Dynamic Multiple Object Segmentation with Spatio-Temporal Filtering.

This article primarily focuses on the localization and extraction of multiple moving objects in images taken from a moving camera platform, such as image sequences captured by drones. The positions of moving objects in the images are influenced by both the camera's motion and the movement of the objects themselves, while the background position in the images is related to the camera's motion. The main objective of this article was to extract all moving objects from the background in an image. We first constructed a motion feature space containing motion distance and direction, to map the trajectories of feature points. Subsequently, we employed a clustering algorithm based on trajectory distinctiveness to differentiate between moving objects and the background, as well as feature points corresponding to different moving objects. The pixels between the feature points were then designated as source points. Within local regions, complete moving objects were segmented by identifying these pixels. We validated the algorithm on some sequences in the Video Verification of Identity (VIVID) program database and compared it with relevant algorithms. The experimental results demonstrated that, in the test sequences when the feature point trajectories exceed 10 frames, there was a significant difference in the feature space between the feature points on the moving objects and those on the background. Correctly classified frames with feature points accounted for 67% of the total frames.The positions of the moving objects in the images were accurately localized, with an average IOU value of 0.76 and an average contour accuracy of 0.57. This indicated that our algorithm effectively localized and segmented the moving objects in images captured by moving cameras.

Hybrid spatial-spectral generative adversarial network for hyperspectral image classification.

In recent years, generative adversarial networks (GNAs), consisting of two competing 2D convolutional neural networks (CNNs) that are used as a generator and a discriminator, have shown their promising capabilities in hyperspectral image (HSI) classification tasks. Essentially, the performance of HSI classification lies in the feature extraction ability of both spectral and spatial information. The 3D CNN has excellent advantages in simultaneously mining the above two types of features but has rarely been used due to its high computational complexity. This paper proposes a hybrid spatial-spectral generative adversarial network (HSSGAN) for effective HSI classification. The hybrid CNN structure is developed for the construction of the generator and the discriminator. For the discriminator, the 3D CNN is utilized to extract the multi-band spatial-spectral feature, and then we use the 2D CNN to further represent the spatial information. To reduce the accuracy loss caused by information redundancy, a channel and spatial attention mechanism (CSAM) is specially designed. To be specific, a channel attention mechanism is exploited to enhance the discriminative spectral features. Furthermore, the spatial self-attention mechanism is developed to learn the long-term spatial similarity, which can effectively suppress invalid spatial features. Both quantitative and qualitative experiments implemented on four widely used hyperspectral datasets show that the proposed HSSGAN has a satisfactory classification effect compared to conventional methods, especially with few training samples.

Left ventricular strain and myocardial work in short-term peritoneal dialysis patients.

BACKGROUND: Initiation of dialysis encompasses new cardiovascular challenges on patients with end-stage renal disease (ESRD). This study used two-dimensional speckle-tracking echocardiography (2D-STE) to investigate the change of left ventricular (LV) myocardial function undergoing peritoneal dialysis (PD) within 1-3 months. METHODS: A total of 56 patients with ESRD and 27 healthy controls were enrolled in this prospective study. Mean duration of PD was 44.41 ± 16.44 days. We evaluated LV myocardial function of patients with ESRD in baseline and within 1-3 months after PD by 2D-STE with global longitudinal strains (GLS) and myocardial work (MW). Based on the level of serum phosphate before PD, patients were divided into two groups: the group with normal serum phosphate or hyperphosphatemia. RESULTS: Compared with healthy controls, patients with ESRD had impaired GLS (p < .001) and increased global work index (GWI) (p = .034), global constructive work (GCW) (p < .001), global wasted work (GWW) (p < .001), and lower global work efficiency (GWE) (p = .002). After PD therapy, GWI (p = .001), GCW (p < .001), and GWW (p = .023) decreased and closed to healthy subjects (p > .05) and no significant improvement was observed in GLS (p = .387). GLS of basal segments worsened in the hyperphosphatemia group (p = .005) and GWW reduced remarkably in the group with normal serum phosphate after PD treatment (p = .008). The change of left ventricular internal diameter in diastole (LVIDd) was the only parameter influenced GWI in post-dialysis patients (ß = 0.324, p = .013). CONCLUSIONS: Short-term PD treatment improved LV MW in ESRD patients. They benefited more when receiving treatment before the increase of serum phosphorus.

Maritime Infrared Small Target Detection Based on the Appearance Stable Isotropy Measure in Heavy Sea Clutter Environments.

Infrared small target detection plays a crucial role in maritime security. However, detecting small targets within heavy sea clutter environments remains challenging. Existing methods often fail to deliver satisfactory performance in the presence of substantial clutter interference. This paper analyzes the spatial-temporal appearance characteristics of small targets and sea clutter. Based on this analysis, we propose a novel detection method based on the appearance stable isotropy measure (ASIM). First, the original images are processed using the Top-Hat transformation to obtain the salient regions. Next, a preliminary threshold operation is employed to extract the candidate targets from these salient regions, forming a candidate target array image. Third, to distinguish between small targets and sea clutter, we introduce two characteristics: the gradient histogram equalization measure (GHEM) and the local optical flow consistency measure (LOFCM). GHEM evaluates the isotropy of the candidate targets by examining their gradient histogram equalization, while LOFCM assesses their appearance stability based on local optical flow consistency. To effectively combine the complementary information provided by GHEM and LOFCM, we propose ASIM as a fusion characteristic, which can effectively enhance the real target. Finally, a threshold operation is applied to determine the final targets. Experimental results demonstrate that our proposed method exhibits superior comprehensive performance compared to baseline methods.

The TOX subfamily: all-round players in the immune system.

The thymocyte selection-related HMG box protein (TOX) subfamily comprises evolutionarily conserved DNA-binding proteins, and is expressed in certain immune cell subsets and plays key roles in the development of CD4+ T cells, innate lymphoid cells (ILCs), T follicular helper (Tfh) cells, and in CD8+ T-cell exhaustion. Although its roles in CD4+ T and natural killer (NK) cells have been extensively studied, recent findings have demonstrated previously unknown roles for TOX in the development of ILCs, Tfh cells, as well as CD8+ T-cell exhaustion; however, the molecular mechanism underlying TOX regulation of these immune cells remains to be elucidated. In this review, we discuss recent studies on the influence of TOX on the development of various immune cells and CD8+ T-cell exhaustion and the roles of specific TOX family members in the immune system. Moreover, this review suggests candidate regulatory targets for cell therapy and immunotherapies.

Active non-uniform illumination-based underwater polarization imaging method for objects with complex polarization properties.

Active polarization imaging is one of the most effective underwater optical imaging methods that can eliminate the degradation of image contrast and clarity caused by macro-molecule scattering. However, the non-uniformity of active illumination and the diversity of object polarization properties may decrease the quality of underwater imaging. This paper proposes a non-uniform illumination-based active polarization imaging method for underwater objects with complex optical properties. Firstly, illumination homogenization in the frequency domain is proposed to extract and homogenize the natural incident light from the total receiving light. Then, the weight values of the polarized and non-polarized images are computed according to each pixel's degree of linear polarization (DoLP) in the original underwater image. By this means, the two images can be fused to overcome the problem of reflected light loss generated by the complex polarization properties of underwater objects. Finally, the fusion image is normalized as the final result of the proposed underwater polarization imaging method. Both qualitative and quantitative experimental results show that the presented method can effectively eliminate the uneven brightness of the whole image and obtain the underwater fusion image with significantly improved contrast and clarity. In addition, the ablation experiment of different operation combinations shows that each component of the proposed method has noticeable enhancement effects on underwater polarization imaging. Our codes are available in Code 1.

Unsupervised end-to-end infrared and visible image fusion network using learnable fusion strategy.

Infrared and visible image fusion aims to reconstruct fused images with comprehensive visual information by merging the complementary features of source images captured by different imaging sensors. This technology has been widely used in civil and military fields, such as urban security monitoring, remote sensing measurement, and battlefield reconnaissance. However, the existing methods still suffer from the preset fusion strategies that cannot be adjustable to different fusion demands and the loss of information during the feature propagation process, thereby leading to the poor generalization ability and limited fusion performance. Therefore, we propose an unsupervised end-to-end network with learnable fusion strategy for infrared and visible image fusion in this paper. The presented network mainly consists of three parts, including the feature extraction module, the fusion strategy module, and the image reconstruction module. First, in order to preserve more information during the process of feature propagation, dense connections and residual connections are applied to the feature extraction module and the image reconstruction module, respectively. Second, a new convolutional neural network is designed to adaptively learn the fusion strategy, which is able to enhance the generalization ability of our algorithm. Third, due to the lack of ground truth in fusion tasks, a loss function that consists of saliency loss and detail loss is exploited to guide the training direction and balance the retention of different types of information. Finally, the experimental results verify that the proposed algorithm delivers competitive performance when compared with several state-of-the-art algorithms in terms of both subjective and objective evaluations. Our codes are available at https://github.com/MinjieWan/Unsupervised-end-to-end-infrared-and-visible-image-fusion-network-using-learnable-fusion-strategy.

Postsystolic shortening and early systolic lengthening for early detection of myocardial involvement in patients with systemic lupus erythematosus.

BACKGROUND: Cardiac involvement predicts a poor prognosis in patients with systemic lupus erythematosus (SLE). Two-dimensional speckle-tracking echocardiography (2D-STE) are used to identify subclinical myocardial involvement in various diseases. This study objected to evaluate postsystolic shortening (PSS) and early systolic lengthening (ESL) by 2D-STE for early detection of myocardial involvement in patients with SLE. METHODS: A total of 121 patients with preserved left ventricular ejection fraction (LVEF) in SLE and 30 healthy controls underwent standard 2D-STE in our study. According to SLE disease activity index (SLEDAI), we divided SLE patients into two groups: the group of inactive disease (SLEDAI ≤ 4) and active disease (SLEDAI ≥ 5). The maximum of postsystolic strain index (PSImax ) and early systolic strain index (ESImax ) were acquired from 17 segments of left ventricular (LV). We also compared the PSImax and ESImax of basal, medial, and apical segments between SLE patients and controls. RESULTS: Compared with healthy controls and the group of SLEDAI ≤ 4, the group of SLEDAI ≥ 5 had higher PSImax and ESImax value of global LV and basal segments. The absolute value of global longitudinal strain (GLS) had no difference between the group of active disease and inactive disease. Multivariate analysis demonstrated that PSS was independently associated with SLEDAI and diabetes mellitus. CONCLUSIONS: Detection of PSS and ESL enable to identify LV systolic impairment in SLE patients at an early stage.

Bandwidth extension method based on the field-shunting effect in a high-gain photoelectric receiver circuit.

In the high-gain photoelectric receiver circuit, the method based on the field-shunting effect is applied to improve the bandwidth of the transimpedance amplifier. This method is implemented by adding a ground trace under the gain resistor, which reduces the parasitic capacitance of the gain resistor and thus increases the bandwidth. To obtain the specific impact of this method on bandwidth, a series of simulations are carried out, including electromagnetic simulations of a three-dimensional structure of circuit gain part and simulation program with integrated circuit emphasis (SPICE) simulations of the high-gain voltage-current feedback transimpedance amplifier. Finally, the optimal simulation result shows that selecting a 1206 size chip fixed resistor and setting the ground trace width to 1.1 mm can greatly reduce the influence of resistor parasitic effects on the circuit, thereby achieving the best performance of bandwidth extension. Further, the comparative experiment also verifies the effectiveness of the method for bandwidth enhancement.

Pulsed light time-of-flight measurement based on a differential hysteresis timing discrimination method.

In the pulsed light time-of-flight (ToF) measurement, the timing point generated in the receiver channel is very important to the measurement accuracy. Therefore, a differential hysteresis timing discrimination method is proposed to generate timing points of the receiver channel. This method is based on utilizing the unbalanced characteristics of the fully differential operational amplifier circuit as well as introducing extra hysteresis levels to achieve the stable generation of timing points. With this method, fewer circuit components are consumed and the dynamic range of the receiver channel is not limited by its linear range. The experiments demonstrate that a receiver channel applying the proposed discrimination reaches better single shot accuracy compared to that using leading-edge timing discrimination. This method is also suitable for the timing walk error compensation by means of pulse width. Finally, these results verify the effectiveness of the proposed method in pulsed light ToF measurement.

Calculating the model of a nondiagonal rotation invariant angle for a complicated scatterer.

In the research on scattering polarimetry, a scattering mechanism is described by an internal degree of freedom called rotation invariant parameter α. Traditionally, α is calculated by the diagonal scattering matrix of the single scatterer. However, when the research object is a scatterer with complicated surface and microstructure, the traditional calculation of parameter α is biased, since the corresponding scattering matrix is a nondiagonal matrix. To address this problem, this paper proposes a scientific model based on Cameron decomposition to raise the accuracy of parameter α in the complicated scatterer. The rotation invariant parameter with higher accuracy is renamed as a nondiagonal rotation invariant angle. In the verified experiments, the experimental values of each nondiagonal rotation invariant angle are compared with the referenced values calculated by optical constant and incident angle. The results demonstrate that fewer residuals are achieved than by the traditional method. Based on the presented calculating model, the scattering mechanism difference interval between two different materials is proposed as a judging area to distinguish the differences between scattering mechanisms in application.

Stokes-vector-based polarimetric imaging system for adaptive target/background contrast enhancement.

A novel method to optimize the polarization state of a polarimetric imaging system is proposed to solve the problem of target/background contrast enhancement in an outdoor environment adaptively. First, the last three elements of the Stokes vector are selected to be the observed object's polarization features, the discriminant projection of which is regarded as the detecting function of our imaging system. Then, the polarization state of the system, which can be seen as a physical classifier, is calculated by training samples with a support vector machine method. Finally, images processed by the system with the designed optimal polarization state become discriminative output directly. By this means, the target/background contrast is enhanced greatly, which results in a more accurate and convenient target discrimination. Experimental results demonstrate that the effectiveness and discriminative ability of the optimal polarization state are credible and stable.

Robust infrared small target detection via non-negativity constraint-based sparse representation.

Infrared (IR) small target detection is one of the vital techniques in many military applications, including IR remote sensing, early warning, and IR precise guidance. Over-complete dictionary based sparse representation is an effective image representation method that can capture geometrical features of IR small targets by the redundancy of the dictionary. In this paper, we concentrate on solving the problem of robust infrared small target detection under various scenes via sparse representation theory. First, a frequency saliency detection based preprocessing is developed to extract suspected regions that may possibly contain the target so that the subsequent computing load is reduced. Second, a target over-complete dictionary is constructed by a varietal two-dimensional Gaussian model with an extent feature constraint and a background term. Third, a sparse representation model with a non-negativity constraint is proposed for the suspected regions to calculate the corresponding coefficient vectors. Fourth, the detection problem is skillfully converted to an l1-regularized optimization through an accelerated proximal gradient (APG) method. Finally, based on the distinct sparsity difference, an evaluation index called sparse rate (SR) is presented to extract the real target by an adaptive segmentation directly. Large numbers of experiments demonstrate both the effectiveness and robustness of this method.

Regulation of DNA-PK activity promotes the progression of TNBC via enhancing the immunosuppressive function of myeloid-derived suppressor cells.

BACKGROUND: DNA-dependent protein kinase (DNA-PK) is engaged in DNA damage repair and is significantly expressed in triple negative breast cancer (TNBC). Inhibiting DNA-PK to reduce DNA damage repair provides a possibility of tumor treatment. NU7441, a DNA-PK inhibitor, can regulate the function and differentiation of CD4+ T cells and effectively enhance immunogenicity of monocyte-derived dendritic cells. However, the effect of NU7441 on the tumor progression activity of immunosuppressive myeloid-derived suppressor cells (MDSCs) in TNBC remains unclear. RESULTS: In this study, we found that NU7441 alone significantly increased tumor growth in 4 T1 (a mouse TNBC cell line) tumor-bearing mice. Bioinformatics analysis showed that DNA-PK and functional markers of MDSCs (iNOS, Arg1, and IDO) tended to coexist in breast cancer patients. The mutations of these genes were significantly correlated with lower survival in breast cancer patients. Moreover, NU7441 significantly decreased the percentage of MDSCs in peripheral blood mononuclear cells (PBMCs), spleen and tumor, but enhanced the immunosuppressive function of splenic MDSCs. Furthermore, NU7441 increased MDSCs' DNA-PK and pDNA-PK protein levels in PBMCs and in the spleen and increased DNA-PK mRNA expression and expression of MDSCs functional markers in splenic MDSCs from tumor-bearing mice. NU7441 combined with gemcitabine reduced tumor volume, which may be because gemcitabine eliminated the remaining MDSCs with enhanced immunosuppressive ability. CONCLUSIONS: These findings highlight that the regulation of DNA-PK activity by NU7441 promotes TNBC progression via enhancing the immunosuppressive function of MDSCs. Moreover, NU7441 combined with gemcitabine offers an efficient therapeutic approach for TNBC and merits deeper investigation.

5-HT induces regulatory B cells in fighting against inflammation-driven ulcerative colitis.

Serotonin (5-hydroxytryptamine or 5-HT) is a neuroendocrine peptide endowed with immunomodulatory functions. Regulatory B cells (Bregs) play an important role in maintaining intestinal immune homeostasis. We analyzed the differences of 5-HT and Bregs between peripheral blood of ulcerative colitis (UC) and healthy controls (HC). Besides, 5-HT-treated B cells were adoptively transferred into colitis mice to elucidate the role of 5-HT in regulating Bregs. The level of serum 5-HT and IL-10 in UC patients was lower and both were negatively correlated with disease activity. 5-HT7 receptor (5-HT7R) was higher expressed on Bregs in UC. 5-HT promoted IL-10 production in Bregs through the activation of STAT3. And adoptive transfer of 5-HT-treated B cells alleviated intestinal inflammation via inducing IL-10-producing B cells in mice. Our results suggest that 5-HT/5-HT7R signaling pathway facilitate functional Bregs in constraining inflammation in UC, which may be a new potential prospect in the treatment of UC.

Non-invasive left ventricular myocardial work identifies subclinical myocardial involvement in patients with systemic lupus erythematosus.

OBJECTIVE: Global myocardial work (MW) is a novel indicator that accounts for deformation and afterload, which may provide additional value for assessment of myocardial function. Non-invasive echocardiographic estimated left ventricular (LV) MW incorporates longitudinal strain curves and blood pressure data. This study sought to assess MW in systemic lupus erythematosus (SLE) patients with normal LV ejection fraction (LVEF) by two-dimensional speckle-tracking imaging (2D-STI) to reflect subclinical myocardial damage. METHODS: 98 SLE patients and 98 gender and age-matched healthy subjects were included. The patients with SLE were divided into mild activity (SLE disease activity index (SLEDAI) ≤ 4; n = 45), moderate activity (5 ≤ SLEDAI ≤ 9; n = 23), and high activity (SLEDAI ≥ 10; n = 30) subgroups. Standard transthoracic echocardiography was applied to evaluate the systolic myocardial function of the global LV. The parameters of non-invasive MW including global wasted work (GWW) and global work efficiency (GWE) were calculated from echocardiographic LV pressure-strain loops (PSL) and blood pressure at rest. RESULTS: The SLE group had a significantly higher GWW (75.7 ± 39.1 mmHg% vs 37.9 ± 18.0 mmHg%, P < 0.001) and decreased GWE ratio (95.5 ± 2.0% vs 97.4 ± 1.0%, P < 0.001) compared with the controls. Among the subgroups with elevating level of disease activity, SLE patients with preserved LVEF had a significantly higher GWW (61.6 ± 29.9 mmHg% to 96.2 ± 42.2 mmHg%, P for trend = 0.001) and markedly decreased GWE (96.4 ± 1.5% to 94.4 ± 2.0%, P for trend = 0.001). In two separate multiple linear regression analyses, SLEDAI were independently associated with GWW (ß = 0.271, P = 0.005) and GWE (ß = -0.354, P<0.001). CONCLUSION: GWW and GWE are promising novel tools for the early detection of subclinical LV dysfunction. GWW and GWE could distinguish distinct patterns in different grades of SLEDAI.

An improved U-net based retinal vessel image segmentation method.

Diabetic retinopathy is not just the most common complication of diabetes but also the leading cause of adult blindness. Currently, doctors determine the cause of diabetic retinopathy primarily by diagnosing fundus images. Large-scale manual screening is difficult to achieve for retinal health screen. In this paper, we proposed an improved U-net network for segmenting retinal vessels. Firstly, due to the lack of retinal data, pre-processing of the raw data is required. The data processed by grayscale transformation, normalization, CLAHE, gamma transformation. Data augmentation can prevent overfitting in the training process. Secondly, the basic network structure model U-net is built, and the Bi-FPN network is fused based on U-net. Datasets from a public challenge are used to evaluate the performance of the proposed method, which is able to detect vessel SP of 0.8604, SE of 0.9767, ACC of 0.9651, and AUC of 0.9787.

Long non-coding RNA profiles in plasma exosomes of patients with gastric high-grade intraepithelial neoplasia.

Long non-coding (lnc) RNAs in circulating exosomes are a new class of promising cancer biomarkers; however, their expression in exosomes derived from gastric high-grade intraepithelial neoplasia (GHGIN) has not been reported. In the present study, differentially expressed (DE) lncRNAs were analyzed in the peripheral blood collected from 5 patients with GHGIN and 5 healthy donors using high-throughput sequencing. Reverse transcription-quantitative PCR analysis was performed on 6 randomly selected DE lncRNAs to validate the reliability of the sequencing results. The potential roles of the DE lncRNAs in GHGIN were investigated using Gene Ontology (GO) and Kyoto Encyclopedia of Gene and Genome (KEGG) pathway enrichment analyses. A total of 25,145 lncRNAs were identified in all the samples and 83 DE lncRNAs were further screened, including 76 upregulated and 7 downregulated DE lncRNAs. GO and KEGG analyses predicted that the DE lncRNAs played notable roles in 'protein/macromolecule glycosylation', 'regulation of protein ubiquitination', 'renin-angiotensin system' and 'MAPK signaling pathways'. A lncRNA-micro (mi)RNA-mRNA interaction network was constructed and used to perform association analyses. It was found that 83 lncRNAs were abnormally expressed in GHGIN, with some potential functions associated with gastric cancer. Furthermore, the lncRNA-miRNA-mRNA interaction network indicated that 7 DE lncRNAs may play a notable role in the occurrence and development of GHGIN. The results of the present study showed the expression profiles of lncRNAs in human GHGIN, elucidated some of the molecular changes associated with GHGIN and improved the understanding of the molecular mechanisms underlying GHGIN and gastric cancer.

Myeloid-Derived Suppressor Cells in Patients With Acute Pancreatitis With Increased Inhibitory Function.

Acute pancreatitis (AP) is pancreatic or systemic inflammation without or with motion organ dysfunction. Severe acute pancreatitis (SAP) is the main cause of death for patients with AP. A pro-/anti-inflammatory imbalance is considered the key regulation of disease severity. However, the real mechanism of SAP remains unclear. This study aimed to identify the frequency and specific roll of myeloid-derived suppressor cell (MDSC) in AP. We evaluated MDSC frequency and disease severity by analyzing MDSCs in the peripheral blood of healthy controls (HCs) and patients with mild acute pancreatitis (MAP) and SAP by flow cytometry. We also compared the frequency and inhibitory ability of MDSCs from HCs and SAP, and finally detected the reason for the difference in inhibitory ability. AP was marked by expansion of MDSCs as well as its subsets, granulocytic MDSCs (G-MDSCs) and monocytic MDSCs (M-MDSCs). The proportion of MDSC in the peripheral blood mononuclear cells of patients with AP was increased and positively correlated with AP severity. The frequency of MDSC was decreased after treatment compared with pre-treatment. CD3+ T cells were remarkably inhibited by MDSC derived from the patients with SAP. In the expression of arginase-1 (Arg-1) and reactive oxygen species (ROS), the MDSCs from patients with SAP increased. These findings demonstrated that MDSCs expanded in the peripheral blood in patients with AP, especially in those with SAP. Moreover, the inhibitory ability of MDSCs was increased in the patients with SAP compared with that in the HCs. The enhanced suppressive function was possibly caused by an overexpression of Arg-1 and ROS.

Novel Immune Subsets and Related Cytokines: Emerging Players in the Progression of Liver Fibrosis.

Liver fibrosis is a pathological process caused by persistent chronic injury of the liver. Kupffer cells, natural killer (NK) cells, NKT cells, and dendritic cells (DCs), which are in close contact with T and B cells, serve to bridge innate and adaptive immunity in the liver. Meanwhile, an imbalanced inflammatory response constitutes a challenge in liver disease. The dichotomous roles of novel immune cells, including T helper 17 (Th17), regulatory T cells (Tregs), mucosa-associated invariant T cells (MAIT), and innate lymphoid cells (ILCs) in liver fibrosis have gradually been revealed. These cells not only induce damage during liver fibrosis but also promote tissue repair. Hence, immune cells have unique, and often opposing, roles during the various stages of fibrosis. Due to this heterogeneity, the treatment, or reversal of fibrosis through the target of immune cells have attracted much attention. Moreover, activation of hepatic stellate cells (HSCs) constitutes the core of fibrosis. This activation is regulated by various immune mediators, including Th17, Th22, and Th9, MAIT, ILCs, and γδ T cells, as well as their related cytokines. Thus, liver fibrosis results from the complex interaction of these immune mediators, thereby complicating the ability to elucidate the mechanisms of action elicited by each cell type. Future developments in biotechnology will certainly aid in this feat to inform the design of novel therapeutic targets. Therefore, the aim of this review was to summarize the role of specific immune cells in liver fibrosis, as well as biomarkers and treatment methods related to these cells.