Head Nodes Research Articles

Lipid metabolism marker CD36 is associated with 18FDG-PET/CT false negative lymph nodes in head and neck squamous cell carcinoma.

Lymph node metastasis frequently occurs in head and neck squamous cell carcinoma (HNSCC) patients, and [18F] fluorodeoxyglucose positron emission tomography with computed tomography (18FDG-PET/CT) examination for lymph node metastasis could result in false negativity and delay following treatment. However, the mechanism and resolution for 18FDG-PET/CT false negatives remain unclear. Our study was aim to found biomarkers for false negativity and true positivity from a metabolic perspective. Ninety-two patients diagnosed with HNSCC who underwent preoperative 18FDG-PET/CT and subsequent surgery in our institution were reviewed. Immunohistochemistry (IHC) examinations of glucose metabolism (GLUT1 and GLUT5), amino acid metabolism4 (GLS and SLC1A5), and lipid metabolism (CPT1A and CD36) markers were conducted on their primary lesion and lymph node sections. We identified specific metabolic patterns of the false-negative group. Significantly, CD36 IHC score of primary lesions was higher in false-negative group than true-positive group. Moreover, we validated pro-invasive biological effects of CD36 by bioinformatics analysis as well as experiments. Conclusion: IHC examination of CD36 expression, which is a lipid metabolism marker, in primary lesions could distinguish HNSCC patients' lymph nodes false negatives in 18FDG-PET/CT.

PO-1614 Head and Neck high-risk lymph nodes detection – a three-dimensional deep learning proposal

PO-1614 Head and Neck high-risk lymph nodes detection – a three-dimensional deep learning proposal

Confidentiality Enhancement Using Spread Spectrum Modulation Technique for Aggregated Data in Wireless Sensor Networks

Wireless Sensor Networks (WSN) play a crucial role in transmitting bulk data remotely on Internet of Things (IoT). The data transmitted by the tiny sensor nodes of the network are basically the physical parameters of the environment like temperature, vibration, pressure etc. There exists a huge co-relation in the data sent by the nodes. This bulk data needs to be aggregated in order to conserve the energy of the network and thereby enhance the network lifetime. Although the aggregation might lead to loss of actual data, aggregating helps in reducing energy requirement during transmission which is of primary importance and hence is considered as a suitable method in WSN. Adding confidentiality to aggregated data helps in sending the data more securely. Spread spectrum modulation is a widely used technique to provide confidentiality in communication systems. In this research work, we implement a data aggregation technique and apply the spread spectrum modulation technique to provide confidentiality to the aggregated data that needs to be transmitted from the cluster head node to the gateway. Here, data aggregation process consists of averaging the number of data that are sensed by a sensor node and transmitting only its average value to the gateway. This reduces the amount of data transmission and helps in conserving energy. Further, the spread spectrum technique implements Binary Phase Shift Keying (BPSK) method with Frequency Hopping Spread Spectrum (FHSS) carried out using six different frequencies on MATLAB 2020a. The simulation results evaluate the performance of the system. The graphs are plotted for modulated and demodulated signals, spread and de-spread sequences, Bit Error Rate (BER) of BPSK/ FHSS over Rayleigh flat fading channel, Power Spectral Density (PSD) and Fast Fourier Transform (FFT) of frequency hopped signal. The results show that BER value decreases with increase in Signal to Noise Ratio (SNR). The maximum power consumption in the network is 7.518 mW at 5MHz frequency, for adding confidentiality to the aggregated data to be transmitted. Thus, the proposed work promises efficient energy consumption, longer network lifetimes with added confidentiality.

Design and analysis of routing algorithm with energy efficiency based on clustering and backwards in IoT system

One of the biggest challenges for Internet of Things (IoT) systems is traffic congestion in large networks. For this reason, the bandwidth should be increased in such systems. In addition, the issue of routing is raised in sending packets from the origin to the destination. Therefore, if there are many IoT devices in the network, it will increase the traffic, which makes faultless routing important in these networks. In this paper, a novel routing method based on Routing Protocol for Low-Power (RPL) is presented to minimize the energy consumption of the Internet of Things. Using the backward method based on the A* method to reduce energy consumption in a large graph, promising nodes are selected. A coordinate node is used to manage packets and transfer them. The selection of the coordinator node helps to receive packets with less energy and less delay from its neighbors, and the head node selects the best coordinator node with the shortest distance and the highest residual energy. The proposed method improves the energy consumption criteria, the delay between nodes, and the network overhead criterion by considering the estimated energy to the destination with the A* method.

Network energy optimization and intelligent routing in WSN applicable for IoT using self‐adaptive coyote optimization algorithm

SummaryThe Internet of Things (IoT) is a recent wireless telecommunications platform, which contains a set of sensor nodes linked by wireless sensor networks (WSNs). These approaches split the sensor nodes into clusters, in which each cluster consists of an exclusive cluster head (CH) node. The major scope of this task is to introduce a novel CH selection in WSN applicable to IoT using the self‐adaptive meta‐heuristic algorithm. This paper aids in providing the optimal routing in the network based on direct node (DN) selection, CH selection, and clone cluster head (CCH) selection. DNs are located near the base station, and it is chosen to avoid the load of CH. The adoption of the novel self‐adaptive coyote optimization algorithm (SA‐COA) is used for the DN selection and CCH selection. When the nodes are assigned in the network, DN and CCH selection is performed by the proposed SA‐COA. Then, the computation of residual energy helps to select the CH, by correlating with the threshold energy. CCH is proposed to copy the data from the CH to avoid the loss of data in transmitting. By forming the CCH, the next CH can be easily elected with the optimal CCH using SA‐COA. From the simulation findings, the best value of the designed SA‐COA‐LEACH model is secured at 1.14%, 3.17%, 1.18%, and 7.33% progressed than self‐adaptive whale optimization algorithm (SAWOA), cyclic rider optimization algorithm (C‐ROA), krill herd algorithm (KHA), and COA while taking several nodes 50. The proposed routing of sensor networks specifies better performance than the existing methods.

Imaging of Neck Nodes in Head and Neck Cancers – a Comprehensive Update

Cervical lymph node metastases from head and neck squamous cell cancers significantly reduce disease-free survival and worsen overall prognosis and, hence, deserve more aggressive management and follow-up. As per the eighth edition of the American Joint Committee on Cancer staging manual, extranodal extension, especially in human papillomavirus-negative cancers, has been incorporated in staging as it is important in deciding management and significantly impacts the outcome of head and neck squamous cell cancer. Lymph node imaging with various radiological modalities, including ultrasound, computed tomography and magnetic resonance imaging, has been widely used, not only to demonstrate nodal involvement but also for guided histopathological evaluation and therapeutic intervention. Computed tomography and magnetic resonance imaging, together with positron emission tomography, are used widely for the follow-up of treated patients. Finally, there is an emerging role for artificial intelligence in neck node imaging that has shown promising results, increasing the accuracy of detection of nodal involvement, especially normal-appearing nodes. The aim of this review is to provide a comprehensive overview of the diagnosis and management of involved neck nodes with a focus on sentinel node anatomy, pathogenesis, imaging correlates (including radiogenomics and artificial intelligence) and the role of image-guided interventions.

English

This paper proposes a solution to enhance the key management of Wireless Sensor Networks (WSNs) by integrating blockchain technology with Hierarchical Identity-Based Encryption. The aim is to establish a highly secure WSN structure. The current wireless network is improved by incorporating blockchain technology to increase security and reliability. However, the security of key management in dynamic WSNs is threatened by the vulnerability of untrusted Head Nodes(HN) to attack. The presence of an untrusted HN imposes a heavy burden on the sensors for key management, in addition to the already existing challenges. In order to overcome these issues, blockchain technology has been used to propose a secure key management scheme (BSKM). We offer a secure node movement technique and a safe cluster-building mechanism, with the blockchain acting as a trust machine to replace the major functions of the Head node. We run extensive simulations for security analyses. The findings show that the BSKM scheme is efficient and effective and is more suitable for enhancing the reliability of dynamic WSNs.

A time sequence coding based node-structure feature model oriented to node classification

Graph data mining is an important method for managing complex systems in artificial intelligence (AI). As an important branch of graph data mining, node classification is widely applied in paper classification in citation networks, user classification in social networks, etc. At present, many graph neural networks (GNNs) have been proposed to realize node classification by obtaining node embeddings encoded by aggregating neighborhood features. Most GNNs use one-hot encoding as the initial node feature when there is no node attribute in the graph. However, one-hot encoding is a global encoding technique that cannot reflect the environmental context of specific nodes. To this end, we proposed a graph structure sequential coding (GSSC) model, including a learnable node-structure feature X, sampler, encoder and classifier, to obtain the structural embeddings which can better reflect the topological structure of non-attribute graphs. For non-attribute graphs, we utilized the learnable feature X as the initial node-structure feature, which could be simultaneously trained with the GSSC model. In addition, we used a decoupling scheme to separate the sampling and encoding processes of GSSC. Therefore, we could use different samplers for sparse and dense graphs. In addition, different sampling orders (i.e., temporality) in node sequences could reflect different semantic associations between the nodes. Therefore, we innovatively employed time sequence models (TSMs) as encoders to encode node sequences with temporality. Based on these TSMs, semantic information of the head node in the node sequence can be captured. The experimental results on five datasets confirmed that our GSSC model performs better than other representative methods in node classification for graphs without node attributes.

Vocational Education Information Technology Based on Cross-Attention Fusion Knowledge Map Recommendation Algorithm

With the rapid development of China’s economic development, the demand for technical talents in all walks of life is becoming more and more urgent. Therefore, the research on the intelligent method of vocational education information is becoming more and more important. In this research, the cross-attention fusion module and attention mechanism are introduced into the knowledge map recommendation algorithm to build an algorithm model. The attention mechanism is used to give corresponding attention to each neighbour node of the head node in the knowledge map, and a weight matrix is established to represent different importances of the additional information contained by each neighbour node, which further improves the interpretability of the recommendation. Finally, the model is analysed experimentally. The results show that CAF is superior to other algorithms in Recall and NDCG, which further verifies that attention mechanism plays a significant role in communication. It can be seen that CAF optimisation model is superior to other algorithms in many tests, and is superior to a similar algorithm MKR, which further verifies the effectiveness and superiority of cross-attention fusion module. The CAF model can still maintain its stability in the case of sparse data.

Reliable Data Transmission in Underwater Wireless Sensor Networks Using a Cluster-Based Routing Protocol Endorsed by Member Nodes

Considering Underwater Wireless Sensor Networks (UWSNs) have limited power resources (low bandwidth, long propagation delays, and non-rechargeable batteries), it is critical that they develop solutions to reduce power usage. Clustering is one solution because it not only saves energy consumption but also improves scalability and data integrity. The design of UWSNs is vital to the development of clustering algorithms. The limited energy of sensor nodes, narrow transmission bandwidth, and unpredictable topology of mobile Underwater Acoustic Wireless Sensor Networks (UAWSNs) make it challenging to build an effective and dependable underwater communication network. Despite its success in data dependability, the acoustic underwater communication channel consumes the greatest energy at a node. Recharging and replacing a submerged node’s battery could be prohibitively expensive. We propose a network architecture called Member Nodes Supported Cluster-Based Routing Protocol (MNS-CBRP) to achieve consistent information transfer speeds by using the network’s member nodes. As a result, we use clusters, which are produced by dividing the network’s space into many minute circular sections. Following that, a Cluster Head (CH) node is chosen for each circle. Despite the fact that the source nodes are randomly spread, all of the cluster heads are linked to the circle’s focal point. It is the responsibility of the MNS-CBRP source nodes to communicate the discovered information to the CH. The discovered data will then be sent to the CH that follows it, and so on, until all data packets have been transferred to the surface sinks. We tested our techniques thoroughly using QualNet Simulator to determine their viability.

An Energy-Efficient Data Aggregation Clustering Algorithm for Wireless Sensor Networks Using Hybrid PSO

Extending the lifetime of wireless sensor networks (WSNs) and minimizing energy costs are the two most significant concerns for data transmission. Sensor nodes are powered by their own battery capacity, allowing them to perform critical tasks and interact with other nodes. The quantity of electricity saved from each sensor together in a WSN has been strongly linked to the network’s longevity. Clustering conserves the most power in wireless transmission, but the absence of a mechanism for selecting the most suitable cluster head (CH) node increases the complexity of data collection and the power usage of the sensor nodes. Additionally, the disparity in energy consumption can lead to the premature demise of nodes, reducing the network’s lifetime. Metaheuristics are used to solve non-deterministic polynomial (NP) lossy clustering problems. The primary purpose of this research is to enhance the energy efficiency and network endurance of WSNs. To address this issue, this work proposes a solution where hybrid particle swarm optimization (HPSO) is paired with improved low-energy adaptive clustering hierarchy (HPSO-ILEACH) for CH selection in cases of data aggregation in order to increase energy efficiency and maximize the network stability of the WSN. In this approach, HPSO determines the CH, the distance between the cluster’s member nodes, and the residual energy of the nodes. Then, ILEACH is used to minimize energy expenditure during the clustering process by adjusting the CH. Finally, the HPSO-ILEACH algorithm was successfully implemented for aggregating data and saving energy, and its performance was compared with three other algorithms: low energy-adaptive clustering hierarchy (LEACH), improved low energy adaptive clustering hierarchy (ILEACH), and enhanced PSO-LEACH (ESO-LEACH). The results of the simulation studies show that HPSO-ILEACH increased the network lifetime, with an average of 55% of nodes staying alive, while reducing energy consumption average by 28% compared to the other mentioned techniques.

Depletion of T and B cells in lymphoid tissues of mice induced by oclacitinib, a Janus kinase inhibitor.

The main purpose of the study was to determine the safety of oclacitinib (OCL), a Janus kinase inhibitor, with respect of its effect on CD4 + and CD8 + T cells as well as B cells in the lymphoid tissue. The mice were treated orally with OCL at a dose of 2.7 mg/kg for 14 days and peripheral blood, head and neck lymph nodes (HNLNs), mediastinal lymph nodes (MLNs) and spleen were collected. The study found that OCL induced depletion of CD4 + T cells in the HNLNs and MLNs, while it did not affect the absolute count of CD8 + T cells in these tissues. Also OCL caused a loss of B cells in the HNLNs, although not in the MLNs. Moreover, OCL depleted B cells in the peripheral blood, but did not affect the absolute count of CD4 + and CD8 + T cells. Thus, it can be concluded that OCL may induce a depletive effect on CD4 + and CD8 + T cells as well as B cells in the lymphoid tissue. This effect should be seen as an unfavorable one, especially in patients with infections. Therefore, a clinical implication is that in such patients, the benefit/risk ratio should be thoroughly considered by clinicians. Moreover, OCL reduced the absolute count of eosinophils, basophils, neutrophils and monocytes. However, it is uncertain whether this effect should be considered to be of clinical importance because the levels of these cells were within the physiological range. It is possible that the depletive effect of OCL toward T and B cells, as well as eosinophils and basophils may contribute to the beneficial effects of the drug in the treatment of skin allergic diseases.

Research on Hierarchical Clustering LEACH Protocol Optimization Algorithm Based on Chebyshev Distance

For the problems of premature death of cluster head nodes, short network life cycle and fast energy consumption in hierarchical clustering LEACH protocol in wireless sensor networks, an improved routing protocol K-means LEACH distance optimization algorithm based on LEACH is proposed in this paper. The improved routing algorithm takes into account the problems such as random selection of cluster heads and tedious calculation of distance formula in LEACH algorithm, so K-means clustering algorithm is introduced. The specific scheme is to first divide the data into K classes, then select the appropriate cluster heads, and introduce the Chebyshev distance formula on the basis of K-means clustering algorithm. From the comparative analysis of the simulation results, the K-means LEACH distance optimization algorithm compared to the original LEACH algorithm wireless network node life cycle by 64%, node energy consumption by 26%, which is conducive to improving the stability of WSN, reducing node energy consumption and extending the network survival time.

Deep reinforcement learning-based collaborative routing algorithm for clustered MANETs

Flexible adaptation to differentiated quality of service (QoS) is quite important for future 6G network with a variety of services. Mobile ad hoc networks (MANETs) are able to provide flexible communication services to users through self-configuration and rapid deployment. However, the dynamic wireless environment, the limited resources, and complex QoS requirements have presented great challenges for network routing problems. Motivated by the development of artificial intelligence, a deep reinforcement learning-based collaborative routing (DRLCR) algorithm is proposed. Both routing policy and subchannel allocation are considered jointly, aiming at minimizing the end-to-end (E2E) delay and improving the network capacity. After sufficient training by the cluster head node, the Q-network can be synchronized to each member node to select the next hop based on local observation. Moreover, we improve the performance of training by considering historical observations, which can improve the adaptability of routing policies to dynamic environments. Simulation results show that the proposed DRLCR algorithm outperforms other algorithms in terms of resource utilization and E2E delay by optimizing network load to avoid congestion. In addition, the effectiveness of the routing policy in a dynamic environment is verified.

PIRAP: Intelligent Hybrid Approach for Secure Data Transmission in Wireless Sensor Networks

The wireless sensor network (WSN) is a mobile adhoc network which has no support of infrastructure. The complexities in network management are critical to resolve even when choosing the cluster head is more efficient. From clustering procedure, appropriate cluster heads are selected with attention to energy saving among member nodes. When it comes to WSN security, trust-based cluster head selection is critical, assuming the cooperation of all sensor nodes. In light of this assumption, existing approaches were unable to assist in identifying the network’s ideal cluster head. Due to the dynamic topology and mobility of nodes in WSNs, security is a challenge. However, secrecy is often accomplished end-to-end via symmetric keys between two corresponding applications. The symmetric key is incompatible with the WSN environment. Additionally, the WSN nodes vary in their qualification for which portion of the data is accessible in which context. This work proposes an elliptic curve cryptography-enabled ciphertext policy attribute-based encryption (ECC-CP-ABE) algorithm for secure data transmission during intra-cluster communication and inter-cluster communication. In this work, we select a reliable node-based trust value mechanism and this reliable node acts as an attribute authority that inter-cluster provides a decryption private key to cluster members who are involved in intra- and inter-cluster communications. In ECC-CP-ABE, the cluster head node (CHN) and cluster member node (CMN) utilize CP-ABE for the encryption of network messages using an access policy matrix A that is computed from an AND–OR operation-based monotonic tree access structure which is defined over a various set of attributes. To ensure authenticity and integrity, the cluster member and CHN sign each ciphertext using an ECC algorithm. For performance evaluation, we use packet delivery ratio, energy consumption, encryption time, network lifetime, and decryption time as metric measures and compared the results of proposed ECC-CP-ABE with two benchmark methods, Secured WSN and Taylor-based Cat Salp Swarm Algorithm. From the results, we analyze that the proposed ECC-CP-ABE reduces energy consumption by 53.2%; increases packet delivery ratio by 98.6%; increases network lifespan by 97.5%, and reduces encryption time by 20[Formula: see text]s and decryption time by 15[Formula: see text]s.

Double Q‐learning‐based adaptive trajectory selection for energy‐efficient data collection in wireless sensor networks

SummaryThis paper proposes a novel distributed stochastic routing strategy using mobile sink based on double Q‐learning algorithm to improve the network performance in wireless sensor network with uncertain communication links. Furthermore, in order to extend network lifetime, a modified leach‐based clustering technique is proposed. To balance the energy dissipation between nodes, the selected cluster head nodes are then rotated based on the newly suggested threshold energy value. The simulation results demonstrate that the proposed algorithms outperform the QWRP, QLMS, ESRP and HACDC in terms of network lifetime by 18.33%, 35.1%, 39.7% and 44.7%, respectively. Moreover, the proposed algorithms considerably enhances the learning rate and hence reduces the data collection latency.

Deep learning for automatic head and neck lymph node level delineation provides expert-level accuracy.

Deep learning-based head and neck lymph node level (HN_LNL) autodelineation is of high relevance to radiotherapy research and clinical treatment planning but still underinvestigated in academic literature. In particular, there is no publicly available open-source solution for large-scale autosegmentation of HN_LNL in the research setting. An expert-delineated cohort of 35 planning CTs was used for training of an nnU-net 3D-fullres/2D-ensemble model for autosegmentation of 20 different HN_LNL. A second cohort acquired at the same institution later in time served as the test set (n = 20). In a completely blinded evaluation, 3 clinical experts rated the quality of deep learning autosegmentations in a head-to-head comparison with expert-created contours. For a subgroup of 10 cases, intraobserver variability was compared to the average deep learning autosegmentation accuracy on the original and recontoured set of expert segmentations. A postprocessing step to adjust craniocaudal boundaries of level autosegmentations to the CT slice plane was introduced and the effect of autocontour consistency with CT slice plane orientation on geometric accuracy and expert rating was investigated. Blinded expert ratings for deep learning segmentations and expert-created contours were not significantly different. Deep learning segmentations with slice plane adjustment were rated numerically higher (mean, 81.0 vs. 79.6, p = 0.185) and deep learning segmentations without slice plane adjustment were rated numerically lower (77.2 vs. 79.6, p = 0.167) than manually drawn contours. In a head-to-head comparison, deep learning segmentations with CT slice plane adjustment were rated significantly better than deep learning contours without slice plane adjustment (81.0 vs. 77.2, p = 0.004). Geometric accuracy of deep learning segmentations was not different from intraobserver variability (mean Dice per level, 0.76 vs. 0.77, p = 0.307). Clinical significance of contour consistency with CT slice plane orientation was not represented by geometric accuracy metrics (volumetric Dice, 0.78 vs. 0.78, p = 0.703). We show that a nnU-net 3D-fullres/2D-ensemble model can be used for highly accurate autodelineation of HN_LNL using only a limited training dataset that is ideally suited for large-scale standardized autodelineation of HN_LNL in the research setting. Geometric accuracy metrics are only an imperfect surrogate for blinded expert rating.

EEDLABA: Energy-Efficient Distance- and Link-Aware Body Area Routing Protocol Based on Clustering Mechanism for Wireless Body Sensor Network

In medical environments, a wireless body sensor network (WBSN) is used to operate remotely, and sensor nodes are employed. It consists of sensor nodes installed on a human body to monitor a patient’s condition, such as heartbeat, temperature, and blood sugar level, and are functionalized and controlled by remote devices. A WBSN consists of nodes that are actually sensors in nature and are operated with a short range of communication. These sensor nodes are fixed with limited computation power and the main concern is energy consumption and path loss. In this paper, we propose a new protocol named energy-efficient distance- and link-aware body area (EEDLABA) with a clustering mechanism and compare it with the current link-aware and energy-efficient body area (LAEEBA) and distance-aware relaying energy-efficient (DARE) routing protocols in a WBSN. The proposed protocol is an extended type of LAEEBA and DARE in which the positive features have been deployed. The clustering mechanism has been presented and deployed in EEDLABA for better performance. To solve these issues in LAEEBA and DARE, the EEDLABA protocol has been proposed to overcome these. Path loss and energy consumption are the major concerns in this network. For that purpose, the path loss and distance models are proposed in which the cluster head (CH) node, coordinator (C) node, and other nodes, for a total of nine nodes, are deployed on a human body. The results have been derived from MATLAB simulations in which the performance of the suggested EEDLABA has been observed in assessment with the LAEEBA and DARE. From the results, it has been concluded that the proposed protocol can perform well in the considered situations for WBSNs.

Ameliorate the performance using soft computing approaches in wireless networks

<span lang="EN-US">Wireless sensor networks are an innovative and rapidly advanced network occupying the broad spectrum of wireless networks. It works on the principle of “use with less expense, effort and with more comfort.” In these networks, routing provides efficient and effective data transmission between different sources to access points using the clustering technique. This work addresses the low-energy adaptive clustering hierarchy (LEACH) protocol’s main backdrop of choosing head nodes based on a random value. In this, the soft computing methods are used, namely the fuzzy approach, to overcome this barrier in LEACH. Our approach’s primary goal is to extend the network lifetime with efficient energy consumption and by choosing the appropriate head node in each cluster based on the fuzzy parameters. The proposed clustering algorithm focused on two fuzzy inference structures, namely Mamdani and Sugeno fuzzy logic models in two scenarios, respectively. We compared our approach with four existing works, the conventional LEACH, LEACH using the fuzzy method, multicriteria cluster head delegation, and fuzzy-based energy efficient clustering approach (FEECA) in wireless sensor network. The proposed scenario based fuzzy LEACH protocol approaches are better than the four existing methods regarding stability, network survivability, and energy consumption.</span>

A hybrid blockchain-based identity authentication scheme for Mobile Crowd Sensing

With the continuous innovative development and popularization of mobile smart devices, the application of Mobile Crowd Sensing (MCS) continues to be studied extensively. However, existing centralized MCS applications that use servers for task publishing and data collection exhibit common problems, such as single points of failure and security vulnerabilities. Accordingly, we proposed a hybrid blockchain-based identity authentication scheme for MCS called HBIA, which uses blockchain technology to resolve the single-point failure problem. HBIA builds a cluster structure based on factors such as geographical location and balance, and uses it to construct a hybrid blockchain, with the cluster head node and internal cluster node authenticating on the public and private chains, respectively. We also implemented zero-knowledge proof (ZKP) to ensure the privacy of participants’ identities, thus balancing the contradiction between blockchain transparency and security. In addition, HBIA uses the zero-knowledge succinct non-interactive argument of knowledge (zk-SNARK) technology to enable off-chain computing and on-chain verification, further reducing the blockchain’s workload. Finally,​ HBIA was evaluated based on the pavement crack detection task and tested on the Ethereum public test network known as Ropsten. The test results indicate that the identity authentication scheme proposed in this paper is superior to existing schemes in terms of authentication time.