A wireless sensor network is a large collection of sensor nodes with limited power supply and constrained computational capability. Due to the restricted communication range and high density of sensor nodes, packet forwarding in sensor networks is usually performed through multi-hop data transmission. Therefore, routing in wireless sensor networks has been considered an important field of research over the past decade. Nowadays, multipath routing approach is widely used in wireless sensor networks to improve network performance through efficient utilization of available network resources. Accordingly, the main aim of this survey is to present the concept of the multipath routing approach and its fundamental challenges, as well as the basic motivations for utilizing this technique in wireless sensor networks. In addition, we present a comprehensive taxonomy on the existing multipath routing protocols, which are especially designed for wireless sensor networks. We highlight the primary motivation behind the development of each protocol category and explain the operation of different protocols in detail, with emphasis on their advantages and disadvantages. Furthermore, this paper compares and summarizes the state-of-the-art multipath routing techniques from the network application point of view. Finally, we identify open issues for further research in the development of multipath routing protocols for wireless sensor networks.

Pests and diseases can cause severe damage to citrus fruits. Farmers used to rely on experienced experts to recognize them, which is a time consuming and costly process. With the popularity of image sensors and the development of computer vision technology, using convolutional neural network (CNN) models to identify pests and diseases has become a recent trend in the field of agriculture. However, many researchers refer to pre-trained models of ImageNet to execute different recognition tasks without considering their own dataset scale, resulting in a waste of computational resources. In this paper, a simple but effective CNN model was developed based on our image dataset. The proposed network was designed from the aspect of parameter efficiency. To achieve this goal, the complexity of cross-channel operation was increased and the frequency of feature reuse was adapted to network depth. Experiment results showed that Weakly DenseNet-16 got the highest classification accuracy with fewer parameters. Because this network is lightweight, it can be used in mobile devices.

The purpose of this study is to analyze the clinical results of patients with scaphoid nonunion treated with arthroscopic bone grafting and K (Kirschner)-wires fixation. We retrospectively reviewed the records of 27 patients with scaphoid nonunion who had been treated with arthroscopic bone grafting and K-wires fixation method from November 2008 to February 2014. The average patient age was 35 years. The time from injury to treatment averaged 45 months. The average follow-up period was 18 months. Bone union was assessed using serial plain radiographs. The functional outcome was evaluated by comparing the modified Mayo wrist score with the visual analog scale (VAS) for pain, which were measured at the time of preoperation and at final follow-up. Union was achieved in 26 of the 27 nonunions (96.29%). The average radiologic union time was 10 weeks. The average VAS score decreased from 6.38 (range, 3–10) preoperatively to 1.59 (range, 0–3) at the final follow-up. The average modified Mayo wrist score improved from 60.19 preoperatively to 83.46 at the final follow-up. According to this score, there were 12 excellent, 6 good, and 9 fair results at the final follow-up. Arthroscopic bone grafting and percutaneous K-wires fixation is an effective treatment method for a scaphoid nonunion and has the advantages of allowing thorough assessment, enabling a comprehensive management approach for scaphoid nonunion in a minimally invasive manner, and this method can also be used for the scaphoid nonunion with SNAC stage I.