Visually impaired people (VIPs) encounter various challenges in their daily lives, and there is a need for portable, user-friendly device for real-time assistance to give them guidance regarding their surroundings. This article presents an artificial-intelligence-based innovative wearable assistive device—artificial intelligence (AI)-SenseVision—to analyze visual and sensory information about the objects and obstacles present in the scene to perceive the surrounding environment. The device is a complete amalgamation of sensor and computer-vision-based technologies that generate auditory information with the name of identified objects or audio warnings for detected obstacles. The performance of the trained deep-learning model is rigorously tested in complex and real-life scenarios using various statistical parameters for experimental validation. Moreover, the trained deep-learning models have been integrated into a low-cost single-board processor to make a standalone cost-effective device. All data processing is done within an optimized single hardware setup, and the user can easily access different modes, such as indoor and outdoor mode, while also enabling object counting in observed scenes. The proposed system has low-cost sensors, multiple operational modes, easy integration, and small volume, making this assistive device helpful for VIPs for independent navigation and collision prevention.

Abstract Blood‐related diseases are one of the major concerns in the biomedical domain and most of the disease symptoms are reflected through the analysis of blood cells. The diagnosis by experts in laboratories is very costly and time‐consuming, thus artificial intelligence‐based systems can help in the automatic diagnosis and monitoring of an individual's health. In this study, a CNN‐based architecture Microcell‐Net is proposed which is trained on a microscopic image dataset of peripheral blood cells in eight different classes. The images have several inter‐class and intra‐class diversity with different magnification levels and the noise present in the images makes the classification task significantly challenging. Experimental results indicated that the proposed model can efficiently classify various types of microscopic blood cells with good accuracy. The experimental findings accomplished 98.76% validation accuracy and 97.65% test accuracy in complex background conditions. The performance of the model is compared with other state‐of‐the‐art models and the proposed deep neural network performs significantly better than others. The proposed model can be utilized in a real‐time diagnosis system because it is fast, automatic and efficient, which can assist in taking clinical decisions and early diagnosis of haematological disorders.

Abstract Plant diseases are a critical issue in the farming industry, and early identification is essential for plant monitoring. The leaves of plants represent the majority of disease symptoms, however, leaf analysis by specialists in laboratories is expensive and time‐consuming. Hence, there is a necessity for automated and more accurate plant disease detection techniques which can help in diagnosing early symptoms to turn down the economic loss. A deep‐learning‐based technique for detecting and classifying plant diseases from leaf images is provided in this research. A diversified image dataset of plant leaves with 12 distinct crops in 22 different categories was used for this work. Several intra‐class and inter‐class variations in the training dataset make it more complex and challenging to train a deep‐learning model. An exhaustive analysis of different deep neural networks has been done with different combinations of optimizers and learning rates. Additionally, five‐fold cross‐validation and testing on separate test images have been done for a detailed investigation of the trained model in different statistical parameters. The proposed approach extended the results to an average cross‐validation accuracy of 98.68%, and average test accuracy of 97.69% is obtained on unseen images having intra‐class and inter‐class variations.

Cardiac disorders are one of the leading causes of mortality around the globe and early diagnosis of heart diseases can be beneficial for its mitigation. In this article, an artificial intelligence (AI) based device has been proposed, which allows for an automatic and real-time diagnosis of cardiac diseases based on deep learning techniques. The heart sound (phonocardiogram) signal is acquired by a customized designed stethoscope and the signal is processed before analysis using AI methods for the classification of four major cardiac diseases (Aortic Stenosis, Mitral Regurgitation, Mitral Stenosis, and Mitral Valve Prolapse). Two deep learning-based neural networks, one-dimensional (1-D) convolutional neural network (CNN) and spectrogram based 2-D-CNN models from the analysis of these signals has been integrated with a low-cost single-board processor to make a standalone device. All data processing is done in a single hardware setup and user interface is provided allowing the user to control the data accessibility and visibility to generate the diagnostic report. As a result, the developed device has demonstrated to be a valuable low-cost diagnostic tool for both medical professionals and personal usage at home.