skilled workforce that understands Industry 4.0, ineffective legislation and controls, ineffective performance framework, and short-term corporate goals. The study finds that ineffective strategies for the integration of industry 4.0 with sustainability measures, combined with a lack of funds for industry 4.0 initiatives, are just two of the major barriers. The findings of the study will help organizations to develop an effective and integrated strategic approach that will foster sustainable operations through the utilization of improved knowledge of Industry 4.0 and the circular economy.

InstaCovNet-19: A deep learning classification model for the detection of COVID-19 patients using Chest X-ray.

Abstract: Recently, the whole world became infected by the newly discovered coronavirus (COVID-19). SARS-CoV-2, or widely known as COVID-19, has proved to be a hazardous virus severely affecting the health of people. It causes respiratory illness, especially in people who already suffer from other diseases. Limited availability of test kits as well as symptoms similar to other diseases such as pneumonia has made this disease deadly, claiming the lives of millions of people. Artificial intelligence models are found to be very successful in the diagnosis of various diseases in the biomedical field In this paper, an integrated stacked deep convolution network InstaCovNet-19 is proposed. The proposed model makes use of various pre-trained models such as ResNet101, Xception, InceptionV3, MobileNet, and NASNet to compensate for a relatively small amount of training data. The proposed model detects COVID-19 and pneumonia by identifying the abnormalities caused by such diseases in Chest X-ray images of the person infected. The proposed model achieves an accuracy of 99.08% on 3 class (COVID-19, Pneumonia, Normal) classification while achieving an accuracy of 99.53% on 2 class (COVID, NON-COVID) classification. The proposed model achieves an average recall, F1 score, and precision of 99%, 99%, and 99%, respectively on ternary classification, while achieving a 100% precision and a recall of 99% on the binary class., while achieving a 100% precision and a recall of 99% on the COVID class. InstaCovNet-19's ability to detect COVID-19 without any human intervention at an economical cost with high accuracy can benefit humankind greatly in this age of Quarantine.

Stationary wavelet transform based ECG signal denoising method.

Abstract: Electrocardiogram (ECG) signals are used to diagnose cardiovascular diseases. During ECG signal acquisition, various noises like power line interference, baseline wandering, motion artifacts, and electromyogram noise corrupt the ECG signal. As an ECG signal is non-stationary, removing these noises from the recorded ECG signal is quite tricky. In this paper, along with the proposed denoising technique using stationary wavelet transform, various denoising techniques like lowpass filtering, highpass filtering, empirical mode decomposition, Fourier decomposition method, discrete wavelet transform are studied to denoise an ECG signal corrupted with noise. Signal-to-noise ratio, percentage root-mean-square difference, and root mean square error are used to compare the ECG signal denoising performance. The experimental result showed that the proposed stationary wavelet transform based ECG denoising technique outperformed the other ECG denoising techniques as more ECG signal components are preserved than other denoising algorithms.

Recent advances in 3D printing technologies for wearable (bio)sensors

Abstract: Wearable (bio)sensors driven through emerging three-dimensional (3D) printing technologies are currently considered the next-generation tools for various healthcare applications due to their exciting characteristics such as high stretchability, super flexibility, low cost, ultra-thinness, and lightweight. In this context, 3D printing, an emerging advanced additive manufacturing technology has revolutionized the concept of free form construction and end-user customization owing to its multifarious peculiarities that involve ease of operation, on-demand and rapid fabrication, precise and controlled deposition, as well as versatility with various soft functional materials. The customized functional structures with controllable geometry and design can be autonomously printed on the desired surfaces using the 3D printing technologies excluding the prerequisite amenities of microfabrication technologies. To accomplish this, both academics and industry experts have worked persistently to fabricate smaller, faster, and more efficient wearable devices using readily available 3D printing technologies. The contribution of 3D printing technologies in developing novel 3D structures for wearable applications using printable soft and functional materials is highlighted in this article. Moreover, the process of 3D printing along with major techniques, namely vat photopolymerization, material jetting, and material extrusion are summarized. Besides this, a number of 3D printed (bio)sensing platforms such as glucose sensors, lactate sensors, sweat sensors, strain sensors, tactile sensors, wearable oximeters, smart bandages, artificial skin, tattoo sensors, electroencephalography (EEG), electrocardiography (ECG) sensors, etc., are discussed in terms of design specifications and fabrication strategies of devices obtained via 3D printing techniques.

The impact of microplastics on marine environment: A review

Abstract: With increase in population, waste management is becoming a major issue, further recent studies also highlighted another serious issue of marine litter. It was observed that the human generated waste is accumulating in marine environment, with presence of high amounts of microplastics in water bodies such as rivers, lakes, seas and oceans. Research has highlighted that U.V light and low temperature helps in the breakdown of normal plastic into smaller pieces, which we generally referred as microplastics and through runoff, it enters into marine environment. Generally microplastics composed of polyvinyl chloride (PVC), polyethylene terephthalate (PET), polystyrene (PS) and nylon etc. With lack of efficient management, the concentration of these microplastics is increasing at an alarming rate, which not only affect the marine environment, but it is directly affecting the marine life. Some recent investigations have shown that marine environment closer to urban areas have higher levels of microplastics and aquatic animals of these areas have shown high accumulation of microplastics in their tissues. Further, it has also been reported that the other water pollutants, such as dyes, heavy metals and other chemicals can easily attach with microplastics and these microplastics also act as a carrier of other pollutants in the body of aquatic animals, which further enters into food chain. The present review provides an overview of the microplastics, its fate and its toxic effects to environment and marine health.

Strategic framework for developing resilience in Agri-Food Supply Chains during COVID 19 pandemic

Abstract: The Agri-food Supply Chains (AFSC) dealing with perishable items have suffered a lot due to the COVID 19 pandemic. During this uncertain time, developing resiliency has become a priority for manage...

Machine Learning Techniques for Heart Disease Prediction: A Comparative Study and Analysis

Abstract: Nowadays, people are getting caught in their day-to-day lives doing their work and other things and ignoring their health. Due to this hectic life and ignorance towards their health, the number of people getting sick increases every day. Moreover, most of the people are suffering from a disease like heart disease. Global deaths of almost 31% population are due to heart-related disease as data contributed by the World Health Organization (WHO). So, the prediction of happening heart disease or not becomes important for the medical field. However, data received by the medical sector or hospitals is so huge that sometimes it becomes difficult to analyze. Using machine learning techniques for this prediction and handling of data can become very efficient for medical people. Hence in this study, we have discussed the heart disease and its risk factors and explained machine learning techniques. Using that machine learning techniques, we have predicted heart disease and provided a comparative analysis of the algorithms for machine learning used for the experiment of the prediction. The goal or objective of this research is completely related to the prediction of heart disease via a machine learning technique and analysis of them.

Multivariate adaptive regression splines model for reinforced soil foundations

Abstract: In this study, a multivariate adaptive regression splines (MARS) model has been developed to predict the settlement of shallow reinforced sandy soil foundations (RSSFs). The potential of the MARS m...

Global, regional, national burden of breast cancer in 185 countries: evidence from GLOBOCAN 2018

Abstract: This study aims to examine the burden of breast cancer in 185 countries in 2018. The estimates of incidence, mortality, and prevalence of breast cancer were drawn from GLOBOCAN 2018. The overall burden of breast cancer was gauged using breast cancer burden index (BRCBI)—a novel index comprising age-standardized incidence rate (ASIR), age-standardized mortality rate (ASMR), mortality-to-incidence ratio (MIR), prevalence-to-incidence ratio (PIR), and prevalence-to-mortality ratio (PMR). The socioeconomic status of countries was measured using human development index (HDI) Globally, breast cancer was responsible for an estimated 626,679 deaths at age-standardized rate of 13/100,000; there were 2.1 million cases diagnosed in 2018 at age-standardized rate of 46.3/100,000. The ASIR varied 22-fold from 5/100,000 (Bhutan) to 113.2/100,000 (Belgium). The ASMR varied 13-fold from 2.7/100,000 (Bhutan) to 36.9/100,000 (Fiji). The HDI exhibited a positive gradient with ASIR (r = 0.73), PIR (r = 0.98), and PMR (r = 0.85); with MIR, however, it exhibited a negative association (r = − 0.83). The BRCBI spanned from 0.70 in Somalia to 78.92 in South Korea and exhibited a positive association with HDI (r = 0.76). An additional 46,823 female lives in 2018 and a cumulative total of 333,304 lives could have been saved over 2013–2018, had countries performed as per their HDI. The substantial burden of breast cancer in developing and low-resource economies calls for a holistic approach to cancer management and control that includes oncologic infrastructure to provide cost-effective screening, diagnostic, therapeutic, and palliative services, greater breast cancer awareness, and mitigation of risk factors.

A deeply coupled ConvNet for human activity recognition using dynamic and RGB images

Abstract: This work is motivated by the tremendous achievement of deep learning models for computer vision tasks, particularly for human activity recognition. It is gaining more attention due to the numerous applications in real life, for example smart surveillance system, human–computer interaction, sports action analysis, elderly healthcare, etc. Recent days, the acquisition and interface of multimodal data are straightforward due to the invention of low-cost depth devices. Several approaches have been developed based on RGB-D (depth) evidence at the cost of additional equipment’s setup and high complexity. Contrarily, the methods that utilize RGB frames provide inferior performance due to the absence of depth evidence and these approaches require to less hardware, simple and easy to generalize using only color cameras. In this work, a deeply coupled ConvNet for human activity recognition proposed that utilizes the RGB frames at the top layer with bi-directional long short-term memory (Bi-LSTM). At the bottom layer, the CNN model is trained with a single dynamic motion image. For the RGB frames, the CNN-Bi-LSTM model is trained end-to-end learning to refine the feature of the pre-trained CNN, while dynamic images stream is fine-tuned with the top layers of the pre-trained model to extract temporal information in videos. The features obtained from both the data streams are fused at decision level after the softmax layer with different late fusion techniques and achieved high accuracy with max fusion. The performance accuracy of the model is assessed using four standard single as well as multiple person activities RGB-D (depth) datasets. The highest classification accuracies achieved on human action datasets are compared with similar state of the art and found significantly higher margin such as 2% on SBU Interaction, 4% on MIVIA Action, 1% on MSR Action Pair, and 4% on MSR Daily Activity.

An intelligent approach for predicting the strength of geosynthetic-reinforced subgrade soil

Abstract: In the recent times, the use of geosynthetic-reinforced soil (GRS) technology has become popular for constructing safe and sustainable pavement structures. The strength of the subgrade soil is rout...

A Systematic Review on Firefly Algorithm: Past, Present, and Future

Abstract: Firefly Algorithm (FA) is one of the popular algorithm of Swarm Intelligence domain that can be used in most of the areas of optimization. FA and its variants are simple to implement and easily understood. These can be used to successfully solve the problems of different areas. Modification in original FA or hybrid FA algorithms are required to solve diverse range of engineering problems. In this paper, a comprehensive review of firefly algorithm is presented and various characteristics are discussed. The various variant of FA such as binary, multi-objective and hybrid with other meta-heuristics are discussed. The applications and performance evolution metric are presented. This paper provides the possible future research direction of FA.

A Parallel Military-Dog-Based Algorithm for Clustering Big Data in Cognitive Industrial Internet of Things

Abstract: With the advancement of wireless communication, Internet of Things (IoT), and big data, high performance data analytic tools and algorithms are required. Data clustering, a promising analytic technique is widely used to solve the IoT and big-data-based problems, since it does not require labeled datasets. Recently, metaheuristic algorithms have been efficiently used to solve various clustering problems. However, to handle big datasets produced from IoT devices, these algorithm fail to respond within the desired time due to high computation cost. This article presents a new metaheuristic-based clustering method to solve the big data problems by leveraging the strength of MapReduce. The proposed methods leverages the searching potential of military dog squad to find the optimal centroids and MapReduce architecture to handle the big datasets. The optimization efficacy the proposed method is validated against 17 benchmark functions, and the results are compared with five other recent algorithms, namely, bat, particle swarm optimization, artificial bee colony, multiverse optimization, and whale optimization algorithm. Furthermore, a parallel version of the proposed method is introduced using MapReduce [MapReduce-based MDBO (MR-MDBO)] for clustering the big datasets produced from industrial IoT. Moreover, the performance of MR-MDBO is studied on two benchmark UCI datasets and three real IoT-based datasets produced from industry. The F-measure and computation time of the MR-MDBO is compared with the six other state-of-the-art methods. The experimental results witness that the proposed MR-MDBO-based clustering outperforms the other considered algorithms in terms of clustering accuracy and computation times.

Gold nanomaterials for optical biosensing and bioimaging

Abstract: Gold nanoparticles (AuNPs) are highly compelling nanomaterials for biomedical studies due to their unique optical properties. By leveraging the versatile optical properties of different gold nanostructures, the performance of biosensing and biomedical imaging can be dramatically improved in terms of their sensitivity, specificity, speed, contrast, resolution and penetration depth. Here we review recent advances of optical biosensing and bioimaging techniques based on three major optical properties of AuNPs: surface plasmon resonance, surface enhanced Raman scattering and luminescence. We summarize the fabrication methods and optical properties of different types of AuNPs, highlight the emerging applications of these AuNPs for novel optical biosensors and biomedical imaging innovations, and discuss the future trends of AuNP-based optical biosensors and bioimaging as well as the challenges of implementing these techniques in preclinical and clinical investigations.

Advances in applications of Non-Destructive Testing (NDT): A review

Abstract: Manufacturing processes such as casting, rolling, forging, extrusion, material removal processes, etc are some of the common techniques used today in manufacturing industries However, these proce

A Saliva-Based RNA Extraction-Free Workflow Integrated With Cas13a for SARS-CoV-2 Detection.

Abstract: A major bottleneck in scaling-up COVID-19 testing is the need for sophisticated instruments and well-trained healthcare professionals, which are already overwhelmed due to the pandemic Moreover, the high-sensitive SARS-CoV-2 diagnostics are contingent on an RNA extraction step, which, in turn, is restricted by constraints in the supply chain Here, we present CASSPIT (Cas13 Assisted Saliva-based & Smartphone Integrated Testing), which will allow direct use of saliva samples without the need for an extra RNA extraction step for SARS-CoV-2 detection CASSPIT utilizes CRISPR-Cas13a based SARS-CoV-2 RNA detection, and lateral-flow assay (LFA) readout of the test results The sample preparation workflow includes an optimized chemical treatment and heat inactivation method, which, when applied to COVID-19 clinical samples, showed a 97% positive agreement with the RNA extraction method With CASSPIT, LFA based visual limit of detection (LoD) for a given SARS-CoV-2 RNA spiked into the saliva samples was ~200 copies; image analysis-based quantification further improved the analytical sensitivity to ~100 copies Upon validation of clinical sensitivity on RNA extraction-free saliva samples (n = 76), a 98% agreement between the lateral-flow readout and RT-qPCR data was found (Ct<35) To enable user-friendly test results with provision for data storage and online consultation, we subsequently integrated lateral-flow strips with a smartphone application We believe CASSPIT will eliminate our reliance on RT-qPCR by providing comparable sensitivity and will be a step toward establishing nucleic acid-based point-of-care (POC) testing for COVID-19