Adrenal Volume Quantitative Visualization Tool by Multiple Parameters and an nnU-Net Deep Learning Automatic Segmentation Model.

Abnormalities in adrenal gland size may be associated with various diseases. Monitoring the volume of adrenal gland can provide a quantitative imaging indicator for such conditions as adrenal hyperplasia, adrenal adenoma, and adrenal cortical adenocarcinoma. However, current adrenal gland segmentation models have notable limitations in sample selection and imaging parameters, particularly the need for more training on low-dose imaging parameters, which limits the generalization ability of the models, restricting their widespread application in routine clinical practice. We developed a fully automated adrenal gland volume quantification and visualization tool based on the no new U-Net (nnU-Net) for the automatic segmentation of deep learning models to address these issues. We established this tool by using a large dataset with multiple parameters, machine types, radiation doses, slice thicknesses, scanning modes, phases, and adrenal gland morphologies to achieve high accuracy and broad adaptability. The tool can meet clinical needs such as screening, monitoring, and preoperative visualization assistance for adrenal gland diseases. Experimental results demonstrate that our model achieves an overall dice coefficient of 0.88 on all images and 0.87 on low-dose CT scans. Compared to other deep learning models and nnU-Net model tools, our model exhibits higher accuracy and broader adaptability in adrenal gland segmentation.

Deep learning for colorectal cancer detection in contrast-enhanced CT without bowel preparation: a retrospective, multicentre study.

BACKGROUND: Contrast-enhanced CT scans provide a means to detect unsuspected colorectal cancer. However, colorectal cancers in contrast-enhanced CT without bowel preparation may elude detection by radiologists. We aimed to develop a deep learning (DL) model for accurate detection of colorectal cancer, and evaluate whether it could improve the detection performance of radiologists. METHODS: We developed a DL model using a manually annotated dataset (1196 cancer vs 1034 normal). The DL model was tested using an internal test set (98 vs 115), two external test sets (202 vs 265 in 1, and 252 vs 481 in 2), and a real-world test set (53 vs 1524). We compared the detection performance of the DL model with radiologists, and evaluated its capacity to enhance radiologists' detection performance. FINDINGS: In the four test sets, the DL model had the area under the receiver operating characteristic curves (AUCs) ranging between 0.957 and 0.994. In both the internal test set and external test set 1, the DL model yielded higher accuracy than that of radiologists (97.2% vs 86.0%, p < 0.0001; 94.9% vs 85.3%, p < 0.0001), and significantly improved the accuracy of radiologists (93.4% vs 86.0%, p < 0.0001; 93.6% vs 85.3%, p < 0.0001). In the real-world test set, the DL model delivered sensitivity comparable to that of radiologists who had been informed about clinical indications for most cancer cases (94.3% vs 96.2%, p > 0.99), and it detected 2 cases that had been missed by radiologists. INTERPRETATION: The developed DL model can accurately detect colorectal cancer and improve radiologists' detection performance, showing its potential as an effective computer-aided detection tool. FUNDING: This study was supported by National Science Fund for Distinguished Young Scholars of China (No. 81925023); Regional Innovation and Development Joint Fund of National Natural Science Foundation of China (No. U22A20345); National Natural Science Foundation of China (No. 82072090 and No. 82371954); Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application (No. 2022B1212010011); High-level Hospital Construction Project (No. DFJHBF202105).

Preoperative CT-based radiomics and deep learning model for predicting risk stratification of gastric gastrointestinal stromal tumors.

BACKGROUND: Gastrointestinal stromal tumors (GISTs) are clinically heterogeneous with various malignant potential in different individuals. It is crucial to explore a reliable method for preoperative risk stratification of gastric GISTs noninvasively. PURPOSE: To establish and evaluate a machine learning model using the combination of computed tomography (CT) morphology, radiomics, and deep learning features to predict the risk stratification of primary gastric GISTs preoperatively. METHODS: The 193 gastric GISTs lesions were randomly divided into training set, validation set, and test set in a ratio of 6:2:2. The qualitative and quantitative CT morphological features were assessed by two radiologists. The tumors were segmented manually, and then radiomic features were extracted using PyRadiomics and the deep learning features were extracted using pre-trained Resnet50 from arterial phase and venous phase CT images, respectively. Pearson correlation analysis and recursive feature elimination were used for feature selection. Support vector machines were employed to build a classifier for predicting the risk stratification of GISTs. This study compared the performance of models using different pre-trained convolutional neural networks (CNNs) to extract deep features for classification, as well as the performance of modeling features from single-phase and dual-phase images. The arterial phase, venous phase and dual-phase machine learning models were built, respectively, and the morphological features were added to the dual-phase machine learning model to construct a combined model. Receiver operating characteristic (ROC) curves were used to evaluate the efficacy of each model. The clinical application value of the combined model was determined through the decision curve analysis (DCA) and the net reclassification index (NRI) was analyzed. RESULTS: The area under the curve (AUC) of the dual-phase machine learning model was 0.876, which was higher than that of the arterial phase model or venous phase model (0.813, 0.838, respectively). The combined model had best predictive performance than the above models with an AUC of 0.941 (95% CI: 0.887-0.974) (p = 0.012, Delong test). DCA demonstrated that the combined model had good clinical application value with an NRI of 0.575 (95% CI: 0.357-0.891). CONCLUSION: In this study, we established a combined model that incorporated dual-phase morphology, radiomics, and deep learning characteristics, which can be used to predict the preoperative risk stratification of gastric GISTs.

Optimal debulking surgery in ovarian cancer patients: MRI may predict the necessity of rectosigmoid resection.

OBJECTIVES: To determine whether MRI can predict the necessity of rectosigmoid resection (RR) for optimal debulking surgery (ODS) in ovarian cancer (OC) patients and to compare the predictive accuracy of pre- and post-neoadjuvant chemotherapy (NACT) MRI. METHODS: The MRI of 82 OC were retrospectively analyzed, including six bowel signs (length, transverse axis, thickness, circumference, muscularis involvement, and submucosal edema) and four para-intestinal signs (vaginal, parametrial, ureteral, and sacro-recto-genital septum involvement). The parameters reflecting the degree of muscularis involvement were measured. Patients were divided into non-RR and RR groups based on the operation and postoperative outcomes. The independent predictors of the need for RR were identified by multivariate logistic regression analysis. RESULTS: Imaging for 82 patients was evaluated (67 without and 15 with NACT). Submucosal edema and muscularis involvement (OR 13.33 and 8.40, respectively) were independent predictors of the need for RR, with sensitivities of 83.3% and 94.4% and specificities of 93.9% and 81.6%, respectively. Among the parameters reflecting the degree of muscularis involvement, circumference ≥ 3/12 had the highest prediction accuracy, increasing the specificity from 81.6% for muscularis involvement only to 98.0%, with only a slight decrease in sensitivity (from 94.4% to 88.9%). The predictive sensitivities of pre-NACT and post-NACT MRI were 100.0% and 12.5%, respectively, and the specificities were 85.7% and 100.0%, respectively. CONCLUSIONS: MRI analysis of rectosigmoid muscularis involvement and its circumference can help predict the necessity of RR in OC patients, and pre-NACT MRI may be more suitable for evaluation. CRITICAL RELEVANCE STATEMENT: We analyzed preoperative pelvic MRI in OC patients. Our findings suggest that MRI has predictive potential for identifying patients who require RR to achieve ODS. KEY POINTS: The need for RR must be determined to optimize treatment for OC patients. Muscularis involvement circumference ≥ 3/12 could help predict RR. Pre-NACT MRI may be superior to post-NACT MRI in predicting RR.

Renal fat deposition measured on dixon-based MRI is significantly associated with early kidney damage in obesity.

PURPOSE: To investigate the renal fat deposition on Dixon-based magnetic resonance imaging (MRI) and to explore the predictive value of renal fat biomarkers of magnetic resonance (MR-RFBs) for early kidney damage in obesity. METHODS: This prospective study included 56 obese volunteers and 47 non-obese healthy volunteers. All volunteers underwent renal magnetic resonance examinations. The differences in MR-RFBs [including renal proton density fat fraction (PDFF), renal sinus fat volume (RSFV), and perirenal fat thickness (PRFT)] measured on Dixon-based MRI between the obese and non-obese volunteers were analyzed using a general linear model, taking sex, age, diabetes, and hypertension as covariates. The relationship between estimated glomerular filtration rate (eGFR) and demographic, laboratory, and imaging parameters in obese volunteers was examined by correlation analysis. RESULTS: Obese volunteers had higher MR-RFBs than non-obese volunteers after controlling for confounders (all p < 0.001). Renal PDFF (r = - 0.383; p = 0.004), RSFV (r = - 0.368; p = 0.005), and PRFT (r = - 0.451; p < 0.001) were significantly negatively correlated with eGFR in obesity. After adjusting for age, sex, body mass index, diabetes, hypertension, visceral adipose tissue, subcutaneous adipose tissue, renal PDFF, and RSFV, PRFT remained independently negatively associated with eGFR (ß = - 0.587; p = 0.003). CONCLUSIONS: All MR-RFBs are negatively correlated with eGFR in obesity. The MR-RFBs, especially PRFT, may have predictive value for early kidney damage in obesity.

Measurements of peri-prostatic adipose tissue by MRI predict bone metastasis in patients with newly diagnosed prostate cancer.

Objectives: To investigate the role of MRI measurements of peri-prostatic adipose tissue (PPAT) in predicting bone metastasis (BM) in patients with newly diagnosed prostate cancer (PCa). Methods: We performed a retrospective study on 156 patients newly diagnosed with PCa by prostate biopsy between October 2010 and November 2022. Clinicopathologic characteristics were collected. Measurements including PPAT volume and prostate volume were calculated by MRI, and the normalized PPAT (PPAT volume/prostate volume) was computed. Independent predictors of BM were determined by univariate and multivariate logistic regression analysis, and a new nomogram was developed based on the predictors. Receiver operating characteristic (ROC) curves were used to estimate predictive performance. Results: PPAT and normalized PPAT were associated with BM (P<0.001). Normalized PPAT positively correlated with clinical T stage(cT), clinical N stage(cN), and Grading Groups(P<0.05). The results of ROC curves indicated that PPAT and normalized PPAT had promising predictive value for BM with the AUC of 0.684 and 0.775 respectively. Univariate and multivariate analysis revealed that high normalized PPAT, cN, and alkaline phosphatase(ALP) were independently predictors of BM. The nomogram was developed and the concordance index(C-index) was 0.856. Conclusions: Normalized PPAT is an independent predictor for BM among with cN, and ALP. Normalized PPAT may help predict BM in patients with newly diagnosed prostate cancer, thus providing adjunctive information for BM risk stratification and bone scan selection.

Genomic insights into local adaptation of upland cotton in China and Pakistan.

KEY MESSAGE: Different kinship and resistance to cotton leaf curl disease (CLCuD) and heat were found between upland cotton cultivars from China and Pakistan. 175 SNPs and 82 InDels loci related to yield, fiber quality, CLCuD, and heat resistance were identified. Elite alleles found in Pakistani accessions aided local adaptation to climatic condition of two countries. Adaptation of upland cotton (Gossypium hirsutum) beyond its center of origin is expected to be driven by tailoring of the genome and genes to enhance yield and quality in new ecological niches. Here, resequencing of 456 upland cotton accessions revealed two distinct kinships according to the associated country. Fiber quality and lint percentage were consistent across kinships, but resistance to cotton leaf curl disease (CLCuD) and heat was distinctly exhibited by accessions from Pakistan, illustrating highly local adaption. A total of 175 SNP and 82 InDel loci related to yield, fiber quality, CLCuD and heat resistance were identified; among them, only two overlapped between Pakistani and Chinese accessions underscoring the divergent domestication and improvement targets in each country. Loci associated with resistance alleles to leaf curl disease and high temperature were largely found in Pakistani accessions to counter these stresses prevalent in Pakistan. These results revealed that breeding activities led to the accumulation of unique alleles and helped upland cotton become adapted to the respective climatic conditions, which will contribute to elucidating the genetic mechanisms that underlie resilience traits and help develop climate-resilient cotton cultivars for use worldwide.

A computed tomography-based multitask deep learning model for predicting tumour stroma ratio and treatment outcomes in patients with colorectal cancer: a multicentre cohort study.

BACKGROUND: Tumour-stroma interactions, as indicated by tumour-stroma ratio (TSR), offer valuable prognostic stratification information. Current histological assessment of TSR is limited by tissue accessibility and spatial heterogeneity. The authors aimed to develop a multitask deep learning (MDL) model to noninvasively predict TSR and prognosis in colorectal cancer (CRC). MATERIALS AND METHODS: In this retrospective study including 2268 patients with resected CRC recruited from four centres, the authors developed an MDL model using preoperative computed tomography (CT) images for the simultaneous prediction of TSR and overall survival. Patients in the training cohort ( n =956) and internal validation cohort (IVC, n =240) were randomly selected from centre I. Patients in the external validation cohort 1 (EVC1, n =509), EVC2 ( n =203), and EVC3 ( n =360) were recruited from other three centres. Model performance was evaluated with respect to discrimination and calibration. Furthermore, the authors evaluated whether the model could predict the benefit from adjuvant chemotherapy. RESULTS: The MDL model demonstrated strong TSR discrimination, yielding areas under the receiver operating curves (AUCs) of 0.855 (95% CI, 0.800-0.910), 0.838 (95% CI, 0.802-0.874), and 0.857 (95% CI, 0.804-0.909) in the three validation cohorts, respectively. The MDL model was also able to predict overall survival and disease-free survival across all cohorts. In multivariable Cox analysis, the MDL score (MDLS) remained an independent prognostic factor after adjusting for clinicopathological variables (all P <0.05). For stage II and stage III disease, patients with a high MDLS benefited from adjuvant chemotherapy [hazard ratio (HR) 0.391 (95% CI, 0.230-0.666), P =0.0003; HR=0.467 (95% CI, 0.331-0.659), P <0.0001, respectively], whereas those with a low MDLS did not. CONCLUSION: The multitask DL model based on preoperative CT images effectively predicted TSR status and survival in CRC patients, offering valuable guidance for personalized treatment. Prospective studies are needed to confirm its potential to select patients who might benefit from chemotherapy.

Comparative severe dermatologic toxicities of immune checkpoint inhibitors in malignant melanoma: A systematic review and network meta-analysis.

BACKGROUND: Immune checkpoint inhibitors (ICIs) have advanced the therapeutic landscape for malignant melanoma patients. However, they can cause permanent and irreversible dermatologic immune-related adverse events (irAEs) that may lead to interruption of ICI treatment or become life-threatening. To assess the risk of severe dermatologic irAEs (grade 3 or higher) among ICIs for advanced melanoma, we conducted a network meta-analysis (NMA). METHODS: Phase II/III randomized controlled clinical trials (RCTs) involving ICIs were retrieved from various databases, including PubMed, Embase, Cochrane Library, and Web of Science. These trials were published from the inception of databases to October 15, 2022. In addition, the risk of severe dermatologic irAEs associated with ICI types and doses was evaluated and compared by NMA. RESULTS: This study included 20 Phase II/III RCTs with a total of 10 575 patients. The results indicated that ICIs carry a higher risk of severe dermatologic irAEs compared to chemotherapy. Additionally, the combinational therapy of Nivolumab + Ipilimumab was associated with a higher risk than ICI monotherapy. Comparatively, the latest treatment option involving dual ICI therapy with Relatlimab + Nivolumab showed a lower toxicity risk, but higher than Ipilimumab alone. Lastly, Nivolumab, at a dose of 3 mg/kg every 2 weeks, was observed as the lowest-risk dosing regimen for severe dermatologic irAEs in patients with advanced melanoma. CONCLUSION: The findings suggest that Nivolumab (1 mg/kg) + Ipilimumab (3 mg/kg) administered every 3 weeks should be used cautiously in patients with advanced melanoma at high risk for dermatologic irAEs. While we recommend the preferred regimen of Nivolumab (dose = 3 mg/kg, every 2 weeks).

Ni-Catalyzed 1,1- and 1,3-Aminoboration of Unactivated Alkenes.

Alkene borylfunctionalization reactions have emerged as useful methods for chemical synthesis. While much progress has been made on 1,2-borylamination reactions, the related 1,1- and 1,3-borylaminations have not been reported. Herein, a Ni-catalyzed 1,1-borylamination of 1,1-disubstituted and monosubstituted alkenes and a 1,3-borylamination of cyclic alkenes are presented. Key to development of these reactions was the identification of an alkyllithium activator in combination with Mg salts. The utility of the products and the mechanistic details are discussed.

Design and Synthesis of Bicyclo[4.3.0]nonene Nucleoside Analogues.

Nucleoside analogues are effective antiviral agents, and the continuous emergence of pathogenic viruses demands the development of novel and structurally diverse analogues. Here, we present the design and synthesis of novel nucleoside analogues with a carbobicyclic core, which mimics the conformation of natural ribonucleosides. Employing a divergent synthetic route featuring an intermolecular Diels-Alder reaction, we successfully synthesized carbobicyclic nucleoside analogues with high antiviral efficacy against respiratory syncytial virus.

Outcomes of different parenchymal-sparing hepatectomies in patients with colorectal liver metastases and prognostic impact of peritumoral imaging features.

OBJECTIVES: Parenchymal-sparing hepatectomy (PSH) is recommended in patients with colorectal liver metastases (CRLM). Based on the principle of PSH, to investigate the impact of anatomical resection (AR) and non-anatomic resection (NAR) on the outcome of CRLM and to evaluate the potential prognostic impact of three peritumoral imaging features. METHODS: Fifty-six patients who had abdominal gadoxetic acid-enhanced magnetic resonance imaging (MRI) before CRLM surgery were included in this retrospective research. Peritumoral early enhancement, peritumoral hypointensity on hepatobiliary phase (HBP), and biliary dilatation to the CRLM at MRI were evaluated. Survival estimates were calculated using the Kaplan-Meier method, and multivariate analysis was conducted to identify independent predictors of liver recurrence-free survival (LRFS), recurrence-free survival (RFS) and overall survival (OS). RESULTS: NAR had a lower 3-year LRFS compared with AR (36.6% vs. 78.6%, p = 0.012). No significant differences were found in 3-year RFS (34.1% vs. 41.7%) and OS (61.7% vs. 81.3%) (p > 0.05). In NAR group, peritumoral early enhancement was associated with poor LRFS (p = < 0.001, hazard ratio [HR] = 6.260; 95% confidence interval [CI], 2.322,16.876]) and poor RFS (p = 0.035, HR =2.516; 95% CI, 1.069,5.919). No independent predictors of CRLM were identified in the AR group. CONCLUSIONS: In patients with CRLM, peritumoral early enhancement was a predictor of LRFS and RFS after NAR according to the principle of PSH.

Paeoniflorin inhibits colorectal cancer cell stemness through the miR-3194-5p/catenin beta-interacting protein 1 axis.

Paeoniflorin (PF) is a natural plant ingredient with remarkable antitumor effects. Herein, we investigated the biological effects and mechanism of PF in colorectal cancer (CRC) cell stemness. The messenger RNA (mRNA) and protein expressions were assessed using quantitative real-time polymerase chain reaction and western blot. The viability, proliferation, and migration and invasion of CRC cells were evaluated using cell counting kit-8, clone-formation, and transwell migration and invasion assays, respectively. The sphere-formation capacity was determined using the sphere-formation assay. A dual-luciferase reporter gene assay was employed to analyze the interaction between miR-3194-5p and catenin beta-interacting protein 1 (CTNNBIP1). The viability, migration, invasion, epithelial-mesenchymal transition, and stemness of CRC cells were repressed by PF. MiR-3194-5p was upregulated in CRC tissues and cells. MiR-3194-5p knockdown suppressed CRC cell stemness, while miR-3194-5p overexpression had the opposite effect. In addition, the inhibition of CRC cell stemness caused by PF was eliminated by miR-3194-5p overexpression. CTNNBIP1 functioned as the target of miR-3194-5p, whose knockdown abrogated the repression of CRC cell stemness and Wnt/ß-catenin signaling activation by PF.PF regulated the miR-3194-5p/CTNNBIP1/Wnt/ß-catenin axis to repress CRC cell stemness.

First report of Leaf Blight of Liquidambar formosana Hance caused by Alternaria tenuissima in China.

Liquidambar formosana Hance is widely planted in urban landscapes in China owing to its ornamental red leaves. In June 2020, a distinctive leaf spot disease was observed on L. formosana in Nanjing Forestry University, Jiangsu Province of China (32°4'49"N, 118°48'56"E). Approximately 61% (14 out of 23) of the trees displayed leaf spots. The diseased symptoms included irregularly distributed spots that showed black or dark brown, and occasionally with pale green halo. Two representative trees were selected for sampling and five leaves with typical symptoms were selected randomly for isolation. The tissues from the margin of the lesions (0.2 cm × 0.2 cm) were cut and disinfected in 1% sodium hypochlorite for 90 s, rinsed in sterile water twice for 30 s, and dried with sterile paper. Then, 20 tissues were incubated on 2% potato dextrose agar (PDA) supplemented with 100 mg/L Ampicillin Sodium and incubated in the dark at 25℃ for 4 days. Seventeen single-spore fungi were isolated from lesion tissues as described by Woudenberg et al. (2013). The colony morphology of 17 isolates was extremely similar, so 3 isolates (NFUA01, NFUA02, and NFUA03) were selected randomly for further study. Colonies on PDA were circular, gray, and slightly raised loose cotton mycelium, while the reverse side was olive green in the center with white margins. Conidiophores were brown, simple or branched, and produced numerous conidia in short chains. Conidia were obclavate or ellipsoid, brown, with 1-5 transverse septa and 0-3 longitudinal septa, and measured 7.1 to 32.5 × 3.3 to 13.3 µm (n=50). The morphological observations were consistent with the description of the genus Alternaria sp. (Woudenberg et al. 2013). Six gene fragments, including SSU, LSU, ITS, GAPDH, RPB2 and EF-1 region, were amplified and sequenced. The primers of six nuclear loci were used by NS1 / NS4((White et al. 1990), LSU1Fd (Crous et al. 2009)/ LR5 (Vilgalys & Hester 1990), V9G (De Hoog & Gerrits van den Ende 1998)/ ITS4 (White et al. 1990), gpd1 / gpd2 (Berbee et al. 1999), RPB2-5F2 / fRPB2-7cR (Liu et al. 1999), and EF1-728F / EF1-986R (Carbone & Kohn 1999). The sequences were submitted in GenBank (SSU, ON237470 to ON237472; LSU, ON237464 to ON237466; ITS, ON197354 to ON197356; GAPDH, ON237476 to ON237478; RPB2, ON237467 to ON237469; EF-1, ON237473 to ON237475). BLAST result showed that SSU, LSU, ITS, GAPDH, RPB2, and EF-1 sequences of NFUA01, NFUA02, and NFUA03 were identical to A. tenuissima at a high level (>99%, Table 1). A maximum likelihood and Bayesian posterior probability analysis were performed by IQtree v. 1.6.8 and Mr. Bayes v. 3.2.6 with the concatenated sequences (Guindon et al. 2010; Ronquist et al. 2012). The representative strains which selected for Phylogenetic analyses were chosen from the strains which mentioned by Woudenberg et al (2013) and obtained the sequences from NCBI. The concatenated sequences placed NFUA01, NFUA02 and NFUA03 in the clade of Alternaria tenuissima with a high confidence level (ML/BI= 100/1). A pathogenicity assay was done using isolate NFUA01 on 3-year-old L. formosana seedlings. L. formosana leaves were wounded by a sterilized needle (0.5-mm-diam), and inoculated with spore suspension (106 conidia/mL), and L. formosana leaves inoculated with sterile water were used as the control. Each treatment had 5 leaves, and incubated at 25℃ under high moisture conditions. The experiments were conducted three times. Seven days after inoculation, leaves inoculated with spore suspension showed brown leaf blights resembling the original disease symptoms, whereas the control remained healthy. The fungus was reisolated from the lesions and was confirmed as A. tenuissima based on morphologically characteristics and ITS sequence analysis. To our knowledge, this is the first report of A. tenuissima associated with leaf blight on L. formosana. The finding provides clear pathogen information for further evaluation of the disease control strategies.

Desmoplastic Reaction Associates with Prognosis and Adjuvant Chemotherapy Response in Colorectal Cancer: A Multicenter Retrospective Study.

Desmoplastic reaction (DR) is one of many tumor-host interactions and is associated with the overall survival (OS) of patients with colorectal cancer. However, the clinical significance of DR requires further study in large multicenter cohorts and its predictive value in adjuvant chemotherapy (ACT) response remains unclear. Here, a total of 2,225 patients with colorectal cancer from five independent institutions were divided into primary (N = 1,012 from two centers) and validation (N = 1,213 from three centers) cohorts. DR was classified as immature, middle, or mature depending on the presence of myxoid stroma and hyalinized collagen bundles at the invasive front of the primary tumor. OS among different subgroups were compared, and the correlations of DR type with tumor-infiltrating lymphocytes (TILs) within stroma, tumor stroma ratio (TSR), and Stroma AReactive Invasion Front Areas (SARIFA) were also analyzed. In the primary cohort, patients with mature DR had the highest 5-year survival rate. These findings were confirmed in validation cohort. In addition, for stage II colorectal cancer, patients classified as non-mature DR would benefit from ACT compared with surgery alone. Furthermore, immature and middle DR were more associated with high TSR, less distribution of TILs within stroma and positive SARIFA compared with mature. Taken together, these data suggest that DR is a robust-independent prognostic factor for patients with colorectal cancer. For patients with stage II colorectal cancer, non-mature DR could be a potential marker for recognizing high-risk patients who may benefit from ACT. Significance: DR has the potential to identify patients with high-risk colorectal cancer and predict the efficacy of adjuvant chemotherapy in patients with stage II colorectal cancer. Our findings support reporting DR types as additional pathologic parameters in clinical practice for more precise risk stratification.

Neural Network-Based Prediction for Secret Key Rate of Underwater Continuous-Variable Quantum Key Distribution through a Seawater Channel.

Continuous-variable quantum key distribution (CVQKD) plays an important role in quantum communications, because of its compatible setup for optical implementation with low cost. For this paper, we considered a neural network approach to predicting the secret key rate of CVQKD with discrete modulation (DM) through an underwater channel. A long-short-term-memory-(LSTM)-based neural network (NN) model was employed, in order to demonstrate performance improvement when taking into account the secret key rate. The numerical simulations showed that the lower bound of the secret key rate could be achieved for a finite-size analysis, where the LSTM-based neural network (NN) was much better than that of the backward-propagation-(BP)-based neural network (NN). This approach helped to realize the fast derivation of the secret key rate of CVQKD through an underwater channel, indicating that it can be used for improving performance in practical quantum communications.

Species of the Colletotrichum spp., the Causal Agents of Leaf Spot on European Hornbeam (Carpinus betulus).

European hornbeam (Carpinus betulus L.) is widely planted in landscaping. In October 2021 and August 2022, leaf spot was observed on C. betulus in Xuzhou, Jiangsu Province, China. To identify the causal agent of anthracnose disease on C. betulus, 23 isolates were obtained from the symptomatic leaves. Based on ITS sequences and colony morphology, these isolates were divided into four Colletotrichum groups. Koch's postulates of four Colletotrichum species showed similar symptoms observed in the field. Combining the morphological characteristics and multi-gene phylogenetic analysis of the concatenated sequences of the internal transcribed spacer (ITS) gene, Apn2-Mat1-2 intergenic spacer (ApMat) gene, the calmodulin (CAL) gene, glyceraldehyde3-phosphate dehydrogenase (GAPDH) gene, Glutamine synthetase (GS) gene, and beta-tubulin 2 (TUB2) genes, the four Colletotrichum groups were identified as C. gloeosporioides, C. fructicola, C. aenigma, and C. siamense. This study is the first report of four Colletotrichum species causing leaf spot on European hornbeam in China, and it provides clear pathogen information for the further evaluation of the disease control strategies.

Characterization of a novel tetravalent botulism antitoxin based on receptor-binding domain of BoNTs.

Botulinum neurotoxin (BoNTs; serotypes A, B, E, and F) cause botulism disease in humans, which could be effectively treated using antitoxins. Herein, we established a novel receptor-binding domain (RBD)-based antitoxin using recombinant C terminal heavy chain (Hc) domains of BoNTs as immunogens. Immunization of horses with these recombinant Hc domains allowed the purification and digestion of IgGs from hyper-immune sera to produce high-quality and high-efficiency monovalent botulism antitoxin F(ab')2 against each BoNT (M-BATs). However, these M-BATs could not bind or neutralize other serotypes of BoNTs, and that there were no cross-protective effects among these M-BATs. This suggested the need to prepare tetravalent antitoxins to neutralize the four BoNTs simultaneously. Thus, these M-BATs were formulated into a novel tetravalent botulism antitoxin (T-BAT), in which a 10-ml volume contained 10000 IU of BoNT/A and 5000 IU of BoNT/B, BoNT/E, and BoNT/F antitoxins. The novel antitoxin preparation could prevent and treat the four mixed botulinum neurotoxins simultaneously in vivo, representing strong efficacy in an animal poisoning model. Moreover, these antibodies in T-BAT could bind the RBD, whereas conventional antitoxins based on inactivated toxins mainly bind the light chain or heavy chain translocation domain (HN) and weakly bind the important RBD in current experimental conditions. The high levels of RBD-specific novel antitoxins can efficiently bind the RBD and neutralize natural or recombinant toxins containing this RBD. The findings of the present study experimentally support the use of RBD-specific antitoxins to treat BoNT serotype A, B, E, and F-mediated botulism. This study demonstrated the concept of developing potent novel multivalent antitoxins against all BoNTs or other toxins, using the RBD of these toxins as an alternative antigen to inactivated toxins. KEY POINTS: • Antitoxins based on the receptor-binding domains of botulinum neurotoxins were made. • Novel antitoxin binds RBD; traditional antitoxin mainly binds light chain or HN domain. • A tetravalent antitoxin could prevent and treat the four mixed neurotoxins in vivo.

An in-memory computing architecture based on a duplex two-dimensional material structure for in situ machine learning.

The growing computational demand in artificial intelligence calls for hardware solutions that are capable of in situ machine learning, where both training and inference are performed by edge computation. This not only requires extremely energy-efficient architecture (such as in-memory computing) but also memory hardware with tunable properties to simultaneously meet the demand for training and inference. Here we report a duplex device structure based on a ferroelectric field-effect transistor and an atomically thin MoS2 channel, and realize a universal in-memory computing architecture for in situ learning. By exploiting the tunability of the ferroelectric energy landscape, the duplex building block demonstrates an overall excellent performance in endurance (>1013), retention (>10 years), speed (4.8 ns) and energy consumption (22.7 fJ bit-1 µm-2). We implemented a hardware neural network using arrays of two-transistors-one-duplex ferroelectric field-effect transistor cells and achieved 99.86% accuracy in a nonlinear localization task with in situ trained weights. Simulations show that the proposed device architecture could achieve the same level of performance as a graphics processing unit under notably improved energy efficiency. Our device core can be combined with silicon circuitry through three-dimensional heterogeneous integration to give a hardware solution towards general edge intelligence.

Cotton heterosis and hybrid cultivar development.

Cotton, the most important economic crop in the world, displays strong hybrid vigor, and has long been a subject for hybrid cultivar breeding. Here, advances in the theoretical and applied research in cotton heterosis along with its hybrid cultivar development by hand-emasculation and pollination (HEP), cytoplasmic (CMS) and genic male sterile lines (GMS) mainly in China during the past few decades are presented in this review. Three types of hybrids produced by HEP, CMS and GMS facilitating hybrid seed production with hand-pollination have been developed and are being planted simultaneously in cotton production. However, most hybrids commercially planted in production are produced by HEP, therefore, F2 seeds are being extensively planted due to the high cost to produce F1 seed. F2 generations of these combinations exceed the check cultivars in yield usually up to 5~15%. GMS genes (ms2 and ms5ms6) used in hybrid seed production and casual mitochondrial genes for G. harknessii CMS have been cloned. Challenges and opportunities in cotton heterosis and future hybrid cultivar development in cotton are discussed.