Data Mining Practical Machine Learning Tools and Techniques

This paper presents control and coordination algorithms for groups of vehicles. The focus is on autonomous vehicle networks performing distributed sensing tasks where each vehicle plays the role of a mobile tunable sensor. The paper proposes gradient descent algorithms for a class of utility functions which encode optimal coverage and sensing policies. The resulting closed-loop behavior is adaptive, distributed, asynchronous, and verifiably correct.

Coverage control for mobile sensing networks

‡We describe the use of singular value decomposition in transforming genome-wide expression data from genes 3 arrays space to reduced diagonalized ‘‘eigengenes’’ 3 ‘‘eigenarrays’’ space, where the eigengenes (or eigenarrays) are unique orthonormal superpositions of the genes (or arrays). Normalizing the data by filtering out the eigengenes (and eigenarrays) that are inferred to represent noise or experimental artifacts enables meaningful comparison of the expression of different genes across different arrays in different experiments. Sorting the data according to the eigengenes and eigenarrays gives a global picture of the dynamics of gene expression, in which individual genes and arrays appear to be classified into groups of similar regulation and function, or similar cellular state and biological phenotype, respectively. After normalization and sorting, the significant eigengenes and eigenarrays can be associated with observed genome-wide effects of regulators, or with measured samples, in which these regulators are overactive or underactive, respectively.

Singular Value Decomposition for Genome-Wide Expression Data Processing and Modeling

https://www.journal-of-hepatology.eu/article/S0168-8278(15)00259-7/pdf

EASL-ALEH Clinical Practice Guidelines: Non-invasive tests for evaluation of liver disease severity and prognosis.

The latest advances in hepatology were presented in oral and poster presentations. In order to cover the varying subspecialties, the sessions were divided into various sections including 'Acute Liver Failure and Artificial Liver Support', 'Biliary Tract and Immunologic Liver Diseases', 'Cellular and Molecular Biology', 'Clinical and Experimental Hepatobiliary Surgery', 'Hepatotoxicity and Cell Death', 'Transport and Biliary Physiology', 'Viral Hepatitis', 'Evaluation and Treatment of Biliary Disease', 'Necrosis/Apoptosis', 'Portal Hypertension', 'Blood Flow and Vascular Disorders of Cirrhosis', 'Liver Transplantation', 'Fibrogenesis', 'Hepatocellular Carcinoma', 'Metabolism and Genetic Disease', and 'Public Policy, Epidemiology and Decision Analysis'. Drug therapy focused on treatments for viral hepatitis, autoimmune hepatitis, primary biliary cirrhosis and primary sclerosing cholangitis, and recurrent viral disease following liver transplant. High dose interferon therapy or various combinations of interferon/ribavirin (ICN Pharmaceuticals Inc) therapy seem to offer the best current therapy for chronic HCV. PEGylated interferon (F Hoffmann-La Roche Ltd) offers hope for treatment and histologic improvement in patients with chronic HCV. Following liver transplantation, combination interferon/ribavirin therapy may also find success, but caution with new potent immunosuppressant monoclonal antibodies is advised. For HBV, intramuscular H-BIG (NABI) appears to be effective and less costly than iv H-BIG administration following liver transplantation. Percutaneous radiofrequency ablation may hold promise over conventional ethanol injection therapy for small hepatocellular carcinoma. Autoimmune hepatitis may respond to tacrolimus therapy whereas budesonide therapy did not provide any advantage to prednisone therapy. For primary biliary cirrhosis, eicosapentate and ursodeoxycholic acid may provide benefit to some patients while silymarin from milk thistle did not provide any additional benefit. In primary sclerosing cholangitis, high dose ursodeoxycholic acid may provide benefit. Ursodeoxycholic acid may also provide benefit for mothers with intrahepatic cholestasis of pregnancy by decreasing pruritus, lowering laboratory values and allowing deliveries to occur closer to term.

Digestive disease week 2000. American Association for the Study of Liver Diseases

Background and aim Mortality of alcohol-related liver disease (ArLD) is increasing, and liver fibrosis stage is the best mortality predictor. Non-invasive tests (NITs) are increasingly used to detect fibrosis, but their value as prognostic tests in chronic liver disease, and in particular in ArLD, is less well recognized. We aimed to describe the prognostic performance of four widely used NITs (Fibrosis 4 test [FIB4], Enhanced Liver Fibrosis [ELF] test, FibroScan, and FibroTest) in ArLD. Methods Applying systematic review methodology, we searched four databases from inception to May 2020. Inclusion/exclusion criteria were applied to search using Medical Subject Heading terms and keywords. The first and second reviewers independently screened results, extracted data, and performed risk-of-bias assessment using Quality in Prognosis Studies tool. Results Searches produced 25 088 articles. After initial screening, 1020 articles were reviewed independently by both reviewers. Eleven articles remained after screening for eligibility: one on ELF, four on FibroScan, four on FIB4, one on FIB4 + FibroScan, and one on FibroTest + FIB4. Area under the receiver operating characteristic curves for outcome prediction ranged from 0.65 to 0.76 for FibroScan, 0.64 to 0.83 for FIB4, 0.69 to 0.79 for FibroTest, and 0.72 to 0.85 for ELF. Studies scored low-moderate risk of bias for most domains but high risk in confounding/statistical reporting domains. The results were heterogeneous for outcomes and reporting, making pooling of data unfeasible. Conclusions This systematic review returned 11 papers, six of which were conference abstracts and one unpublished manuscript. While the heterogeneity of studies precluded direct comparisons of NITs, each NIT performed well in individual studies in predicting prognosis in ArLD (area under the receiver operating characteristic curves >0.7 in each NIT category) and may add value to prognostication in clinical practice.

Systematic review: Investigating the prognostic performance of four non‐invasive tests in alcohol‐related liver disease

Now days Wireless Sensor Networks (WSNs) are very popular due to their uses in potential range of applications. A set of either random or predefined distributed tiny sensor nodes could perform the desired task as per need. A huge list of different parameters is available in the litrature that must required special attention while designing the WSN system. Energy saving of each node is one of the key issue as nodes are battery operated and replacement of battery is almost impossible. Network coverage partially or completely is also required to meet out the quality of sensing information. To address these issues, we propose a new Hybrid Hterrogeneous LEACH (HHE-LEACH) protocol for sensor nodes in WSNs. We have used the concepts of learning automata (LA) for the WSN in which every node has LA. Direct or indirect communication with base station (BS) will depends upon the distance-based thresholds which makes a partition of WSN. During this process, each LA may get a reward or penalty with respect to finite number of actions performed. Performance of the HHE-LEACH is evaluated using extensive simulations by selecting various metrics. The results obtained show that proposed scheme achieved better stability, network lifetime and partial network coverage in comparison to other existing protocols such as DD, LEACH and SEP.

Learning Automata-Based Coverage Oriented Clustering in HWSNs

This article proposes a SaTya scheme that leverages a blockchain (BC)-based deep learning (DL)-assisted classifier model that forms a trusted chronology in fake news classification. The news collected from newspapers, social handles, and e-mails are web-scrapped, prepossessed, and sent to a proposed Q-global vector for word representations (Q-GloVe) model that captures the fine-grained linguistic semantics in the data. Based on the Q-GloVe output, the data are trained through a proposed bi-directional long short-term memory (Bi-LSTM) model, and the news is classified as real-or-fake news. This reduces the vanishing gradient problem, which optimizes the weights of the model and reduces bias. Once the news is classified, it is stored as a transaction, and the news stakeholders can execute smart contracts (SCs) and trace the news origin. However, only verified trusted news sources are added to the BC network, ensuring credibility in the system. For security evaluation, we propose the associated cost of the Bi-LSTM classifier and propose vulnerability analysis through the smart check tool for potential vulnerabilities. The scheme is compared against discourse-structure analysis, linguistic natural language framework, and entity-based recognition for different performance metrics. The scheme achieves an accuracy of 99.55% compared to 93.62% against discourse structure analysis. Also, it shows an average improvement of 18.76% against other approaches, which indicates its viability against fake-classifier-based models.

SaTYa: Trusted Bi-LSTM-Based Fake News Classification Scheme for Smart Community

The smart healthcare system has improved the patients quality of life (QoL), where the records are being analyzed remotely by distributed stakeholders. It requires a voluminous exchange of data for disease prediction via the open communication channel, i.e., the Internet to train artificial intelligence (AI) models efficiently and effectively. The open nature of communication channels puts data privacy at high risk and affects the model training of collected data at centralized servers. To overcome this, an emerging concept, i.e., federated learning (FL) is a viable solution. It performs training at client nodes and aggregates their results to train the global model. The concept of local training preserves the privacy, confidentiality, and integrity of the patient’s data which contributes effectively to the training process. The applicability of FL in the healthcare domain has various advantages, but it has not been explored to its extent. The existing surveys majorly focused on the role of FL in diverse applications, but there exists no detailed or comprehensive survey on FL in healthcare informatics (HI). We present a relative comparison of recent surveys with the proposed survey. To strengthen healthcare data privacy and increase the QoL of patients, we proposed an FL-based layered healthcare informatics architecture along with the case study on FL-based electronic health records (FL-EHR). We discuss the emerging FL models, and present the statistical and security challenges in FL adoption in medical setups. Thus, the review presents useful insights for both academia and healthcare practitioners to investigate FL application in HI ecosystems.

https://ieeexplore.ieee.org/ielx7/6287639/6514899/09867987.pdf

Adoption of Federated Learning for Healthcare Informatics: Emerging Applications and Future Directions

Regular up-gradation in the standards of the wireless networks and to sustain in the market with the best services, are the key challenges to the services providers of the telecommunication industries Traditional mobile number portability system (TMNPS) is based on the manual process, in which existing service provider and the new service provider have to undergo certain transfers once a porting request is generated by the user Slow processing of request, lack of transparency among the entities involved, and delay during the verification are the major challenges in the TMNPS Moreover, a third-party is required to ensure the security of the stored and regulated data Motivated from these facts, in this paper, we propose a blockchain-based mobile number portability scheme, which uses an Ethereum blockchain technology and call routing mechanism to handle the request of the user The secured royalty contract transactions and secure call routing mechanism are the potential part of the proposed scheme We then evaluated the performance of the proposed scheme and functionalities of the proposed smart contracts (SCs) using experimental setup with respect to various parameters Results show that the transition time taken by the proposed scheme is only 4 seconds for 200 blocks and TMNPS took 6 seconds to execute the 200 blocks

Sudeep Tanwar

Papers

Systematic review: Investigating the prognostic performance of four non‐invasive tests in alcohol‐related liver disease

Learning Automata-Based Coverage Oriented Clustering in HWSNs

SaTYa: Trusted Bi-LSTM-Based Fake News Classification Scheme for Smart Community

Adoption of Federated Learning for Healthcare Informatics: Emerging Applications and Future Directions

Blockchain-based scheme for the mobile number portability