Search

With the rise of powerful computational technologies, healthcare systems are going through a paradigm shift to the era of healthcare 5.0, also known as" smart healthcare". This new-age healthcare system not only enhances the life of patients but also aims to reduce healthcare costs significantly. People's health data collected through various means, including various Internet of Things (IoT) devices, is too sensitive to be regulated by particular organisations owing to increasing privacy and security threats. The sheer size of data makes it difficult for centralised bodies to regulate, due to which third-party involvement is introduced, which again poses threats to the privacy of the data holder. Motivated by the abovementioned gaps, we propose a decentralised approach using Ethereum blockchain technology and Oasis protocol. The patient-driven solution approach effectively solves the privacy problem by providing complete control of data to the owner. Along with this, byproducts such as fine-grained access control and on-chain data processing capabilities are also obtained.

Emotion recognition is a very challenging research field due to its complexity, as individual differences in cognitive–emotional cues involve a wide variety of ways, including language, expressions, and speech. If we use video as the input, we can acquire a plethora of data for analyzing human emotions. In this research, we use features derived from separately pretrained self-supervised learning models to combine text, audio (speech), and visual data modalities. The fusion of features and representation is the biggest challenge in multimodal emotion classification research. Because of the large dimensionality of self-supervised learning characteristics, we present a unique transformer and attention-based fusion method for incorporating multimodal self-supervised learning features that achieved an accuracy of 86.40% for multimodal emotion classification.

Image processing plays an indispensable and significant role in the development of various fields like medical imaging, astronomy, GIS, disaster management, agriculture monitoring, and so on. Medical images which are recorded in digital forms are processed by high-end computers to extract whatever information we desire. In the fast-developing modern world of medical imaging diagnosis and prognosis, where manual photo interpretation is time-consuming, automatic object detection from devices like CT-Scans and MRIs has limited potential to generate the required results. This article addresses the process of identifying Region of Interests in cancer based medical images based on combination of Otsu’s algorithm and Canny edge detection methods. The primary objective of this paper is to derive meaningful and potential information from medical image in different scenarios by applying the image segementation in combination with genetic algorithms in a robust manner to detect region of interest.

The remote sensing surveillance of maritime areas represents an essential task for both security and environmental reasons. Recently, learning strategies belonging to the field of machine learning (ML) have become a niche of interest for the community of remote sensing. Specifically, a major challenge is the automatic classification of ships from satellite imagery, which is needed for traffic surveillance systems, the protection of illegal fisheries, control systems of oil discharge, and the monitoring of sea pollution. Deep learning (DL) is a branch of ML that has emerged in the last few years as a result of advancements in digital technology and data availability. DL has shown capacity and efficacy in tackling difficult learning tasks that were previously intractable. Specifically, DL methods, such as convolutional neural networks (CNNs), have been reported to be efficient in image detection and recognition applications. In this paper, we focused on the development of an automatic ship detection (ASD) approach by using DL methods for assessing the Airbus ship dataset (composed of about 40 K satellite images). The paper explores and analyzes the distinct variations of the YOLO algorithm for the detection of ships from satellite images. A comparison of different versions of YOLO algorithms for ship detection, such as YOLOv3, YOLOv4, and YOLOv5, is presented, after training them on a personal computer with a large dataset of satellite images of the Airbus Ship Challenge and Shipsnet. The differences between the algorithms could be observed on the personal computer. We have confirmed that these algorithms can be used for effective ship detection from satellite images. The conclusion drawn from the conducted research is that the YOLOv5 object detection algorithm outperforms the other versions of the YOLO algorithm, i.e., YOLOv4 and YOLOv3 in terms accuracy of 99% for YOLOv5 compared to 98% and 97% respectively for YOLOv4 and YOLOv3.

Objective: Malignancy-related ascites (MRA) is the terminal stage of many advanced cancers, and the treatment is mainly palliative. This study looked for epidemiology and inpatient hospital outcomes of patients with MRA in the United States using a national database. Methods: The current study was a cross-sectional analysis of 2015 National Inpatient Sample data and consisted of patients ≥18 years with MRA. Descriptive statistics were used for understanding demographics, clinical characteristics, and MRA hospitalization costs. Multivariate regression models were used to identify predictors of length of hospital stay and in-hospital mortality. Results: There were 123 410 MRA hospitalizations in 2015. The median length of stay was 4.7 days (interquartile range [IQR]: 2.5-8.6 days), median cost of hospitalization was US$43 543 (IQR: US$23 485-US$82 248), and in-hospital mortality rate was 8.8% (n = 10 855). Multivariate analyses showed that male sex, black race, and admission to medium and large hospitals were associated with increased hospital length of stay. Factors associated with higher in-hospital mortality rates included male sex; Asian or Pacific Islander race; beneficiaries of private insurance, Medicaid, and self-pay; patients residing in large central and small metro counties; nonelective admission type; and rural and urban nonteaching hospitals. Conclusions: Our study showed that many demographic, socioeconomic, health care, and geographic factors were associated with hospital length of stay and in-hospital mortality and may suggest disparities in quality of care. These factors could be targeted for preventing unplanned hospitalization, decreasing hospital length of stay, and lowering in-hospital mortality for this population.

A new aerophilic species, Diploneis mawsmaii sp. nov., is here described from a speleo-environment. The species is found on mosses growing under the water dripping region of Mawsmai cave, Meghalaya state of India. It is discussed based on detailed light and scanning electron microscopic observations and compared with morphologically similar Diploneis species like D. modica Hustedt, D. modicahassiaca Lange-Bertalot & A. Fuhrmann, D. boldtiana Cleve and D. zula Kulikovskiy & Lange Bertalot. Diploneis mawsmaii possesses distinct morphological features such as a largely expanded central area bearing a prominent circular to oval central nodule and the internal structure exhibiting short striation with acute ends and the flattening of longitudinal canals at the central area. This discovery increases our understanding of cavern biodiversity in the Indo-Burma hotspot region.

OBJECTIVE The objective of this study was to understand recent trends in direct health care expenditures among cancer survivors using novel cost-estimation methods and a nationally representative database. MATERIALS AND METHODS This study was a retrospective analysis of 193,003 adults, ≥18 years of age, using the Medical Expenditure Panel Survey during the years 2009-2016. Manning and Mullahy two-part model was used to calculate adjusted mean and incremental medical expenditures after adjusting for covariates. RESULTS The mean direct annual health care expenditure among cancer survivors ($13,025.0 [$12,572.0 to $13,478.0]) was nearly 3 times greater than noncancer participants ($4689.3 [$4589.2 to $4789.3]) and were mainly spent on inpatient services, office-based visits, and prescription medications. Cancer survivors had an additional health care expenditure of $4407.6 ($3877.6, $4937.6) per person per year, compared with noncancer participants after adjusting for covariates (P

Network attacks are continuously surging, and attackers keep on changing their ways in penetrating a system. A network intrusion detection system is created to monitor traffic in the network and to warn regarding the breach in security by invading foreign entities in the network. Specific experiments have been performed on the NSL-KDD dataset instead of the KDD dataset because it does not have redundant data so the output produced from classifiers will not be biased. The main types of attacks are divided into four categories: denial of service (DoS), probe attack, user to root attack (U2R), remote to local attack (R2L). Overall, this chapter proposes an intense study on linear and ensemble models such as logistic regression, stochastic gradient descent (SGD), naïve bayes, light GBM (LGBM), and XGBoost. Lastly, a stacked model is developed that is trained on the above-mentioned classifiers, and it is applied to detect intrusion in networks. From the plethora of approaches taken into consideration, the authors have found maximum accuracy (98.6%) from stacked model and XGBoost.

One of the most critical issues that the marine surveillance system has to address is the accuracy of its ship detection. Since it is responsible for identifying potential pirate threats, it has to be able to perform its duties efficiently. In this paper, we present a novel deep learning approach that combines the capabilities of a Graph Neural Network (GNN) and a You Only Look Once (YOLOv7) deep learning framework. The main idea of this method is to provide a better understanding of the ship’s presence in harbor areas. The three hyperparameters that are used in the development of this system are the learning rate, batch sizes, and optimization selection. The results of the experiments show that the Adam optimization achieves a 93.4% success rate when compared to the previous generation of the YOLOv7 algorithm. The High-Resolution Satellite Image Dataset (HRSID), which is a high-resolution image of a synthetic aperture radar, was used for the test. This method can be further improved by taking into account the various kinds of neural network architecture that are commonly used in deep learning.

With the large-scale improvements in communication technology and easy accessibility of the same has led to the advent of the IoT technology, which in turn has resulted in more, and more devices being IoT enabled. This research paper embarks a next generation smart mirror with the aim of connecting the conventional mirror into the IoT network along with a certain set of features, which were never introduced before. The whole range of features focuses on increasing the utility of daily mirrors. This mirror along with acting as a piece of glass will also be acting as a huge glass notepad upon which the user can leave any handwritten messages which are to be conveyed to others and can never go unnoticed. Keeping in mind the smudging effect and the fingerprint mark that are left behind when the mirror surface is touched, this new feature allows one to write the message on the mirror without touching the mirror surface itself. Additionally, the mirror will also support some of common features as well namely date, weather forecast, news headlines and daily remainders.

The main purpose of this work is to investigate and compare several deep learning enhanced techniques applied to X-ray and CT-scan medical images for the detection of COVID-19. In this paper, we used four powerful pre-trained CNN models, VGG16, DenseNet121, ResNet50,and ResNet152, for the COVID-19 CT-scan binary classification task. The proposed Fast.AI ResNet framework was designed to find out the best architecture, pre-processing, and training parameters for the models largely automatically. The accuracy and F1-score were both above 96% in the diagnosis of COVID-19 using CT-scan images. In addition, we applied transfer learning techniques to overcome the insufficient data and to improve the training time. The binary and multi-class classification of X-ray images tasks were performed by utilizing enhanced VGG16 deep transfer learning architecture. High accuracy of 99% was achieved by enhanced VGG16 in the detection of X-ray images from COVID-19 and pneumonia. The accuracy and validity of the algorithms were assessed on X-ray and CT-scan well-known public datasets. The proposed methods have better results for COVID-19 diagnosis than other related in literature. In our opinion, our work can help virologists and radiologists to make a better and faster diagnosis in the struggle against the outbreak of COVID-19. The research leading to these results received funding from the Innovative Medicines Innitiative 2 Joint Undertaking (JU) under grant agreement No 853989. The JU receives support from the European Union's Horizon 2020 research and innovation programme and EFPIA and Global Alliance for TB Drug Development non profit organisation, Bill & Melinda Gates Foundation and University of Dundee.

Background: Cancer and cardiovascular diseases (CVDs) are leading causes of morbidity and mortality. We analyzed national data to examine the prevalence of CVD risk factors among adult can...

A patient tracing programme was implemented at an HIV clinic in Trinidad and Tobago to address the problem of defaulters from HIV care and non-adherence to antiretroviral treatment (ART). The study objective was to evaluate the implementation and outcomes of this programme conducted between April and September 2017. Using patient tracing contact methods, trained social workers attempted to contact 1058 patients lost to follow up (LTFU) between July 2016 and March 2017. Of the 1058 LTFU, 192 were ineligible: 27 (2.5%) were transferred to another clinic, 64 (6%) deceased, 35 (3.3%) hospitalised, 50 (4.7%) migrated and 16 (1.5%) incarcerated. Of the 866 eligible patients for patient tracing, 277 (32%) remained permanently LTFU and 589 (68%) were successfully contacted, re-engaged in care and received adherence counselling. Of the 589 who returned to care, 507 (86%) restarted ART. The three most common barriers reported among the 589 who were reengaged were 'forgetting their appointments' (20%), 'being too busy/work' (16%), and 'not wanting to be seen attending the HIV clinic' (12%). The study findings demonstrated the tracing programme as feasible for re-engaging those who are LTFU and highlighted barriers that can be addressed to further improve retention in HIV care among people living with HIV.

While the national HIV infection rate is decreasing, the highest rates of infections continue among men who have sex with men (MSM), particularly minority MSM. It is important to understand attitudes, knowledge, and behaviors surrounding HIV prevention methods, such as pre-exposure prophylaxis (PrEP). In the present study, we created a snapshot of the PrEP continuum of care and identified participant demographic and sources of PrEP awareness factors that were associated with PrEP initiation. Data were collected using anonymous paper–based surveys employing a venue intercept procedure. A total of 188 HIV–negative men completed the survey at Miami Gay Pride 2018. Participants answered questions regarding demographics, PrEP use, and sources of PrEP awareness. The sample was majority Hispanic (55.4%), gay (83.0%), and single (57.7%). The constructed PrEP continuum revealed that a low proportion of those identified as PrEP naïve (n = 143) for HIV infection had PrEP interest (49/143). Moreover, among those who initiated PrEP (n = 45), a high proportion were retained in a PrEP program (37/45), with approximately half achieving medication adherence (25/45). Age group, PrEP knowledge, and source of PrEP awareness were all significantly associated with PrEP initiation. In areas with high HIV infection rates, studies like these offer crucial insight on how public health practitioners should proceed in the goal of decreasing HIV transmission rates. More research is needed to increase PrEP uptake and adherence.

In several fields nowadays, automated emotion recognition has been shown to be a highly powerful tool. Mapping different facial expressions to their respective emotional states is the main objective of facial emotion recognition (FER). In this study, facial expression recognition (FER) was classified using the ResNet-18 model and transformers. This study examines the performance of the Vision Transformer in this task and contrasts our model with cutting-edge models on hybrid datasets. The pipeline and associated procedures for face detection, cropping, and feature extraction using the most recent deep learning model, fine-tuned transformer, are described in this study. The experimental findings demonstrate that our proposed emotion recognition system is capable of being successfully used in practical settings.

Manipulation of a large amount of data becomes a very tedious task. Hence, the authors took the approach of memory networks for the implementation of the chatbot. Traditionally, the LSTM model was used to implement chatbots and QA systems. But the LSTM failed to store relevant information when given a longer information set. On the contrary, the memory networks have an additional memory component with it. This can help in storing long information for further use which is greatly advantageous for the QA and chatbot systems as compared to LSTM. The authors trained and tested their model over Facebook's bAbi dataset which consists of several tasks and has questions regarding each task to retrieve the accuracy of the model. On the pedestal of that dataset, they have presented the accuracy for every task in their study with memory networks.

The concept of the Internet of Energy is derived from the concept of the Internet of Things (IoT), which uses the Internet to access and connect appliances. Energy consumption in heating, ventilation and air conditioning typically accounts for half of energy consumption in buildings. Wireless Home Automation System involves communications, control, and information processing across various systems all done over the internet using IoT. The current prototype has some limitations which are now in the modification stage and are being worked upon for the development of the system. The system facing the issue of high latency in HTTP requests and responses can be resolved by developing more efficient APIs. The system can be made compatible with more electrical appliances unlike the prototype currently compatible only with light appliances. The application of the system can be made available for both iOS and Android platforms with complete integration of all modules with higher efficiency.

Fog computing aims to distribute the storage and processing of data at the network edges to efficiently exploit the capacities of the network. It provides the opportunity to execute interactive and real-time computation-intensive or data-intensive applications. Context-awareness plays an important role in understanding the situation of the environment and acting, accordingly. It has entered fog computing area and several fog-based context-aware systems have been proposed. This survey aims to investigate fog-based context-aware systems. To this end, a framework with the dimensions of domain type and context-awareness is proposed for investigating and classifying the related research. The aim is to technically investigate fog computing, to classify related research, to foster the dissemination of state-of-the-art, and to discuss open research directions in this field.

India has globally been the largest milk-producing country in the world for two decades. About 400 million litres of milk is produced every day. It is the responsibility of a dairy sector to look after the farmers by providing them with various services for their livelihood. The growing financial capital of the dairy industry has enticed various fraudulent behaviour. The majority of suspicious activities are seen during the collection at local collection centres, fake farmer entries, tempered quantity and fat entries manually, and adulteration are the profound malpractices exercised by farmers. So, in this research work, the authors present a profound study on the most popular machine learning methods applied to the problems of farmer churn prediction and fraud detection in the dairies. They applied a plethora of machine learning algorithms to get accurate results for churn and fraud detection. XGBoost Classifier was the best for churn prediction with 93% accuracy, while random forest classifier turns out to be effective for fraud detection with 94% accuracy.

The issue of security is paramount in any organisation. Therefore, the authors intend to aid in the security of such organisations by bringing a video based human authentication system for access control which is a type of cyber physical system (CPS). CPS is an integration of computation and physical processes; here the computation is provided by face detection and recognition algorithm and physical process is the input human face. This system aims to provide a platform that allows any authorized person to enter or leave the premise automatically by using face detection and recognition technology. The system also provides the administrator with the access to the logs, wherein he/she would be able to access the details of the people entering or leaving the organisation along with the live video streaming so that there is no sneaking of any unauthorized person with any other authorized person. The administrator can also do registration on behalf of a new person who requires access to the premises for a restricted amount of time only as specified by the administrator.

The predominant means of communication is speech; however, there are persons whose speaking or hearing abilities are impaired. Communication presents a significant barrier for persons with such disabilities. The use of deep learning methods can help to reduce communication barriers. This paper proposes a deep learning-based model that detects and recognizes the words from a person’s gestures. Deep learning models, namely, LSTM and GRU (feedback-based learning models), are used to recognize signs from isolated Indian Sign Language (ISL) video frames. The four different sequential combinations of LSTM and GRU (as there are two layers of LSTM and two layers of GRU) were used with our own dataset, IISL2020. The proposed model, consisting of a single layer of LSTM followed by GRU, achieves around 97% accuracy over 11 different signs. This method may help persons who are unaware of sign language to communicate with persons whose speech or hearing is impaired.

OBJECTIVE The objective of this study was to understand the trends and characteristics of palliative care delivery among critically ill brain metastasis patients using a nationally representative database. MATERIALS AND METHODS This study was a retrospective analysis of Nationwide Inpatient Sample data collected during 2005 to 2014. This study included critically ill patients, 18 years and above, diagnosed with brain metastasis, identified using International Classification of Diseases-Ninth Revision-Clinical Modification diagnosis and procedure codes. Multivariable logistic regression models were used for predicting factors associated with inpatient palliative care use. RESULTS Among 18,309 critically ill patients with brain metastasis, 3298 (18.0%) received inpatient palliative care. The rate of inpatient palliative care use among these patients increased from 3.2% to 28.5%, during 2005 to 2014 (P

A significant application has been found in making the task of automation easier. Computer vision is a very powerful method for understanding the contents from the images. Weighing sheep for the analysis of their health and growth is a very arduous task, and traditional methods of weighing using a balance are very time-consuming. Estimating sheep weights using computer vision algorithms has already been tried, but these methods are not efficient enough. They also require some extra instruments set-up. We tried to utilize this powerful technology to make the difficult task of estimating sheep weights quick and accurate. We bring a novel approach to image segmentation in order to calculate the area of the sheep directly. The various other features are fed to a neural network-based regression model for achieving better results for the task of estimating sheep’s weight. The proposed approach has enabled us to minimize the complexities of carrying heavy equipment and errors through human involvement in measuring the weight of the sheep.

Introduction: Statins use for secondary cardiovascular disease (CVD) prevention is associated with better health outcomes and lower cost among adults with ASCVD. We aimed to examine the prevalence of statin use among adults with ASCVD in the US and to assess for gender and racial differences in statin use. Methods: We identified adults aged 40-75 years with ASCVD (defined using ICD codes and self-reported history) from the 2008-17 Medical Expenditure Panel Survey (MEPS) data. We estimated the trends in statin use among them, and used joint point regression (JPR) to estimate the annual percentage change in statin use. Logistic regression was used to assess the racial variation in statin use. We applied the weights in the MEPS complex survey design to estimate population sizes, so all results are nationally representative. Results: Between 2008 and 2017, 24896 adults aged 40-75 years with ASCVD were identified. In 2017, there were 2445 patients with ASCVD (equivalent to 24.5 million adults). In 2017, the mean age [SE] was 62.2 [0.3] years, and 42.9% were females. The overall proportion of adults with ASCVD using statin increased from 50.0% in 2008 to 58.7% in 2017, with an average annual percentage change of 0.95% between 2010 and 2017 ( p= 0.01). As shown in Figure 1a , among those with ASCVD, statin use was persistently lower in women than in men. In 2017, compared to White adults with ASCVD, Blacks (OR=0.69; 95% CI: 0.51-0.92) and Hispanics (OR=0.62; 95% CI: 0.45-0.85) were less likely to be on statin. Each year, over 3 million women with ASCVD were not on statin, and majority of them were Blacks and Hispanics. (Figure 1b) Conclusion: Millions of adults with ASCVD are not on statins in the US while gender and racial inequalities in statin use remains, despite persistent guideline recommendations. Herein lies an opportunity to improve CVD-related outcomes and cost by intensifying efforts to use statins for the secondary prevention of CVD and closing gender and racial gaps, especially with the 2030 sustainable development goals in view.

Background: Major adverse cardiovascular and cerebrovascular events (MACCE) are significant causes of perioperative morbidity and mortality but, the incidence and effects following cancer surgeries are unknown. The aims of this study were to evaluate national trends in MACCE after major cancer surgeries and to identify cancer types associated with cardiovascular events using a large national database. Methods: Patients who had major cancer surgeries from 2005 to 2014 were identified from the National Inpatient Sample database. Hospitalizations for surgeries for cancer of prostate, bladder, esophagus, pancreas, lung, liver, breast, colon and rectum were identified by ICD9 diagnosis and procedure codes. The main outcome was perioperative MACCE, defined as in-hospital, all-cause death, acute myocardial infarction (AMI), or acute ischemic stroke, and was evaluated over time. Results: Among 2,854,810 hospitalizations for major cancer surgeries, perioperative MACCE occurred in 67,316 hospitalizations (2.4%). Mean (SE) age of patients was 65.4 (0.07) years and 54.2% were male patients. MACCE occurred most frequently in patients undergoing surgeries for lung (6.8%), pancreatic (4.5%), and colorectal (3.3%) cancers. Between 2005 and 2014, the frequency of MACCE declined from 2.7% to 2.2% ( P P P = 0.002). However, no significant changes were observed for acute ischemic stroke ( P = 0.6) during the study period. Conclusion: Perioperative MACCE occurs in 1 out of every 42 hospitalizations for major cancer surgeries. Despite reductions in the rate of death and AMI among patients undergoing major cardiac surgeries, perioperative ischemic stroke remained constant over time. The lack of improvements in perioperative ischemic stroke rate is concerning and requires additional interventions. Significant efforts should be directed towards improving cardiovascular care during the perioperative period of cancer surgeries.

Introduction: Cardiometabolic syndrome consists of a cluster of metabolic dysfunctions such as impaired glucose tolerance, insulin resistance, dyslipidemia, central adiposity, and hypertension. According to the latest estimates, globally, nearly 25% of all adults have cardiometabolic syndrome. Both cardiometabolic syndrome and cancer pathophysiology commonly involve inflammation and oxidative stress. The objective of this systematic review was to evaluate existing evidences that support the association between cardiometabolic syndrome and risk of developing cancer. Methods: We searched MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and Scopus for relevant articles published from the database inception until October 2019. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines for this review. Using the Oxford Center for Evidence-Based Medicine guidelines individual studies were evaluated. A total of 59 articles were included in this study. Results: Our review showed that cardiometabolic syndrome was associated with increased risk for colorectal, hepatic, endometrial, breast, and bladder cancers. These associations showed variations for sex and geographical locations. For example, the associations were stronger for pancreatic and rectal cancers among women. The strength of these associations was also stronger for sex specific cancers such as breast and endometrial cancers. Studies on European populations showed that these associations were stronger for colorectal cancer among women. However, one study showed that presence of cardiometabolic syndrome contributed protective effects to prostate cancer among American men. In general, strongest associations were found for colorectal cancer among both men and women and hepatic cancer among men. Among cardiometabolic factors, impaired glucose tolerance and central adiposity were the greatest contributors towards increased risk for cancers. Conclusion: Given these results, there should be greater focus on primary prevention to identify and treat cardiometabolic risk factors. In addition, patients with greater cardiometabolic risk factors should be screened earlier and more frequently for cancers. A number of gender and geographical gaps identified in this review could be targeted for improvements as per the goals of 2030 sustainable development initiatives. Future studies should consider cardiometabolic syndrome and cancer together and develop effective interventions for decreasing the incidence and morbidity associated with both the conditions.

Background: In August 2019, FDA approved TAVR for low-risk patients, a paradigm shift in the field of Cardiology. This approval will pave the way for TAVR to rapidly become mainstream, and it will be a preferred procedure for patients with symptomatic aortic stenosis in the coming years. In this study, we considered creating a useful tool to assess the risk of 30-day readmission after TAVR. Methods: We utilized data from ACS National Surgical Quality Improvement Program (ACS NSQIP) between 2015 and 2017. We selected 35 past medical history (such as history of CHF, COPD, etc.) and preoperative lab (such as BUN, WBC count, etc.) variables. A binary variable for 30-day readmission was created. Elastic net, random forest, and gradient boosted classification trees machine learning models were run and best performing model was selected based on the AUROC. Variables were plotted according to the importance. We created a nomogram using logistic regression by including important variables. Methods: About 594 patients underwent TAVR between 2015 and 2017; 56% patients were 80 years or older; and 45% were females. Majority of patients (62.8%) were Non-Hispanic White, had BMI ≥ 25 (71%), and took antihypertensive medication (82%). History of CHF was reported by 26%, and COPD by 17%. There were 68 (11.4%) patients who were readmitted within 30-days for any cause. Gradient boosted classification performed best. Top 15 variables were selected for logistic regression analysis to construct a nomogram. Five of the top variables were Total hospital LOS, pre-op platelet count, pre-op serum sodium, general anesthesia, pre-op BUN. Conclusion: In the US, 1 in 6 patients with TAVR could be readmitted within 30-days due to any cause after TAVR. We presented an interpretable model based on past medical history and pre-operative variables, which can be used to identify high risk patients so that their hospital readmission risk and can be reduced. This will also lower financial burden among patients.