This study proposes an advanced method for plant disease detection utilizing a modified depthwise convolutional neural network (CNN) integrated with squeeze-and-excitation (SE) blocks and improved residual skip connections. In light of increasing global challenges related to food security and sustainable agriculture, this research focuses on developing a highly efficient and accurate automated system for identifying plant diseases, thereby contributing to enhanced crop protection and yield optimization. The proposed model is trained on a comprehensive dataset encompassing various plant species and disease categories, ensuring robust performance and adaptability. By evaluating the model with online random images, demonstrate its significant adaptability and effectiveness in overcoming key challenges, such as achieving high accuracy and meeting the practical demands of agricultural applications. The architectural modifications are specifically designed to enhance feature extraction and classification performance, all while maintaining computational efficiency. The evaluation results further highlight the model’s effectiveness, achieving an accuracy of 98% and an F1 score of 98.2%. These findings emphasize the model’s potential as a practical tool for disease identification in agricultural applications, supporting timely and informed decision-making for crop protection.

The rapid population growth results in a crucial problem in the early detection of diseases in medical research. Among all the cancers unveiled, breast cancer is considered the second most severe cancer. Consequently, an exponential rising in death cases incurred by breast cancer is expected due to the rapid population growth and the lack of resources required for performing medical diagnoses. Utilizing recent advances in machine learning could help medical staff in diagnosing diseases as they offer effective, reliable, and rapid responses, which could help in decreasing the death risk. In this paper, we propose a new algorithm for feature selection based on a hybrid between powerful and recently emerged optimizers, namely, guided whale and dipper throated optimizers. The proposed algorithm is evaluated using four publicly available breast cancer datasets. The evaluation results show the effectiveness of the proposed approach from the accuracy and speed perspectives. To prove the superiority of the proposed algorithm, a set of competing feature selection algorithms were incorporated into the conducted experiments. In addition, a group of statistical analysis experiments was conducted to emphasize the superiority and stability of the proposed algorithm. The best-achieved breast cancer prediction average accuracy based on the proposed algorithm is 99.453%. This result is achieved in an average time of 3.6725 s, the best result among all the competing approaches utilized in the experiments.

Human pose and gesture estimation are crucial in correcting physiotherapy fitness exercises. In recent years, advancements in computer vision and machine learning approaches have led to the development of sophisticated pose estimation models that accurately track and analyze human movements in real time. This technology enables physiotherapists and fitness trainers to gain valuable insights into their client’s exercise forms and techniques, facilitating more effective exercise corrections and personalized training regimens. This research aims to propose an efficient artificial intelligence method for human pose estimation during physiotherapy fitness exercises. We utilized a multi-class exercise dataset based on human skeleton movement points to conduct our experimental research. The dataset comprises 133 features derived from human skeleton movements during various exercises, resulting in high feature dimensionality that affects the performance of human pose estimation with machine learning and deep learning methods. We have introduced a novel Logistic regression Recursive Feature elimination (LogRF) method for feature selection. Extensive experiments demonstrate that using the top twenty selected features, the random forest method outperformed state-of-the-art studies with a high-performance score of 0.998. The performance of each applied method is validated through a k-fold approach and further enhanced using hyperparameter tuning. Our proposed study assists specialists in identifying and addressing potential biomechanical issues, improper postures, and incorrect movement patterns, which are essential for injury prevention and optimizing exercise outcomes. Furthermore, this study enhances the capabilities of remote monitoring and guidance capabilities, allowing physiotherapists to support their patient’s progress with prescribed exercises continually.