A Simple Federated Learning-Based Scheme for Security Enhancement Over Internet of Medical Things.

Nowadays, Federated Learning (FL) over Internet of Medical Things (IoMT) devices has become a current research hotspot. As a new architecture, FL can well protect the data privacy of IoMT devices, but the security of neural network model transmission can not be guaranteed. On the other hand, the sizes of current popular neural network models are usually relatively extensive, and how to deploy them on the IoMT devices has become a challenge. One promising approach to these problems is to reduce the network scale by quantizing the parameters of the neural networks, which can greatly improve the security of data transmission and reduce the transmission cost. In the previous literature, the fixed-point quantizer with stochastic rounding has been shown to have better performance than other quantization methods. However, how to design such quantizer to achieve the minimum square quantization error is still unknown. In addition, how to apply this quantizer in the FL framework also needs investigation. To address these questions, in this paper, we propose FedMSQE - Federated Learning with Minimum Square Quantization Error, that achieves the smallest quantization error for each individual client in the FL setting. Through numerical experiments in both single-node and FL scenarios, we prove that our proposed algorithm can achieve higher accuracy and lower quantization error than other quantization methods.

Conditional Anonymous Remote Healthcare Data Sharing Over Blockchain.

As an important carrier of healthcare data, Electronic Medical Records (EMRs) generated from various sensors, i.e., wearable, implantable, are extremely valuable research materials for artificial intelligence and machine learning. The efficient circulation of EMRs can improve remote medical services and promote the development of the related healthcare industry. However, in traditional centralized data sharing architectures, the balance between privacy and traceability still cannot be well handled. To address the issue that malicious users cannot be locked in the fully anonymous sharing schemes, we propose a trackable anonymous remote healthcare data storing and sharing scheme over decentralized consortium blockchain. Through an "on-chain & off-chain" model, it relieves the massive data storage pressure of medical blockchain. By introducing an improved proxy re-encryption mechanism, the proposed scheme realizes the fine-gained access control of the outsourced data, and can also prevent the collusion between semi-trusted cloud servers and data requestors who try to reveal EMRs without authorization. Compared with the existing schemes, our solution can provide a lower computational overhead in repeated EMRs sharing, resulting in a more efficient overall performance.

Deep Data Analysis-Based Agricultural Products Management for Smart Public Healthcare.

Agricultural is an indispensably public healthcare industry for human beings at any time and smart management of it is of great significance. Since substantial technical advance relies on long-term efforts and continuous progress, reasonably scheduling the distribution of agricultural products acts as a key aspect of smart public healthcare. The most intuitive factor affecting the distribution of agricultural products is its dynamic price. Forecasting price fluctuations in advance can optimize the distribution of agricultural products and pave the way to smart public healthcare. Most researchers study the prices of various agricultural products separately, without considering the interaction of different agricultural products in the time dimension. This study introduces a typical deep learning model named graph neural network (GNN) for this purpose and proposes deep data analysis-based agricultural products management for smart public healthcare (named GNN-APM for short). The highlight of GNN-APM is to take latent correlations among multiple types of agricultural products into consideration when modeling evolving rules of price sequences. A case study is set up with the use of real-world data of the agricultural products market. Simulative results reveal that the designed GNN-APM functions well.

CNN-LSTM Hybrid Real-Time IoT-Based Cognitive Approaches for ISLR with WebRTC: Auditory Impaired Assistive Technology.

In the era of modern technology, people may readily communicate through facial expressions, body language, and other means. As the use of the Internet evolves, it may be a boon to the medical fields. Recently, the Internet of Medical Things (IoMT) has provided a broader platform to handle difficulties linked to healthcare, including people's listening and hearing impairment. Although there are many translators that exist to help people of various linguistic backgrounds communicate more effectively. Using kinesics linguistics, one may assess or comprehend the communications of auditory and hearing-impaired persons who are standing next to each other. When looking at the present COVID-19 scenario, individuals are still linked in some way via online platforms; however, persons with disabilities have communication challenges with online platforms. The work provided in this research serves as a communication bridge inside the challenged community and the rest of the globe. The proposed work for Indian Sign Linguistic Recognition (ISLR) uses three-dimensional convolutional neural networks (3D-CNNs) and long short-term memory (LSTM) technique for analysis. A conventional hand gesture recognition system involves identifying the hand and its location or orientation, extracting certain essential features and applying an appropriate machine learning algorithm to recognise the completed action. In the calling interface of the web application, WebRTC has been implemented. A teleprompting technology is also used in the web app, which transforms sign language into audible sound. The proposed web app's average recognition rate is 97.21%.

Clouds Proportionate Medical Data Stream Analytics for Internet of Things-Based Healthcare Systems.

Internet of Things (IoT) assisted healthcare systems are designed for providing ubiquitous access and recommendations for personal and distributed electronic health services. The heterogeneous IoT platform assists healthcare services with reliable data management through dedicated computing devices. Healthcare services' reliability depends upon the efficient handling of heterogeneous data streams due to variations and errors. A Proportionate Data Analytics (PDA) for heterogeneous healthcare data stream processing is introduced in this manuscript. This analytics method differentiates the data streams based on variations and errors for satisfying the service responses. The classification is streamlined using linear regression for segregating errors from the variations in different time intervals. The time intervals are differentiated recurrently after detecting errors in the stream's variation. This process of differentiation and classification retains a high response ratio for healthcare services through spontaneous regressions. The proposed method's performance is analyzed using the metrics accuracy, identification ratio, delivery, variation factor, and processing time.

Gene Mutation Classification through Text Evidence Facilitating Cancer Tumour Detection.

A cancer tumour consists of thousands of genetic mutations. Even after advancement in technology, the task of distinguishing genetic mutations, which act as driver for the growth of tumour with passengers (Neutral Genetic Mutations), is still being done manually. This is a time-consuming process where pathologists interpret every genetic mutation from the clinical evidence manually. These clinical shreds of evidence belong to a total of nine classes, but the criterion of classification is still unknown. The main aim of this research is to propose a multiclass classifier to classify the genetic mutations based on clinical evidence (i.e., the text description of these genetic mutations) using Natural Language Processing (NLP) techniques. The dataset for this research is taken from Kaggle and is provided by the Memorial Sloan Kettering Cancer Center (MSKCC). The world-class researchers and oncologists contribute the dataset. Three text transformation models, namely, CountVectorizer, TfidfVectorizer, and Word2Vec, are utilized for the conversion of text to a matrix of token counts. Three machine learning classification models, namely, Logistic Regression (LR), Random Forest (RF), and XGBoost (XGB), along with the Recurrent Neural Network (RNN) model of deep learning, are applied to the sparse matrix (keywords count representation) of text descriptions. The accuracy score of all the proposed classifiers is evaluated by using the confusion matrix. Finally, the empirical results show that the RNN model of deep learning has performed better than other proposed classifiers with the highest accuracy of 70%.

Multiconstraint-Aware Routing Mechanism for Wireless Body Sensor Networks.

The merger of wireless sensor technologies, pervasive computing, and biomedical engineering has resulted in the emergence of wireless body sensor network (WBSN). WBSNs assist human beings in various monitoring applications such as health-care, entertainment, rehabilitation systems, and sports. Life-critical health-care applications of WBSNs consider both reliability and delay as major Quality of Service (QoS) parameters. In addition to the common limitations and challenges of wireless sensor networks (WSNs), WBSNs pose distinct constraints due to the behavior and chemistry of the human body. The biomedical sensor nodes (BMSNs) adopt multihop communication while reporting the heterogeneous natured physiological parameters to the nearby base station also called local coordinator. Routing in WBSNs becomes a challenging job due to the necessary QoS considerations, overheated in-body BMSNs, and high and dynamic path loss. To the best of our knowledge, none of the existing routing protocols integrate the aforementioned issues in their designs. In this research work, a multiconstraint-aware routing mechanism (modular-based) is proposed which considers the QoS parameters, dynamic and high path loss, and the overheated nodes issue. Two types of network frameworks, with and without relay/forwarder nodes, are being used. The data packets containing physiological parameters of the human body are categorized into delay-constrained, reliability-constrained, critical (both delay- and reliability-constrained), and nonconstrained data packets. NS-2 is being used to carry out the simulations of the proposed mechanism. The simulation results reveal that the proposed mechanism has improved the QoS-aware routing for WBSNs by adopting the proposed multiconstraint-aware strategy.