Search

Cancer is one of the most dangerous diseases worldwide. Accurate cancer subtype classification facilitates both diagnosis and prognosis. Currently, studies focused on cancer subtype classification mainly utilize invasive histopathological images. However, non-invasive PET/CT images are often the first choice for initial screening and diagnosis of tumors. In addition, existing deep learning models strive to find accurate classification boundaries. It is difficult to make a definitive diagnosis with radiological imaging studies alone. To address these issues, we proposed a classification method embedding atypical patterns for distinguishing tumor subtypes in PET/CT images. First, we introduced a novel pattern class division method, including the original classes representing typical patterns and the extended classes representing atypical patterns. We optimized the distribution of feature vectors through a deep metric learning-based method. Then, we proposed a fuzzy classification loss that employs one or more proxies to represent a pattern class. This loss optimizes the spatial distribution of proxies with their corresponding subsets. Finally, we proposed a fuzzy classification method to predict the class of a sample. The experimental results on two datasets show that the proposed model can separate the samples with atypical features and has high accuracy for samples with typical features. For more difficult tasks, the improvement using the proposed model is more obvious. Furthermore, using the proposed model has good noise immunity and interpretability, which is helpful for clinical auxiliary diagnosis. • Develop a PET/CT tumor subtypes classification method embedding atypical patterns. • Introduce a novel pattern class division method. • Propose a fuzzy classification loss that can separate samples with atypical features. • The proposed method can effectively alleviate the label noise. • The proposed method is more consistent with the diagnostic workflow. [ABSTRACT FROM AUTHOR]

Lymphoma segmentation on positron emission tomography/computed tomography (PET/CT) is a challenge in clinical diagnosis. Existing methods commonly locate lymphoma candidates on PET images and removes false positive candidates using CT images. However, the combined use of PET with CT in segmentation is not trivial, either requiring CT-based multi-organ segmentation or PET/CT registration. The registration of PET/CT will introduce information change in images. Therefore, accurate lymphoma segmentation on PET images is of great importance for medical diagnoses. In this paper, we propose a novel idea of lymphoma segmentation on multi-temporal resolution PET (MTR-PET) images. Instead of using CT information, this method investigates differences of lymphoma and other tissues on different temporal resolution PET images. Two related features, metabolic variation and metabolic heterogeneity, are proposed and combined with traditional features to construct a statistical analysis model for removing false lymphoma candidates. To obtain accurate boundary results, CNN networks are used for lymphoma segmentation on the region of interesting (ROI) images. Our proposed method is evaluated on 53 MTR-PET images with a detection sensibility of 0.9953 and a segmentation Dice coefficient of 0.8667. Experiments and results demonstrate that information of MTR-PET images can significantly improve the performance of lymphoma segmentation. • A 2-stage method for tumor segmentation with convolutional neural network. • Multi-temporal resolution PET imaging analysis. • A statistical analysis model for removing false candidates with multi-temporal resolution PET. • High-precision lymphoma segmentation is accomplished using one modality (PET). [ABSTRACT FROM AUTHOR]

Accurate segmentation of hepatocellular carcinoma (HCC) nuclei is of great importance in automatic pathologic diagnosis. This paper proposes structure convolutional extreme learning machine (SC-ELM) and case-based shape template (CBST) methods for HCC nucleus segmentation, which could tackle complex nucleus scenarios including adhesion or overlap. First, SC-ELM is developed for global segmentation of pathology images, which is used for coarse segmentation. Then, each connected region is considered as a nucleus clump and a probability model with three energy functions is proposed for contour refinement of nucleus clumps. Finally, for complex nucleus clumps, the CBST method combined with pixel-based classification is utilized for unclear or lost boundary inference. Experimentations with 127 liver pathology images demonstrate the performance advantages of our proposed method as compared with related work. [ABSTRACT FROM AUTHOR]

In this study, two polysaccharide fractions (RRP1: Mw = 5.5 kDa, and RRP2: Mw = 425.7 kDa) were isolated from Rhodiola rosea to investigate their antioxidation and hepatoprotective effects. Physicochemical analysis showed that RRP1 was composed of mannose, rhamnose, galacturonic acid, glucose, galactose and arabinose with a relative molar ratio of 0.69:0.11:0.15:1:0.51:7.5 and RRP2 was consisted of mannose, rhamnose, galacturonic acid, glucose, galactose and arabinose (relative molar ratio = 0.15:0.19:1.01:0.18:0.47:1). Periodate oxidation and Smith degradation analysis revealed that, in RRP1, part of the arabinose and glucose residues were 1 → 3,6/1 → 3/1 → 2,3/1 → 3,4/1 → 2,4/1 → 2,3,4-linked, and the mannose, rhamnose and galactose residues were 1 → 2,6/1 → 6/1 → 2/1→/1 → 4,6/1 → 4-linked. In RRP2, the rhamnose, glucose and galactose residues were linked by 1 → 3,6/1 → 3/1 → 2,3/1 → 3,4/1 → 2,4/1 → 2,3,4 linkages, and the arabinose and mannose residues were 1 → 2/1 → 6/1 → 4-linked. The methylation analysis confirmed the structure information of the two fractions. Importantly, fraction RRP1 demonstrated stronger antioxidative activities than RRP2 by scavenging DPPH, hydroxyl and superoxide anion radicals in vitro. Correspondently, RRP1 showed more significant effects than RRP2 on decreasing the levels of ALT, AST and MDA, and increasing the GSH, SOD and CAT levels in the CCl 4 -treated mice. These data demonstrated that the polysaccharide RRP1 could be developed as a promising candidate for preventing and treating liver damage induced by toxic chemicals. [ABSTRACT FROM AUTHOR]

A robust PET-CT segmentation network should guarantee that models trained on the PET-CT images will still work when only CT images are available. It is particularly important due to the radioactivity and expensive cost of PET imaging, in many cases only CT images can be obtained. Disentanglement and Generative Adversarial Networks (GAN) are two commonly used strategies to deal with the missing modality. Disentanglement methods cannot successfully disentangle PET-CT images into modal features and anatomical features because PET-CT images do not satisfy anatomical information consistency constraints. GAN networks tend to ignore information that is critical for downstream tasks, such as tumor information. To address above issues, we propose a siamese semi-disentanglement network. We extract high-level shared tumor features from PET images and CT images instead of anatomical features for downstream segmentation tasks. Meanwhile, in order to leverage low-level entanglement features during segmentation, GAN is used to generate synthetic PET images from CT images. Siamese Consistency Module (SCM) is proposed to ensure that the entanglement low-level features of the synthetic PET images are consistent with the real PET images. The motivation of our proposed method is that the entanglement information discarded by the semi-disentanglement is compensated by GAN to get rid of the anatomical information consistency constraints. Also, the GAN can better retain tumor information through semi-disentanglement. We do experiments on two public PET-CT datasets and one private dataset: Soft-Tissue-Sarcoma (STS) dataset, HeadNeck dataset and LiverTumor dataset. The results show that our proposed method can successfully achieve robust PET-CT segmentation. Our proposed method outperforms other disentanglement methods and generative networks in the absence of PET modality. In the inference stage, with missing PET images, using the siamese semi-disentanglement network proposed in this paper can achieve comparable results to the full modal segmentation. • A novelty disentanglement strategy for robust PET-CT segmentation is proposed. • Semi-disentanglement images into shared tumor features instead of anatomical features. • The GAN can better retain tumor information through semi-disentanglement. • The performance of proposed is better than disentanglement and GAN methods. [ABSTRACT FROM AUTHOR]

Cancer is considered one of the leading causes of death. We can detect cancers through the anatomical and functional imaging provided by PET/CT. However, many tumors in PET/CT are obvious in only one modality, and PET contains many non-lesional hypermetabolic regions, which increases the difficulty of segmentation. Furthermore, traditional two-stage segmentation improves segmentation efficiency by breaking down a segmentation task into two independent subtasks. The second stage loses most of the feature information obtained in the first stage. To address these problems, we propose a hard segmentation network guided by soft segmentation for tumor segmentation on PET/CT images. The proposed network has a soft segmentation branch and a hard segmentation branch. The output of the soft segmentation branch is a logits map composed of gradient values, which is corrected with the soft ground truth by the proposed similarity loss function so that the logits map and the soft ground truth are approximately consistent in the high-dimensional vector space. The output of the hard segmentation branch is the final prediction map. The two branches are connected by a soft segmentation-guided mechanism. This guidance mechanism can generate a soft segmentation-guided map with stable distribution according to the logits map obtained by the soft segmentation branch. We validated the proposed network on two datasets. The Dice of 0.7324 on the public soft tissue sarcoma dataset and 0.7693 on the private liver tumor dataset. By only using U-Net as the backbone network, our method achieves the best performance. [ABSTRACT FROM AUTHOR]

Segmenting the liver and tumor regions using CT scans is crucial for the subsequent treatment in clinical practice and radiotherapy. Recently, liver and tumor segmentation techniques based on U-Net have gained popularity. However, there are numerous varieties of liver tumors, and they differ greatly in terms of their shapes and textures. It is unreasonable to regard all liver tumors as one class for learning. Meanwhile, texture information is crucial for the identification of liver tumors. We propose a plug-and-play Texture-based Auto Pseudo Label (TAPL) module to take use of the texture information of tumors and enable the neural network actively learn the texture differences between various tumors to increase the segmentation accuracy, especially for small tumors. The TPAL module consists of two parts, texture enhancement and texture-based pseudo label generator. To highlight the regions where the texture varies significantly, we enhance the textured areas of the CT image. Based on their texture information, tumors are automatically divided into several classes by the texture-based pseudo label generator. The multi-class tumors produced by the neural network during the prediction step are combined into a single tumor label, which is then used as the outcome of the segmentation. Experiments on clinical dataset and public dataset Lits2017 show that the proposed algorithm outperforms single liver tumor label segmentation methods and is more friendly to small tumors. • A TAPL module is to solve the small liver tumor segmentation problem. • The TAPL module can be added to any U-Net based networks. • The TAPL module force the network to learn the texture differences between tumors. [ABSTRACT FROM AUTHOR]

Tumor segmentation is a key step in computer-aided diagnosis. The PET–CT co-segmentation method combines the high sensitivity of PET images and the anatomical information of CT images. For whole-body multiple tumors, such as soft tissue sarcoma, lymphoma, etc., due to the different lesion location and size, it is necessary to segment the tumor area according to the whole body anatomical information. How to effectively leverage whole-body contextual information and the fusion of multimodal information is the key to the problem. To address this issue, we propose a spatial squeeze and multimodal feature fusion attention network for whole-body multiple tumors segmentation based on PET–CT volumes. Our proposed method consists of two parts, a Coronal-Spatial Squeeze Attention Extraction Network (CSAE-Net) and a Precise PET–CT Fusion Attention Segmentation Network (PFAS-Net), respectively. In CSAE-Net, we squeeze a 3D PET–CT volume along the coronal plane into m 2D images, and obtain 3D Coronal Spatial Squeeze Attention Volume based on these 2D images. In PFAS-Net, the input is a 2D axial PET–CT slice, and the previously obtained coronal spatial squeeze attention map is used to guide the segmentation. Moreover, a Multimodal Fusion Attention (MFA) module is proposed to fuse the metabolic information of PET and the anatomical information of CT. We perform experiments on PET–CT datasets of two whole-body multiple tumors, Soft Tissue Sarcoma (STS) and Lymphoma. The results show that our proposed method improved Dice values by 8.03% in STS and 1.74% in Lymphoma. Also the visualization results show that our proposed method is able to suppress high-uptake regions of normal tissues. [ABSTRACT FROM AUTHOR]

Tumor segmentation is a necessary step in clinical processing that can help doctors diagnose tumors and plan surgical treatments. Since tumors are usually small, the locations and appearances vary substantially across individuals, and the contrast between tumors and adjacent normal tissues is low, tumor segmentation is still a challenging task. Although convolutional neural networks (CNNs) have achieved good results in tumor segmentation, the information about tumor boundaries has been rarely explored. To solve the problem, this paper proposes a new method for automatic tumor segmentation in PET/CT images based on context-coordination and boundary-aware, termed as C 2 BA-UNet. We employ a UNet-like backbone network and replace the encoder with EfficientNet-B0 for efficiency. To acquire potential tumor boundaries, we propose a new multi-atlas boundary-aware (MABA) module based on gradient atlas, uncertainty atlas, and level set atlas, that focuses on uncertain regions between tumors and adjacent tissues. Furthermore, we propose a new context coordination module (CCM) to combine multi-scale context information with attention mechanism to optimize skip connection in high-level layers. To validate the superiority of our method, we conduct experiments on a publicly available soft tissue sarcoma (STS) dataset and a lymphoma dataset, and the results show our method is competitive with other comparison methods. • A boundary-aware module learns uncertain regions between tumors and adjacent tissues. • A context coordination module enhances high-level features and tumor predictions. • C 2 BA-UNet optimizes tumor context information and segmentations. [ABSTRACT FROM AUTHOR]

• MdWRKY41 inhibits the expression of key genes MdMYB12, MdANR and MdUFGT in anthocyanin and procyanidin biosynthesis. • MdWRKY41 can form a repressor with MdMYB16. • MdWRKY41 is a downstream target of MdHY5, and negatively regulates anthocyanin and proanthocyanidin accumulation. Red-fleshed apple fruits are popular because of their high flavonoid content. Although MdMYB10 and its homologs have been identified as crucial regulators of the fruit coloring process, other transcription factors (TFs) contributing to the differences in flesh coloration have not been fully characterized. In this study, we investigated the regulatory effects of MdWRKY41 on anthocyanin and proanthocyanidin (PA) synthesis in red-fleshed apples. The overexpression of MdWRKY41 in red-fleshed apple calli inhibited anthocyanin and PA accumulation by downregulating the expression of a MYB TF gene (MdMYB12) and specific structural genes (MdLAR , MdUFGT , and MdANR). Furthermore, MdWRKY41 was shown to interact with MdMYB16 to form a complex that can further suppress MdANR and MdUFGT expression. Interestingly, MdWRKY41 was targeted by the photoresponse factor MdHY5 and inhibited its transcription. Overall, our findings provide insights into a novel MdHY5-MdWRKY41-MdMYB regulatory module influencing anthocyanin and PA synthesis in red-fleshed apple fruits. [ABSTRACT FROM AUTHOR]

Anthocyanins in apple species are important secondary metabolites that are beneficial for human health. Previous studies revealed that methyl jasmonate (MeJA) promotes anthocyanin accumulation by up-regulating the transcription of related genes. In this study, we isolated a jasmonate (JA)-induced apple MYB gene, MdMYB24-like (MdMYB24L). The encoded nuclear protein contains a conserved R2R3 domain and is homologous to Arabidopsis thaliana AtMYB24. Additionally, MdMYB24L was observed to interact with JA signaling factors (MdJAZ8, MdJAZ11, and MdMYC2) in yeast and in planta. The MdMYC2 protein was also targeted by MdJAZ8 and MdJAZ11, which are rapidly degraded under MeJA treatment. The overexpression of MdMYB24L resulted in higher anthocyanin contents in the transgenic apple 'Orin' calli than in the wild-type control calli. Moreover, the expression levels of the anthocyanin biosynthesis structural genes MdUFGT and MdDFR were up-regulated in the transgenic calli. Furthermore, MdMYB24L positively regulated the transcription of MdDFR and MdUFGT by binding to the MYB-binding site motifs in their promoters. Interestingly, the interaction between MdMYC2 and MdMYB24L further enhanced the transcription of MdUFGT , whereas MdJAZ8 and MdJAZ11 attenuated this effect. We herein provide new details regarding the molecular mechanism by which MYB transcription factors help regulate anthocyanin biosynthesis via JA signaling pathways. • A R2R3 MYB transcription factor, MdMYB24L, is homologous to AtMYB24. • MdMYB24L is nuclear-localized and positively regulated by methyl jasmonate. • MdMYB24L interact with JA signaling factors. • MdMYB24L overexpression promoted anthocyanin accumulation in apple. [ABSTRACT FROM AUTHOR]

Salt stress is an important environmental factor affecting the growth and production of agricultural crops and fruits worldwide, including apple (Malus × domestica). In this study, we demonstrate that a salt-responsive MYB transcription factor (TF), designated as MdMYB63, promotes survival under salt stress. Overexpression of MdMYB63 in apple calli significantly enhanced salt tolerance. Screening of the AP2/ERF family of TFs identified MdERF106 as an interaction partner of MdMYB63. Further analyses showed that the MdMYB63–MdERF106 complex significantly promotes the expression of downstream MdSOS1 , thereby improving the Na+ expulsion and salt tolerance of apple. These functional analyses of MdMYB63 have provided valuable insights into the regulatory network of salt tolerance, and lay a theoretical foundation for the cultivation of new salt-tolerant apple varieties. • Functional analyses of MdMYB63 provide valuable insights into the salt tolerance. • Overexpression of MdMYB63 in apple calli promotes its survival under salt stress. • MdMYB63 binds to MBS on the promoter of MdSOS1 to enhance its transcription. • MdERF106 can significantly promote the effect of MdMYB63 on the expression of downstream MdSOS1 by forming a protein complex. • MYBs can directly regulate SOS1 to participate in the salt tolerance of plants. [ABSTRACT FROM AUTHOR]

• A novel apple MYB-like protein, MdMYBDL1, is highly similar to AtMYBD. • MdMYBDL1 is a downstream target of MdHY5 and promotes anthocyanin accumulation in response to light. • Both MdHY5 and MdMYBDL1 inhibit the expression of MdMYB16/308 , which encode anthocyanin inhibitors. Light is an important environmental factor affecting plant growth and development. Additionally, HY5 is a central factor that coordinates light signal transduction and regulates the expression of flower color-related genes. However, there are few reports describing the co-regulation of apple fruit coloration by MdHY5 and MYB transcription factors. In this study, we detected a light-inducible gene, MdMYBDL1 , which encodes a MYB-like domain and is homologous to AtMYBD in Arabidopsis thaliana. Moreover, we observed that MdHY5 binds to the G-box element of the MdMYBDL1 promoter to upregulate expression. The overexpression of MdMYBDL1 enhanced anthocyanin accumulation in apple calli and inhibited the expression of MdMYB16 and its homolog, MdMYB308. Furthermore, MdMYB16 can form a dimer with MdMYB308 and functions as a negative regulator of anthocyanin biosynthesis. Interestingly, MdMYB16 and MdMYB308 promoter activities were inhibited by MdMYBDL1 and MdHY5. These findings imply that MdHY5 responds to light signals and functions upstream of different types of MYB transcription factors, ultimately regulating anthocyanin accumulation in apples. [ABSTRACT FROM AUTHOR]

• In this paper, the pathway of WRKY transcription factor in regulating apple anthocyanin was described. • This article focuses on the mechanism of Ultraviolet-induced MdWRKY72 promoting anthocyanin synthesis in apples. • This enriches the genetic resources of anthocyanin synthesis and provides a theoretical basis. Ultraviolet-B (UV-B) radiation promotes anthocyanin synthesis in many plants. Although several transcription factors promote anthocyanin synthesis in response to UV-B radiation, the underlying mechanism remains unclear. In this study, the MdWRKY72 transcription factor gene was isolated from the 'Taishanzaoxia' apple genome. Quantitative real-time PCR analyses revealed that the genes encoding enzymes and transcription factors involved in the anthocyanin synthesis pathway (MdANS , MdDFR , MdUFGT , and MdMYB1) were more highly expressed in MdWRKY72 -overexpressing transgenic calli than in the wild-type 'Orin' apple calli. The results indicated that MdWRKY72 increases anthocyanin synthesis in transgenic calli exposed to UV-B radiation. The results of a gel shift assay and chromatin immunoprecipitation proved that MdWRKY72 promotes MdMYB1 expression indirectly by binding to a W-box element in the MdHY5 promoter and directly by binding to a W-box element in the MdMYB1 promoter. Thus, MdWRKY72 increases anthocyanin synthesis via direct and indirect mechanisms. These findings may be useful for elucidating the molecular mechanism underlying UV-B-induced anthocyanin synthesis mediated by MdWRKY72. [ABSTRACT FROM AUTHOR]