ABSTRACT Privacy‐Preserving Data Sharing (PPDS) masks the individual's collected data (e.g., medical healthcare data) before being disseminated by organizations for analysis and research. Patient data contains sensitive values that must be dealt with while ensuring certain privacy conditions are met. This minimizes the risk of re‐identification of an individual record from the group of privacy‐preserved data. However, with the advancement in technology (i.e., Big Data, the Internet of Things (IoT), and Blockchain), the existing classical privacy‐preserving techniques are becoming obsolete. In this paper, we propose a blockchain‐based secure data sharing technique named “SecureChain”, which preserves the privacy of an individual record using local differential privacy (LDP). The three distinguished features of the proposed approach are lower latency, higher throughput, and improved privacy. The proposed model outperforms the benchmarks in terms of both latency and throughput. In terms of precision, the proposed method improves the accuracy to 88.53% compared to its counterparts, which achieved 49% and 85% accuracy. The results of the experiment verify that the proposed approach outperforms its counterparts.

Human emotion recognition (HER) has rapidly advanced, with applications in intelligent customer service, adaptive system training, human–robot interaction (HRI), and mental health monitoring. HER’s primary goal is to accurately recognize and classify emotions from digital inputs. Emotion recognition (ER) and feature extraction have long been core elements of HER, with deep neural networks (DNNs), particularly convolutional neural networks (CNNs), playing a critical role due to their superior visual feature extraction capabilities. This study proposes improving HER by integrating EfficientNet with transfer learning (TL) to train CNNs. Initially, an efficient R‐CNN accurately recognizes faces in online and offline videos. The ensemble classification model is trained by combining features from four CNN models using feature pooling. The novel VGG‐19 block is used to enhance the Faster R‐CNN learning block, boosting face recognition efficiency and accuracy. The model benefits from fully connected mean pooling, dense pooling, and global dropout layers, solving the evanescent gradient issue. Tested on CK+, FER‐2013, and the custom novel HER dataset (HERD), the approach shows significant accuracy improvements, reaching 89.23% (CK+), 94.36% (FER‐2013), and 97.01% (HERD), proving its robustness and effectiveness.

Medical images, particularly chest radiographs and computed tomography (CT), are essential noninvasive diagnostic tools for detecting chest diseases, such as long-term obstructive pulmonary disease, and other respiratory diseases. However, the manual process of detecting diseases and writing their radiology reports is labor-intensive and time-consuming for radiologists. As a result, there has been a growing interest in the development of automated systems for generating radiological chest X-ray (CXR) reports. Although previous research has concentrated on improving the quantitative performance of these automated reports, the quality of the reports has often been neglected. Radiology reports are crucial for communicating with physicians and patients. Therefore, developing automated systems to generate these reports can significantly reduce radiologists’ workload and enhance efficiency in clinical practice by automating the diagnosis of CXR through artificial intelligence (AI) and generating CXR reports. This paper aims to ease the burden especially on radiologists. In this paper, we propose a novel approach using the long short-term memory (LSTM) based model and gated recurrent unit (GRU) model as a decoder in combination with five different convolutional neural network (CNN) models, in which the DenseNet169 model shows the most promising results. Our LSTM-based decoder with the DenseNet169 model achieves the highest results in terms of the BLEU score B1 at 0.5856, B2 at 0.4982, B3 at 0.3470, B4 at 0.1269, ROUGE at 0.4534, and CIDEr at 0.463. Although our GRU-based model achieves the results in terms of the BLEU score (B1: 0.5619, B2: 0.4682, B3: 0.3370, B4: 0.1169), ROUGE (0.4401), and CIDEr (0.4320), the empirical evaluations demonstrate that our proposed approach, especially when combined with DenseNet169, achieves more accurate disease identification and generates reports of fluent and accurate radiological findings. The approach also compared with existing baseline methods.