Auscultation-Based Pulmonary Disease Detection through Parallel Transformation and Deep Learning.

Respiratory diseases are among the leading causes of death, with many individuals in a population frequently affected by various types of pulmonary disorders. Early diagnosis and patient monitoring (traditionally involving lung auscultation) are essential for the effective management of respiratory diseases. However, the interpretation of lung sounds is a subjective and labor-intensive process that demands considerable medical expertise, and there is a good chance of misclassification. To address this problem, we propose a hybrid deep learning technique that incorporates signal processing techniques. Parallel transformation is applied to adventitious respiratory sounds, transforming lung sound signals into two distinct time-frequency scalograms: the continuous wavelet transform and the mel spectrogram. Furthermore, parallel convolutional autoencoders are employed to extract features from scalograms, and the resulting latent space features are fused into a hybrid feature pool. Finally, leveraging a long short-term memory model, a feature from the latent space is used as input for classifying various types of respiratory diseases. Our work is evaluated using the ICBHI-2017 lung sound dataset. The experimental findings indicate that our proposed method achieves promising predictive performance, with average values for accuracy, sensitivity, specificity, and F1-score of 94.16%, 89.56%, 99.10%, and 89.56%, respectively, for eight-class respiratory diseases; 79.61%, 78.55%, 92.49%, and 78.67%, respectively, for four-class diseases; and 85.61%, 83.44%, 83.44%, and 84.21%, respectively, for binary-class (normal vs. abnormal) lung sounds.

Multichannel-Sensing Scheduling and Transmission-Energy Optimizing in Cognitive Radio Networks with Energy Harvesting.

This paper considers cognitive radio networks (CRNs) utilizing multiple time-slotted primary channels in which cognitive users (CUs) are powered by energy harvesters. The CUs are under the consideration that hardware constraints on radio devices only allow them to sense and transmit on one channel at a time. For a scenario where the arrival of harvested energy packets and the battery capacity are finite, we propose a scheme to optimize (i) the channel-sensing schedule (consisting of finding the optimal action (silent or active) and sensing order of channels) and (ii) the optimal transmission energy set corresponding to the channels in the sensing order for the operation of the CU in order to maximize the expected throughput of the CRN over multiple time slots. Frequency-switching delay, energy-switching cost, correlation in spectrum occupancy across time and frequency and errors in spectrum sensing are also considered in this work. The performance of the proposed scheme is evaluated via simulation. The simulation results show that the throughput of the proposed scheme is greatly improved, in comparison to related schemes in the literature. The collision ratio on the primary channels is also investigated.

Development of a cause analysis system for a CPCS trip by using the rule-base deduction method.

A Core Protection Calculator System (CPCS) was developed to initiate a Reactor Trip under the circumstance of certain transients by a Combustion Engineering Company. The major function of the Core Protection Calculator System is to generate contact outputs for the Departure from Nucleate Boiling Ratio (DNBR) Trip and a Local Power Density (LPD) Trip. But in a Core Protection Calculator System, a trip cause cannot be identified, thus only trip signals are transferred to the Plant Protection System (PPS) and only the trip status is displayed. It could take a considerable amount of time and effort for a plant operator to analyze the trip causes of a Core Protection Calculator System. So, a Cause Analysis System for a Core Protection Calculator System (CASCPCS) has been developed by using the rule-base deduction method to assist operators in a Nuclear Power Plant. CASCPCS consists of three major parts. Inference engine has a role of controlling the searching knowledge base, executing the rules and tracking the inference process by using the depth-first searching method. Knowledge base consists of four major parts: rules, data base constants, trip buffer variables and causes. And a user interface is implemented by using menu-driven and window display techniques. The advantage of CASCPCS is that it saves time and effort to diagnose the trip causes of a Core Protection Calculator System, it increases a plant's availability and reliability, and it makes it easy to manage CASCPCS because of using only a cursor control.