Evolutionary tree-based quasi identifier and federated gradient privacy preservations over big healthcare data

<span>Big data has remodeled the way organizations supervise, examine and leverage data in any industry. To safeguard sensitive data from public contraventions, several countries investigated this issue and carried out privacy protection mechanism. With the aid of quasi-identifiers privacy is not said to be preserved to a greater extent. This paper proposes a method called evolutionary tree-based quasi-identifier and federated gradient (ETQI-FD) for privacy preservations over big healthcare data. The first step involved in the ETQI-FD is learning quasi-identifiers. Learning quasi-identifiers by employing information loss function separately for categorical and numerical attributes accomplishes both the largest dissimilarities and partition without a comprehensive exploration between tuples of features or attributes. Next with the learnt quasi-identifiers, privacy preservation of data item is made by applying federated gradient arbitrary privacy preservation learning model. This model attains optimal balance between privacy and accuracy. In the federated gradient privacy preservation learning model, we evaluate the determinant of each attribute to the outputs. Then injecting Adaptive Lorentz noise to data attributes our ETQI-FD significantly minimizes the influence of noise on the final results and therefore contributing to privacy and accuracy. An experimental evaluation of ETQI-FD method achieves better accuracy and privacy than the existing methods.</span>

Navigating the hydroxymethylome: experimental biases and quality control tools for the tandem bisulfite and oxidative bisulfite Illumina microarrays

Aim: Tandem bisulfite (BS) and oxidative bisulfite (oxBS) conversion on DNA followed by hybridization to Infinium HumanMethylation BeadChips allows nucleotide resolution of 5-hydroxymethylcytosine genome-wide. Here, the authors compared data quality acquired from BS-treated and oxBS-treated samples. Materials & methods: Raw BeadArray data from 417 pairs of samples across 12 independent datasets were included in the study. Probe call rates were compared between paired BS and oxBS treatments controlling for technical variables. Results: oxBS-treated samples had a significantly lower call-rate. Among technical variables, DNA-specific extraction kits performed better with higher call rates after oxBS conversion. Conclusion: The authors emphasize the importance of quality control during oxBS conversion to minimize information loss and recommend using a DNA-specific extraction kit for DNA extraction and an oxBSQC package for data preprocessing.

Rethinking Pooling Operation for Liver and Liver-Tumor Segmentations

Deep convolutional neural networks (DCNNs) have been widely used in medical image segmentation due to their excellent feature learning ability. In these DCNNs, the pooling operation is usually used for image down-sampling, which can gradually reduce the image resolution and thus expands the receptive field of convolution kernel. Although the pooling operation has the above advantages, it inevitably causes information loss during the down-sampling of the pooling process. This paper proposes an effective weighted pooling operation to address the problem of information loss. First, we set up a pooling window with learnable parameters, and then update these parameters during the training process. Secondly, we use weighted pooling to improve the full-scale skip connection and enhance the multi-scale feature fusion. We evaluated weighted pooling on two public benchmark datasets, the LiTS2017 and the CHAOS. The experimental results show that the proposed weighted pooling operation effectively improve network performance and improve the accuracy of liver and liver-tumor segmentation.

Performance Evaluation of Solar Energy Cells Using the Interval-Valued T-Spherical Fuzzy Bonferroni Mean Operators

To find the correspondence between every number of attributes, the Bonferroni mean (BM) operator is most widely used and proven to be a flexible approach. To express uncertain information, the frame of the interval-valued T-spherical fuzzy set (IVTSFS) is a recent development in fuzzy settings which discusses four aspects of uncertain information using closed sub-intervals of 0, 1 and hence reduces the information loss greatly. In this research study, we introduced the principle of BM operators with IVTSFS to develop the principle of the inter-valued T-spherical fuzzy (IVTSF) BM (IVTSFBM) operator, the IVTSF-weighted BM (IVTSFWBM) operator, the IVTSF geometric BM (IVTSFGBM) operator, and the IVTSF-weighted geometric BM (IVTSFWGBM) operator. To see the significance of the proposed BM operators, we applied these BM operators to evaluate the performance of solar cells that play an important role in the field of energy storage. To do so, we developed a multi-attribute group decision-making (MAGDM) procedure based on IVTSF information and applied it to the problem of solar cells to evaluate their performance under uncertainty, where four aspects of opinion about solar cells were taken into consideration. We studied the results obtained using BM operators with some previous operators to see the significance of the proposed IVTSF BM operators.

Pricing and quality decisions in a supply chain with consumers' privacy concern

<p style='text-indent:20px;'>This study considers a supply chain consists of one manufacturer produces a product with a quality level and sells it through one retailer. A stylized model is developed to investigate the impacts of consumers' privacy concerns on pricing, quality decisions, and profitability through the relationship between product quality and personal information. When consumers' privacy concern is considered, the product quality level, the wholesale price, the payoffs of the manufacturer and retailer, and consumer surplus decrease with the personal information loss, whereas the selling price increases if this loss is low. Our results also show that the retailer prefers to charge a high selling price if the information benefit and the personal information loss are low, or the information benefit is relatively high. Moreover, a "win-win-win" outcome can be achieved among the manufacturer, retailer, and consumers if the personal information loss is sufficiently low. In the case of quality-differentiated products, however, although the manufacturer improves the product quality level, the wholesale prices are increased if the information benefit and the personal information loss are low, or the information benefit is high.</p>

Exponential Ant-Lion Rider Optimization for Privacy Preservation in Cloud Computing

The innovative trend of cloud computing is outsourcing data to the cloud servers by individuals or enterprises. Recently, various techniques are devised for facilitating privacy protection on untrusted cloud platforms. However, the classical privacy-preserving techniques failed to prevent leakage and cause huge information loss. This paper devises a novel methodology, namely the Exponential-Ant-lion Rider optimization algorithm based bilinear map coefficient Generation (Exponential-AROA based BMCG) method for privacy preservation in cloud infrastructure. The proposed Exponential-AROA is devised by integrating Exponential weighted moving average (EWMA), Ant Lion optimizer (ALO), and Rider optimization algorithm (ROA). The input data is fed to the privacy preservation process wherein the data matrix, and bilinear map coefficient Generation (BMCG) coefficient are multiplied through Hilbert space-based tensor product. Here, the bilinear map coefficient is obtained by multiplying the original data matrix and with modified elliptical curve cryptography (MECC) encryption to maintain data security. The bilinear map coefficient is used to handle both the utility and the sensitive information. Hence, an optimization-driven algorithm is utilized to evaluate the optimal bilinear map coefficient. Here, the fitness function is newly devised considering privacy and utility. The proposed Exponential-AROA based BMCG provided superior performance with maximal accuracy of 94.024%, maximal fitness of 1, and minimal Information loss of 5.977%.

Capturing a Comprehensive Picture of Biological Events From Adverse Outcome Pathways in the Drug Exposome

Background: The chemical part of the exposome, including drugs, may explain the increase of health effects with outcomes such as infertility, allergies, metabolic disorders, which cannot be only explained by the genetic changes. To better understand how drug exposure can impact human health, the concepts of adverse outcome pathways (AOPs) and AOP networks (AONs), which are representations of causally linked events at different biological levels leading to adverse health, could be used for drug safety assessment.Methods: To explore the action of drugs across multiple scales of the biological organization, we investigated the use of a network-based approach in the known AOP space. Considering the drugs and their associations to biological events, such as molecular initiating event and key event, a bipartite network was developed. This bipartite network was projected into a monopartite network capturing the event–event linkages. Nevertheless, such transformation of a bipartite network to a monopartite network had a huge risk of information loss. A way to solve this problem is to quantify the network reduction. We calculated two scoring systems, one measuring the uncertainty and a second one describing the loss of coverage on the developed event–event network to better investigate events from AOPs linked to drugs.Results: This AON analysis allowed us to identify biological events that are highly connected to drugs, such as events involving nuclear receptors (ER, AR, and PXR/SXR). Furthermore, we observed that the number of events involved in a linkage pattern with drugs is a key factor that influences information loss during monopartite network projection. Such scores have the potential to quantify the uncertainty of an event involved in an AON, and could be valuable for the weight of evidence assessment of AOPs. A case study related to infertility, more specifically to “decrease, male agenital distance” is presented.Conclusion: This study highlights that computational approaches based on network science may help to understand the complexity of drug health effects, with the aim to support drug safety assessment.

Automated Drilling Data Quality Control Using Application of AI Technologies

Drilling data quality is notoriously a challenge for any analytics application, due to complexity of the real-time data acquisition system which routinely generates: (i) Time related issues caused by irregular sampling, (ii) Channel related issues in terms of non-uniform names and units, missing or wrong values, and (iii) Depth related issues caused block position resets, and depth compensation (for floating rigs). On the other hand, artificial intelligence drilling applications typically require a consistent stream of high-quality data as an input for their algorithms, as well as for visualization. In this work we present an automated workflow enhanced by data driven techniques that resolves complex quality issues, harmonize sensor drilling data, and report the quality of the dataset to be used for advanced analytics. The approach proposes an automated data quality workflow which formalizes the characteristics, requirements and constraints of sensor data within the context of drilling operations. The workflow leverages machine learning algorithms, statistics, signal processing and rule-based engines for detection of data quality issues including error values, outliers, bias, drifts, noise, and missing values. Further, once data quality issues are classified, they are scored and treated on a context specific basis in order to recover the maximum volume of data while avoiding information loss. This results into a data quality and preparation engine that organizes drilling data for further advanced analytics, and reports the quality of the dataset through key performance indicators. This novel data processing workflow allowed to recover more than 90% of a drilling dataset made of 18 offshore wells, that otherwise could not be used for analytics. This was achieved by resolving specific issues including, resampling timeseries with gaps and different sampling rates, smart imputation of wrong/missing data while preserving consistency of dataset across all channels. Additional improvement would include recovering data values that felt outside a meaningful range because of sensor drifting or depth resets. The present work automates the end-to-end workflow for data quality control of drilling sensor data leveraging advanced Artificial Intelligence (AI) algorithms. It allows to detect and classify patterns of wrong/missing data, and to recover them through a context driven approach that prevents information loss. As a result, the maximum amount of data is recovered for artificial intelligence drilling applications. The workflow also enables optimal time synchronization of different sensors streaming data at different frequencies, within discontinuous time intervals.