Search

In this paper, a novel rotation and scale invariant approach for texture classification based on Gabor filters has been proposed. These filters are designed to capture the visual content of the images based on their impulse responses which are sensitive to rotation and scaling in the images. The filter responses are rearranged according to the filter exhibiting the response having largest amplitude, followed by the calculation of patterns after binarizing the responses based on a particular threshold. This threshold is obtained as the average energy of Gabor filter responses at a particular pixel. The binary patterns are converted to decimal numbers, the histograms of which are used as texture features. The proposed features are used to classify the images from two famous texture datasets: Brodatz, CUReT and UMD texture albums. Experiments show that the proposed feature extraction method performs really well when compared with several other state-of-the-art methods considered in this paper and is more robust to noise.

A major concern related to the healthcare industry is uniformity in healthcare delivery. There is considerable variation in the diagnosis and treatment of patients depending on the experience and expertise of the doctors. Information technology can play a major role in addressing this issue. Research investigating the use of data-driven approaches and knowledge-driven clinical pathways to achieve uniformity in the delivery of healthcare is ongoing. Specifically, the integration of data and knowledge-driven approaches can be used to ensure uniformity in the delivery of patient care, thus avoiding inappropriate variance. The data-driven approach can utilize the bulk of medical data being generated. The knowledge-driven approach in the form of clinical pathways incorporates evidence-based care. In this context, knowledge and data-driven computerized physician order entry (CPOE) systems are gaining importance in healthcare delivery systems. This paper proposes a knowledge and data-driven framework for a CPOE system. This work is based on a knowledge base populated with disease quadruples, each of which comprises a list of symptoms, tests, results, and medications for a particular disease. The data used in the proposed system are obtained from two datasets, namely, the MIMIC (Medical Information Mart for Intensive Care) and the Disease-Symptom Knowledge Database, which is based on the operational data of New York-Presbyterian Hospital (NYPH). We combined both datasets using the common attribute of disease to generate data with more attributes to aid decision making. This was performed with the help of specialists and clinical knowledge. The resulting patient data are further integrated with clinical pathways before the extraction of disease quadruples. The novelty of this work lies in the extraction of disease quadruples from the integration of patient data with clinical pathways. The list dynamically ranks each element of the quadruple based on its association score to facilitate the generation of prescription order sets for the CPOE. The effectiveness of the proposed system in providing uniform patient care delivery has been validated by experts. The proposed system can significantly improve patient safety and the quality of healthcare delivery due to the integration of data-driven capability with clinical pathways.

This paper aims to develop an evolutionary deep learning based hybrid data driven approach for short term load forecasting (STLF) in the context of Bangladesh. With the lapse of time, the power system is getting complex. Penetration of intermittent renewable energy (RE) into the grid, changing prosumer load pattern with the need of demand side management (DSM) has thrown a challenge for dynamic power system operation and control. Load forecasting plays a significant role in this dynamic operation and control. In addition, it directly affects the future planning of network expansion, unit commitment and economic energy mix for power market. Day ahead short-term forecasting is very crucial for day to day operation. As such, various conventional and modified methods have been used over time for short-term prediction. Nevertheless, the existing approaches like age old statistical methods, artificial intelligence (AI), machine learning (ML), deep learning (DL) techniques alone cannot provide effective accuracy all the time. Hence, an integrated genetic algorithm (GA)-bidirectional gated recurrent unit (Bi-GRU) hybrid data driven technique (GA-BiGRU) is proposed in this work. The developed method is validated in Bangladesh power system (BPS) network by providing day ahead forecasting of electrical load of the whole country. Besides, the performance of the prediction model is compared with some existing approaches such as long short-term memory network (LSTM), gated recurrent unit (GRU) and integrated genetic algorithm-gated recurrent unit (GA-GRU) in terms of mean absolute performance error (MAPE) and root mean squared error (RMSE). The outcome gives an indication of better forecasting accuracy of proposed GA-BiGRU evolutionary DL technique compared to others.

Infrastructure-less (sometimes known as ad-hoc) networking paradigm is very appealing and potentially shaping its future into almost all emerging networks (i.e., IoT, wireless sensor networks, vehicular ad-hoc networks, emergency, and tactical radio networks, etc.). However, when conventional networking protocols are used, such networks often perform poorly, mainly because of interference within the network and limited network throughput. Recent advancements in wireless communications have enabled full-duplex (FD) operation by suppressing self-interference, which can theoretically double the network throughput. However, conventional medium access control (MAC) protocols like carrier-sense multiple access with collision avoidance (CSMA/CA) favor half-duplex (HD) operation and fail to benefit from FD transmission opportunities. This article proposes a novel hybrid MAC protocol for full-duplex ad-hoc networks. The proposed MAC combines time division multiple access (TDMA) and IEEE 802.11 distributed coordination function (DCF) strengths in chains of time-slotted contention-based control frames and collision-free data frames. The aim is to fully utilize FD transmission opportunities to increase network throughput. The proposed protocol modifies request-to-send (RTS) and clear-to-send (CTS) frames in IEEE 802.11 DCF MAC to form FD-RTS/CTS frames. These frames are used to enable collision-free FD communications among neighbors. The proposed scheme mitigates conventional MAC issues like hidden-node problem (HNP) and exposed-node problem (ENP). It also allows concurrent FD data transmissions in a collision-free manner. The model is generic and can be applied to any ad-hoc wireless network. In this article, the design is applied to a single channel ad-hoc network with FD transceivers. We compared the proposed design with IEEE 802.11 CSMA/CA protocol with both HD and FD transceivers. The simulation results show a 30% gain in throughput, reduced latency, and fairness among participating wireless nodes.

A massive amount of medical data is available in healthcare industry, which can be utilized to extract useful knowledge. A Clinical Decision Support System (CDSS) is used to improve patient’s safety by minimizing medical errors. Heart disease is one of the major chronic maladies even in todays’ world. Many researchers have employed different data mining techniques to predict heart disease. The objective of proposed framework is to improve the accuracy of heart disease prediction. In this paper, an ensemble based voting scheme is proposed to efficiently predict heart disease. Four benchmark heart disease datasets from UCI repository have been utilized for experimentation and evaluation. The performance of the proposed ensemble is compared with individual classifiers as well as with five different ensemble schemes using various parameters in order to show the effectiveness of the proposed ensemble scheme. The evaluation of results shows that the proposed ensemble scheme has better average accuracy (83%) as compared to other ensemble schemes as well as individual classifiers.