RNA splicing factor Rbm25 underlies heterogeneous preleukemic clonal expansion in mice.

Clonal expansion sets the stage for cancer genesis by allowing for the accumulation of molecular alterations. Although genetic mutations such as Tet2 that induce clonal expansion and malignancy have been identified, these mutations are also frequently found in healthy individuals. Here, we tracked preleukemic clonal expansion using genetic barcoding in an inducible Tet2 knockout mouse model and found that only a small fraction of hematopoietic stem cells (HSCs) expanded excessively upon Tet2 knockout. These overexpanded HSCs expressed significantly lower levels of genes associated with leukemia and RNA splicing than nonoverexpanded Tet2 knockout HSCs. Knocking down Rbm25, an identified RNA splicing factor, accelerated the expansion of Tet2-knockout hematopoietic cells in vitro and in vivo. Our data suggest that mutations of an epigenetic factor Tet2 induce variability in the expression of an RNA splicing factor Rbm25, which subsequently drives heterogeneous preleukemic clonal expansion. This heterogeneous clonal expansion could contribute to the variable disease risks across individuals.

Clonal-level lineage commitment pathways of hematopoietic stem cells in vivo.

While the aggregate differentiation of the hematopoietic stem cell (HSC) population has been extensively studied, little is known about the lineage commitment process of individual HSC clones. Here, we provide lineage commitment maps of HSC clones under homeostasis and after perturbations of the endogenous hematopoietic system. Under homeostasis, all donor-derived HSC clones regenerate blood homogeneously throughout all measured stages and lineages of hematopoiesis. In contrast, after the hematopoietic system has been perturbed by irradiation or by an antagonistic anti-ckit antibody, only a small fraction of donor-derived HSC clones differentiate. Some of these clones dominantly expand and exhibit lineage bias. We identified the cellular origins of clonal dominance and lineage bias and uncovered the lineage commitment pathways that lead HSC clones to different levels of self-renewal and blood production under various transplantation conditions. This study reveals surprising alterations in HSC fate decisions directed by conditioning and identifies the key hematopoiesis stages that may be manipulated to control blood production and balance.

Functional compensation between hematopoietic stem cell clones in vivo.

In most organ systems, regeneration is a coordinated effort that involves many stem cells, but little is known about whether and how individual stem cells compensate for the differentiation deficiencies of other stem cells. Functional compensation is critically important during disease progression and treatment. Here, we show how individual hematopoietic stem cell (HSC) clones heterogeneously compensate for the lymphopoietic deficiencies of other HSCs in a mouse. This compensation rescues the overall blood supply and influences blood cell types outside of the deficient lineages in distinct patterns. We find that highly differentiating HSC clones expand their cell numbers at specific differentiation stages to compensate for the deficiencies of other HSCs. Some of these clones continue to expand after transplantation into secondary recipients. In addition, lymphopoietic compensation involves gene expression changes in HSCs that are characterized by increased lymphoid priming, decreased myeloid priming, and HSC self-renewal. Our data illustrate how HSC clones coordinate to maintain the overall blood supply. Exploiting the innate compensation capacity of stem cell networks may improve the prognosis and treatment of many diseases.

Mesothelin-targeting chimeric antigen receptor-modified T cells by piggyBac transposon system suppress the growth of bile duct carcinoma.

Chimeric antigen receptor modified T cell-based immunotherapy is revolutionizing the field of cancer treatment. However, its potential in treating bile duct carcinoma has not been fully explored. Herein, we developed the second-generation mesothelin-targeting chimeric antigen receptor-modified T cells with the 4-1BB co-stimulatory module by the piggyBac transposon system. Mesothelin-targeting chimeric antigen receptor was expressed by 66.0% of mesothelin-targeting chimeric antigen receptor-modified T cells post electrophoretic transfection and stimulation with K562-meso cells; the expressions of activation markers were tested by flow cytometry assay and showed greater activation of mesothelin-targeting chimeric antigen receptor-modified T cells than control T cells (CD107α: 71.9% vs 48.6%; CD27: 92.1% vs 61.8%; CD137: 55.5% vs 8.4%; CD28: 98.0% vs 82.1%; CD134: 37.5% vs 10.4%). Furthermore, mesothelin-targeting chimeric antigen receptor-modified T cells exerted cytotoxicity toward mesothelin-expressing EH-CA1b and EH-CA1a cells in an effector-to-target ratio-dependent manner, while leaving mesothelin-negative GSC-SD and EH-GB1 cells and normal liver L02 cells almost unharmed. Mesothelin-targeting chimeric antigen receptor-modified T cells secreted cytokines at higher levels when co-cultured with mesothelin-positive EH-CA1a and EH-CA1b cells than with mesothelin-negative GSC-SD and EH-GB1 cells. Enhanced cytotoxicity and cytokine secretion of mesothelin-targeting chimeric antigen receptor-modified T cells compared to control T cells were also observed when co-cultured with 293-meso cells (interferon γ: 85.1% ± 1.47% vs 8.3% ± 2.50%, p = 0.000; tumor necrosis factor α: 90.9% ± 4.67% vs 18.5% ± 3.62%, p = 0.0004; interleukin 2: 60.8% ± 2.00% vs 15.6% ± 2.06%, p = 0.002; interleukin 6: 6.4% ± 2.95% vs 1.7% ± 0.63%, p = 0.055). In addition, mesothelin-targeting chimeric antigen receptor-modified T cells showed greater inhibitory and proliferative capability than control T cells within EH-CA1a cell xenografts. This study shows the potential of mesothelin-targeting chimeric antigen receptor-modified T cells in treating bile duct carcinoma.

Feasibility of Studying a Brief Intervention to Help Chinese Villagers with Problem Alcohol Use After an Earthquake.

AIM: To evaluate the feasibility of conducting a study of structured brief intervention (BI) for reducing problem alcohol use among individuals who experienced earthquake. METHODS: Following the Wenchuan earthquake, 1336 clients from 18 local hospitals were invited to complete the Alcohol Use Disorders Identification Test (AUDIT). Of those, 239 individuals (AUDIT score of greater than or equal to 7) were included in the study. The participants from intervention village hospitals who were assigned to the BI group (n = 118) received a structured BI lasting 15-30 min plus general health education. The participants from the control village hospitals were assigned to the control group (n = 121) only received general health education. Baseline and post-intervention assessments at 12 weeks were conducted using the AUDIT, Substance Abuse Knowledge Scale (SAKS), Self-rating Depression Scale (SDS), Self-rating Anxiety Scale (SAS) and General Well-being Schedule. RESULTS: At 3 months follow-up, the BI group had reduced scores on AUDIT (F = 65.84; P < 0.001) and increased on SAKS (F = 44.45; P < 0.001), but the control group had increased scores on SAS (F = 10.76; P = 0.001) and SDS (F = 18.43; P < 0.001) compared with baseline. BI group showed more decreases for AUDIT scores (group × time effect, F = 34.8; P < 0.001), and had mores increases for SAKS scores (group × time effect, F = 15.7; P < 0.001) compared with control group. CONCLUSION: The study demonstrated the feasibility of a study of BI in problem alcohol users who experienced traumatic events. Further research need to be done to test the effectiveness of BI over a longer period of time, and provide evidence in support of BI as an effective technique in China.

Simultaneous Calibration: A Joint Optimization Approach for Multiple Kinect and External Cameras.

Camera calibration is a crucial problem in many applications, such as 3D reconstruction, structure from motion, object tracking and face alignment. Numerous methods have been proposed to solve the above problem with good performance in the last few decades. However, few methods are targeted at joint calibration of multi-sensors (more than four devices), which normally is a practical issue in the real-time systems. In this paper, we propose a novel method and a corresponding workflow framework to simultaneously calibrate relative poses of a Kinect and three external cameras. By optimizing the final cost function and adding corresponding weights to the external cameras in different locations, an effective joint calibration of multiple devices is constructed. Furthermore, the method is tested in a practical platform, and experiment results show that the proposed joint calibration method can achieve a satisfactory performance in a project real-time system and its accuracy is higher than the manufacturer's calibration.

Low toxic and high soluble camptothecin derivative 2-47 effectively induces apoptosis of tumor cells in vitro.

The cytotoxic activity of camptothecin derivatives is so high that these compounds need to be further modified before their successful application as anti-cancer agents clinically. In this study, we reported the synthesis and biological evaluation of a novel camptothecin derivative called compound 2-47. The changes in structure did not reduce its activity to inhibit DNA topoisomerase I. Compound 2-47 induced apoptosis of many tumor cells including leukemia cells K562, Jurkat, HL-60, breast cancer cell BT-549, colon cancer cell HT-29 and liver cancer cell HepG2 with a half maximal inhibitory concentration (IC50) of 2- to 3-fold lower than HCPT as a control. In particular, 2-47 inhibited the proliferation of Jurkat cells with an IC50 of as low as 40 nM. By making use of Jurkat cell as a model, following treatment of Jurkat cells, compound 2-47 activated caspase-3 and PARP, resulting in a decreased Bcl-2/Bax ratio. These data showed that compound 2-47 induces Jurkat cell death through the mitochondrial apoptotic pathway. In addition, compound 2-47 showed a decreased cytotoxic activity against normal cells and an improved solubility in low-polar solvent. For example, compound 2-47 solutes in CHCl3 130-fold higher than HCPT. Taken together, our data demonstrated that camptothecin derivative 2-47 notably inhibits the tumor cell proliferation through mitochondrial-mediated apoptosis in vitro.

Synthesis and antitumor activity evaluation of a novel series of xanthone derivatives.

A natural xanthone, 1,3,6-trihydroxy-5-methoxyxanthone, was totally synthesized for the first time by six steps in 31% overall yield. The xanthone skeleton was formed by a one-step synthesis in 80% yield, and five of its novel derivatives were also obtained by this approach. This synthetic strategy and all the derivatives could be further used for the preparation of other natural xanthones. All the xanthones were characterized by NMR and ESI-MS, and the cytotoxicity of these xanthones was evaluated against HepG2 and HT-29 cells, and the preliminary structure-activity relationship was evaluated from the results. It was proved that the presence of 3-OH group in the molecule is crucial for its biological activity, while the presence of substituents at C-5 and C-6 may also be beneficial.

Synthesis and Biological Evaluation of Novel Water-Soluble Poly-(ethylene glycol)-10-hydroxycamptothecin Conjugates.

In order to improve the antitumor activity and water solubility of 10-hydroxycamptothecin (HCPT), a series of novel HCPT conjugates were designed and synthesized by conjugating polyethylene glycol (PEG) to the 10-hydroxyl group of HCPT via a valine spacer. The in vitro stability of these synthesized compounds was determined in pH 7.4 buffer at 37 °C, and the results showed that they released HCPT at different rates. All the compounds demonstrated significant antitumor activity in vitro against K562, HepG2 and HT-29 cells. Among them, compounds, 4a, 4d, 4e and 4f, exhibited 2-5 times higher potency than HCPT. The stability and antitumor activity of these conjugates were found to be closely related to the length of PEG and the linker type, conjugates with a relatively short PEG chain and carbamate linkages (compounds 4a and 4f) exhibited controlled release of HCPT and excellent antitumor in vitro activity.

Grasping detection of dual manipulators based on Markov decision process with neural network.

With the development of artificial intelligence, robots are widely used in various fields, grasping detection has been the focus of intelligent robot research. A dual manipulator grasping detection model based on Markov decision process is proposed to realize the stable grasping with complex multiple objects in this paper. Based on the principle of Markov decision process, the cross entropy convolutional neural network and full convolutional neural network are used to parameterize the grasping detection model of dual manipulators which are two-finger manipulator and vacuum sucker manipulator for multi-objective unknown objects. The data set generated in the simulated environment is used to train the two grasping detection networks. By comparing the grasping quality of the detection network output the best grasping by the two grasping methods, the network with better detection effect corresponding to the two grasping methods of two-finger and vacuum sucker is determined, and the dual manipulator grasping detection model is constructed in this paper. Robot grasping experiments are carried out, and the experimental results show that the proposed dual manipulator grasping detection method achieves 90.6% success rate, which is much higher than the other groups of experiments. The feasibility and superiority of the dual manipulator grasping detection method based on Markov decision process are verified.

UvHOS3-mediated histone deacetylation is essential for virulence and negatively regulates ustilaginoidin biosynthesis in Ustilaginoidea virens.

Ustilaginoidea virens is the causal agent of rice false smut, which has recently become one of the most important rice diseases worldwide. Ustilaginoidins, a major type of mycotoxins produced in false smut balls, greatly deteriorates grain quality. Histone acetylation and deacetylation are involved in regulating secondary metabolism in fungi. However, little is yet known on the functions of histone deacetylases (HDACs) in virulence and mycotoxin biosynthesis in U. virens. Here, we characterized the functions of the HDAC UvHOS3 in U. virens. The ΔUvhos3 deletion mutant exhibited the phenotypes of retarded growth, increased mycelial branches and reduced conidiation and virulence. The ΔUvhos3 mutants were more sensitive to sorbitol, sodium dodecyl sulphate and oxidative stress/H2 O2 . ΔUvhos3 generated significantly more ustilaginoidins. RNA-Seq and metabolomics analyses also revealed that UvHOS3 is a key negative player in regulating secondary metabolism, especially mycotoxin biosynthesis. Notably, UvHOS3 mediates deacetylation of H3 and H4 at H3K9, H3K18, H3K27 and H4K8 residues. Chromatin immunoprecipitation assays indicated that UvHOS3 regulates mycotoxin biosynthesis, particularly for ustilaginoidin and sorbicillinoid production, by modulating the acetylation level of H3K18. Collectively, this study deepens the understanding of molecular mechanisms of the HDAC UvHOS3 in regulating virulence and mycotoxin biosynthesis in phytopathogenic fungi.

Quantitative association between gene expression and blood cell production of individual hematopoietic stem cells in mice.

Individual hematopoietic stem cells (HSCs) produce different amounts of blood cells upon transplantation. Taking advantage of the intercellular variation, we developed an experimental and bioinformatic approach to evaluating the quantitative association between gene expression and blood cell production across individual HSCs. We found that most genes associated with blood production exhibit the association only at some levels of blood production. By mapping gene expression with blood production, we identified four distinct patterns of their quantitative association. Some genes consistently correlate with blood production over a range of levels or across all levels, and these genes are found to regulate lymphoid but not myeloid production. Other genes exhibit one or more clear peaks of association. Genes with overlapping peaks are found to be coexpressed in other tissues and share similar molecular functions and regulatory motifs. By dissecting intercellular variations, our findings revealed four quantitative association patterns that reflect distinct dose-response molecular mechanisms modulating the blood cell production of HSCs.

[Construction of keratinocyte growth factor phage active peptides for the promotion of epidermal cell proliferation].

OBJECTIVE: To construct and display the keratinocyte growth factor (KGF) phage active peptides so as to detect the promoting effects of epidermal cell. METHODS: KGF sequences were chosen and their primers were designed. The selected genes of P1, P2 and P4 were obtained by reverse transcription (RT)-PCR. P3 was obtained by direct synthesis. And the KGF genes were subcloned into pComb3 vector. The technique of phage display was employed to display the genes on phage surface. Methyl thiazolyl tetrazolium (MTT) assay was used to evaluate the promoting effects of KGF phage active peptides on the proliferation of epidermal cell. Optical density (A) was determined at 570 nm. Immunofluorescent assay was employed to evaluate the cell affinity of KGF phage active peptides. RESULTS: The four KGF genes were obtained and subcloned into pComb3 vector. The proteins of the KGF genes were expressed on the surface of the pComb3 vector. The MTT data of optical density (A) showed that significant differences existed between the negative control and KGF control (0.293 ± 0.017 vs 0.520 ± 0.043) and KGF phage active peptide groups (0.293 ± 0.017 vs 0.469 ± 0.057, 0.441 ± 0.048, 0.438 ± 0.035, 0.446 ± 0.037) (all P < 0.01). The results of immunofluorescent assay indicated that KGF and KGF phage active peptides had excellent cell affinity. CONCLUSION: KGF phage active peptides are successfully constructed and displayed and they may promote the proliferation of epidermal cell.

Lineage tracking to reveal the fate of hematopoietic stem cells influenced by Flk2- multipotent progenitors after transplantation.

After transplantation, hematopoietic stem cells (HSCs) sustain blood cell regeneration throughout the patient's life. Recent studies suggest that several types of mature blood cells provide feedback signals to regulate HSC fate. However, the potential feedback effect of hematopoietic progenitor cells has not been characterized to date. The present investigation demonstrated that multipotent progenitors (MPPs) promoted T cell production of HSCs when both cell types were cotransplanted in mice. Using genetic barcodes to track individual HSCs in mice, we found that the increased T cell production by HSCs was associated with the combined effects of altered lineage bias and clonal expansion during HSC differentiation. We showed that MPP and HSC co-transplantation promoted the multilineage differentiation of HSCs in the short term while preserving lymphoid-specialized HSC differentiation in the long term. Our findings indicate that MPPs derived from HSCs regulate the fate of HSCs after bone marrow transplantation.

Multi-view fusion network-based gesture recognition using sEMG data.

sEMG(surface electromyography) signals have been widely used in rehabilitation medicine in the past decades because of their non-invasive, convenient and informative features, especially in human action recognition, which has developed rapidly. However, the research on sparse EMG in multi-view fusion has made less progress compared to high-density EMG signals, and for the problem of how to enrich sparse EMG feature information, a method that can effectively reduce the information loss of feature signals in the channel dimension is needed. In this paper, a novel IMSE (Inception-MaxPooling-Squeeze- Excitation) network module is proposed to reduce the loss of feature information during deep learning. Then, multiple feature encoders are constructed to enrich the information of sparse sEMG feature maps based on the multi-core parallel processing method in multi-view fusion networks, while SwT (Swin Transformer) is used as the classification backbone network. By comparing the feature fusion effects of different decision layers of the multi-view fusion network, it is experimentally obtained that the fusion of decision layers can better improve the classification performance of the network. In NinaPro DB1, the proposed network achieves 93.96% average accuracy in gesture action classification with the feature maps obtained in 300ms time window, and the maximum variation range of action recognition rate of individuals is less than 11.2%. The results show that the proposed framework of multi-view learning plays a good role in reducing individuality differences and augmenting channel feature information, which provides a certain reference for non-dense biosignal pattern recognition.

Path Planning Optimization of Intelligent Vehicle Based on Improved Genetic and Ant Colony Hybrid Algorithm.

Intelligent vehicles were widely used in logistics handling, agriculture, medical service, industrial production, and other industries, but they were often not smooth enough in planning the path, and the number of turns was large, resulting in high energy consumption. Aiming at the unsmooth path planning problem of four-wheel intelligent vehicle path planning algorithm, this article proposed an improved genetic and ant colony hybrid algorithm, and the physical model of intelligent vehicle was established. This article first improved ant colony optimization algorithm about heuristic function with the adaptive change of evaporation factor. Then, it improved the genetic algorithm on fitness function, adaptive adjustment of crossover factor, and mutation factor. Last, this article proposed the improved hybrid algorithm with the addition of a deletion operator, adoption of an elite retention strategy, and addition of suboptimal solutions obtained from the improved ant colony algorithm to improved genetic algorithm to obtain optimized new populations. The simulation environment for this article is windows 10, the processor is Intel Core i5-5257U, the running memory is 4GB, the compilation environment is MATLAB2018b, the number of ant samples is 50, the maximum number of iterations is 100, the initial population size of the genetic algorithm is 200, and the maximum number of iterations is 50. Simulation and physical experiments show that the improved hybrid algorithm is effective. Compared with the traditional hybrid algorithm, the improved hybrid algorithm reduced by 46% in the average number of iterations and 75% in the average number of turns in a simple grid. The improved hybrid algorithm reduced by 47% in the average number of iterations and 21% in the average number of turns in a complex grid. The improved hybrid algorithm works better to reduce the number of turns in simple maps.

A Tandem Robotic Arm Inverse Kinematic Solution Based on an Improved Particle Swarm Algorithm.

The analysis of robot inverse kinematic solutions is the basis of robot control and path planning, and is of great importance for research. Due to the limitations of the analytical and geometric methods, intelligent algorithms are more advantageous because they can obtain approximate solutions directly from the robot's positive kinematic equations, saving a large number of computational steps. Particle Swarm Algorithm (PSO), as one of the intelligent algorithms, is widely used due to its simple principle and excellent performance. In this paper, we propose an improved particle swarm algorithm for robot inverse kinematics solving. Since the setting of weights affects the global and local search ability of the algorithm, this paper proposes an adaptive weight adjustment strategy for improving the search ability. Considering the running time of the algorithm, this paper proposes a condition setting based on the limit joints, and introduces the position coefficient k in the velocity factor. Meanwhile, an exponential product form modeling method (POE) based on spinor theory is chosen. Compared with the traditional DH modeling method, the spinor approach describes the motion of a rigid body as a whole and avoids the singularities that arise when described by a local coordinate system. In order to illustrate the advantages of the algorithm in terms of accuracy, time, convergence and adaptability, three experiments were conducted with a general six-degree-of-freedom industrial robotic arm, a PUMA560 robotic arm and a seven-degree-of-freedom robotic arm as the research objects. In all three experiments, the parameters of the robot arm, the range of joint angles, and the initial attitude and position of the end-effector of the robot arm are given, and the attitude and position of the impact point of the end-effector are set to verify whether the joint angles found by the algorithm can reach the specified positions. In Experiments 2 and 3, the algorithm proposed in this paper is compared with the traditional particle swarm algorithm (PSO) and quantum particle swarm algorithm (QPSO) in terms of position and direction solving accuracy, operation time, and algorithm convergence. The results show that compared with the other two algorithms, the algorithm proposed in this paper can ensure higher position accuracy and orientation accuracy of the robotic arm end-effector. the position error of the algorithm proposed in this paper is 0 and the maximum orientation error is 1.29 × 10-8. while the minimum position error of the other two algorithms is -1.64 × 10-5 and the minimum orientation error is -4.03 × 10-6. In terms of operation time, the proposed algorithm in this paper has shorter operation time compared with the other two algorithms. In the last two experiments, the computing time of the proposed algorithm is 0.31851 and 0.30004s respectively, while the shortest computing time of the other two algorithms is 0.33359 and 0.30521s respectively. In terms of algorithm convergence, the proposed algorithm can achieve faster and more stable convergence than the other two algorithms. After changing the experimental subjects, the proposed algorithm still maintains its advantages in terms of accuracy, time and convergence, which indicates that the proposed algorithm is more applicable and has certain potential in solving the multi-arm inverse kinematics solution. This paper provides a new way of thinking for solving the multi-arm inverse kinematics solution problem.

Improved Multi-Stream Convolutional Block Attention Module for sEMG-Based Gesture Recognition.

As a key technology for the non-invasive human-machine interface that has received much attention in the industry and academia, surface EMG (sEMG) signals display great potential and advantages in the field of human-machine collaboration. Currently, gesture recognition based on sEMG signals suffers from inadequate feature extraction, difficulty in distinguishing similar gestures, and low accuracy of multi-gesture recognition. To solve these problems a new sEMG gesture recognition network called Multi-stream Convolutional Block Attention Module-Gate Recurrent Unit (MCBAM-GRU) is proposed, which is based on sEMG signals. The network is a multi-stream attention network formed by embedding a GRU module based on CBAM. Fusing sEMG and ACC signals further improves the accuracy of gesture action recognition. The experimental results show that the proposed method obtains excellent performance on dataset collected in this paper with the recognition accuracies of 94.1%, achieving advanced performance with accuracy of 89.7% on the Ninapro DB1 dataset. The system has high accuracy in classifying 52 kinds of different gestures, and the delay is less than 300 ms, showing excellent performance in terms of real-time human-computer interaction and flexibility of manipulator control.

Real-Time Target Detection Method Based on Lightweight Convolutional Neural Network.

The continuous development of deep learning improves target detection technology day by day. The current research focuses on improving the accuracy of target detection technology, resulting in the target detection model being too large. The number of parameters and detection speed of the target detection model are very important for the practical application of target detection technology in embedded systems. This article proposed a real-time target detection method based on a lightweight convolutional neural network to reduce the number of model parameters and improve the detection speed. In this article, the depthwise separable residual module is constructed by combining depthwise separable convolution and non-bottleneck-free residual module, and the depthwise separable residual module and depthwise separable convolution structure are used to replace the VGG backbone network in the SSD network for feature extraction of the target detection model to reduce parameter quantity and improve detection speed. At the same time, the convolution kernels of 1 × 3 and 3 × 1 are used to replace the standard convolution of 3 × 3 by adding the convolution kernels of 1 × 3 and 3 × 1, respectively, to obtain multiple detection feature graphs corresponding to SSD, and the real-time target detection model based on a lightweight convolutional neural network is established by integrating the information of multiple detection feature graphs. This article used the self-built target detection dataset in complex scenes for comparative experiments; the experimental results verify the effectiveness and superiority of the proposed method. The model is tested on video to verify the real-time performance of the model, and the model is deployed on the Android platform to verify the scalability of the model.

Low-Illumination Image Enhancement Algorithm Based on Improved Multi-Scale Retinex and ABC Algorithm Optimization.

In order to solve the problems of poor image quality, loss of detail information and excessive brightness enhancement during image enhancement in low light environment, we propose a low-light image enhancement algorithm based on improved multi-scale Retinex and Artificial Bee Colony (ABC) algorithm optimization in this paper. First of all, the algorithm makes two copies of the original image, afterwards, the irradiation component of the original image is obtained by used the structure extraction from texture via relative total variation for the first image, and combines it with the multi-scale Retinex algorithm to obtain the reflection component of the original image, which are simultaneously enhanced using histogram equalization, bilateral gamma function correction and bilateral filtering. In the next part, the second image is enhanced by histogram equalization and edge-preserving with Weighted Guided Image Filtering (WGIF). Finally, the weight-optimized image fusion is performed by ABC algorithm. The mean values of Information Entropy (IE), Average Gradient (AG) and Standard Deviation (SD) of the enhanced images are respectively 7.7878, 7.5560 and 67.0154, and the improvement compared to original image is respectively 2.4916, 5.8599 and 52.7553. The results of experiment show that the algorithm proposed in this paper improves the light loss problem in the image enhancement process, enhances the image sharpness, highlights the image details, restores the color of the image, and also reduces image noise with good edge preservation which enables a better visual perception of the image.