With the advent of the "Internet plus" era, the Internet of Things (IoT) is gradually penetrating into various fields, and the scale of its equipment is also showing an explosive growth trend. The age of the "Internet of Everything" is coming. The integration and diversification of IoT terminals and applications make IoT more vulnerable to various intrusion attacks. Therefore, it is particularly important to design an intrusion detection model that guarantees the security, integrity and reliability of the IoT. Traditional intrusion detection technology has the disadvantages of low detection rate and poor scalability, which cannot adapt to the complex and changeable IoT environment. In this paper, we propose a particle swarm optimization-based gradient descent (PSO-LightGBM) for the intrusion detection. In this method, PSO-LightGBM is used to extract the features of the data and inputs it into one-class SVM (OCSVM) to discover and identify malicious data. The UNSW-NB15 dataset is applied to verify the intrusion detection model. The experimental results show that the model we propose is very robust in detecting either normal or various malicious data, especially small sample data such as Backdoor, Shellcode and Worms. INDEX TERMS Intrusion detection, Internet of Things (IoT), particle swarm optimization (PSO), oneclass SVM(OCSVM)

No abstract

Bretschneidera sinensis is an endangered woody species found in East and South China. Comprehensive intraspecies chloroplast genome studies have demonstrated novel genetic resources to assess the genetic variation and diversity of this species. Using genome skimming method, we assembled the whole chloroplast genome of 12 genotypes of B. sinensis from different geographical locations, covering most wild populations. The B. sinensis chloroplast genome size ranged from 158,959 to 159,045 base pairs (bp) and displayed a typical circular quadripartite structure. Comparative analyses of 12 B. sinensis chloroplast genome revealed 33 polymorphic simple sequence repeats (SSRs), 105 polymorphic single nucleotide polymorphisms (SNPs), and 55 indels. Phylogenetic analysis showed that the 12 genotypes were grouped into 2 branches, which is consistent with the geographical distribution (Eastern clade and Western clade). Divergence time estimates showed that the two clades were divergent from 0.6 Ma in the late Pleistocene. Ex situ conservation is essential for this species. In this study, we identified SNPs, indels, and microsatellites of B. sinensis by comparative analyses of chloroplast genomes and determined genetic variation between populations using these genomic markers. Chloroplast genomic resources are also important for further domestication, population genetic, and phylogenetic analysis, possibly in combination with molecular markers of mitochondrial and/or nuclear genomes.

Metastasis is responsible for 90% of all cancer-related deaths in humans, and the development of a rapid and promising solution for an early diagnosis of metastasis is required. The present study proposed a promising method combined with scanned laser pico-projection technique and typical texture feature (i.e., contrast, correlation, energy, entropy, and homogeneity) extraction of gray-level co-occurrence matrix (GLCM) image processing model to classify the low- and high-metastatic cancer cells using five common cancer adenocarcinoma cell line pairs (i.e., HeLa/HeLa-S3, CL1-0/CL1-5, OVTW59-P0/OVTW59-P4, and CE81T-FN/CE81T-FN cell lines). Highly metastatic cancer cells essentially have the highest levels of disorder. Both contrast and entropy refer to the degree of disorder, and energy and homogeneity refer to the degree of uniformity. These four texture features can be effective evaluation indexes for disorder in cancer cells responding to metastatic ability. Texture feature extraction forms reflection images, which are recorded with scanned laser pico-projection system; they effectively bridge the gap in information derived from transmission images. The low- and high-metastatic cancer cells are statistically and effectively classified from the texture feature of GLCM through transmission and reflection images taken with scanned laser pico-projection system. In particular, it only requires several seconds after producing a confluent monolayer of cells and achieves the rapid method with a more reliable diagnostic performance for metastatic ability of cancer cells in vitro or ex vivo.

Oral cancer is one of the most prevalent tumors of the head and neck region. As a result, reliable techniques for detecting are urgently required. Accordingly, the present study proposes an optical method using a Scanned Laser Pico-Projection system (SLPP) and Gray-Level Cooccurrence Matrix (GLCM). The validity of the proposed method is demonstrated using five indexes for quantized criteria for oral cancer classifies. The results show that the cancerous pathological sections have the higher Contrast and Entropy, but the lower Correlation, Homogeneity, and Energy. SLPP system with GLCM image processing can differentiate normal & cancerous pathological sections and it works on both full field analysis and specific tissue analysis. The discrimination of normal and cancerous tissues depends on the disorder caused by unusual proliferation and division of the chromosomes and nuclei. Compared to existing methods, the proposed method approach has many advantages, including a lower cost, a larger sample size and a more reliable diagnostic performance. As a result, it provides a highly promising solution for the pathologists/doctors. 

No abstract

Oral cancer is one of the most widespread tumors of the head and neck region. An earlier diagnosis can help dentist getting a better therapy plan, giving patients a better treatment and the reliable techniques for detecting oral cancer cells are urgently required. This study proposes an optic and automation method using reflection images obtained with scanned laser pico-projection system, and Gray-Level Co-occurrence Matrix for sampling. Moreover, the artificial intelligence technology, Support Vector Machine, was used to classify samples. Normal Oral Keratinocyte and dysplastic oral keratinocyte were simulating the evolvement of cancer to be classified. The accuracy in distinguishing two cells has reached 85.22%. Compared to existing diagnosis methods, the proposed method possesses many advantages, including a lower cost, a larger sample size, an instant, a non-invasive, and a more reliable diagnostic performance. As a result, it provides a highly promising solution for the early diagnosis of oral squamous carcinoma.

No abstract