Cardiovascular diseases (CVDs) are a significant cause of death worldwide. CVDs can be prevented by diagnosing heartbeat sounds and other conventional techniques early to reduce the harmful effects caused by CVDs. However, it is still challenging to segment, extract features, and predict heartbeat sounds in elderly people. The inception of deep learning (DL) algorithms has helped detect various types of heartbeat sounds at an early stage. Motivated by this, we proposed an intelligent architecture categorizing heartbeat into normal and murmurs for elderly people. We have used a standard heartbeat dataset with heartbeat class labels, i.e., normal and murmur. Furthermore, it is augmented and preprocessed by normalization and standardization to significantly reduce computational power and time. The proposed convolutional neural network and bi-directional gated recurrent unit (CNN + BiGRU) attention-based architecture for the classification of heartbeat sound achieves an accuracy of 90% compared to the baseline approaches. Hence, the proposed novel CNN + BiGRU attention-based architecture is superior to other DL models for heartbeat sound classification.

One of the fascinating subjects in the area of Music Information Retrieval (MIR) is the classification of music as it is played into different genres. Machine learning is used in this analysis. To predict the genre of the audio signal, models such as Support Vector Machines (SVM), Random Forests, XGB (eXtreme Gradient Boosting), and Convolutional Neural Networks (CNN) are used. The GTZAN dataset was used for model training and testing. Machine learning and deep learning models each had their own set of features. A comparison analysis is proposed between these models, demonstrating that CNN outperforms machine learning models.

Hyperspectral imaging which is also known as imaging spectroscopy, detects radiation of earth surface features in narrow contiguous spectral regions of the electromagnetic spectrum. The Airborne Visible Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) is an airborne hyperspectral sensor of NASA’s Jet Propulsion Laboratory (JPL) with 425 spectral bands ranging from 380 nm to 2510 nm with a bandwidth of 5 nm and spatial resolution of 4-6 m. This study aims at pixel-wise identification and discrimination of crop types using AVIRIS-NG hyperspectral images, with novel Parallel Convolutional Neural Networks architecture. To tackle the challenge posed by a large number of correlated bands, we compare two band selection techniques using Principal Component Analysis (PCA) and back traversal of pre-trained Artificial Neural Network (ANN). We also propose an automated technique for augmentation of training dataset with a large number of pixels from unlabelled parts of an image, based on Euclidian distance. Experiments show that bands selected by ANN achieve higher accuracy compare to PCA selected bands with automated data augmentation.