An uncertainty-based interpretable deep learning framework for predicting breast cancer outcome.

BACKGROUND: Predicting outcome of breast cancer is important for selecting appropriate treatments and prolonging the survival periods of patients. Recently, different deep learning-based methods have been carefully designed for cancer outcome prediction. However, the application of these methods is still challenged by interpretability. In this study, we proposed a novel multitask deep neural network called UISNet to predict the outcome of breast cancer. The UISNet is able to interpret the importance of features for the prediction model via an uncertainty-based integrated gradients algorithm. UISNet improved the prediction by introducing prior biological pathway knowledge and utilizing patient heterogeneity information. RESULTS: The model was tested in seven public datasets of breast cancer, and showed better performance (average C-index = 0.691) than the state-of-the-art methods (average C-index = 0.650, ranged from 0.619 to 0.677). Importantly, the UISNet identified 20 genes as associated with breast cancer, among which 11 have been proven to be associated with breast cancer by previous studies, and others are novel findings of this study. CONCLUSIONS: Our proposed method is accurate and robust in predicting breast cancer outcomes, and it is an effective way to identify breast cancer-associated genes. The method codes are available at: https://github.com/chh171/UISNet .

Analyzing the topological properties of resting-state brain function network connectivity based on graph theoretical methods in patients with high myopia.

AIM: Recent imaging studies have found significant abnormalities in the brain's functional or structural connectivity among patients with high myopia (HM), indicating a heightened risk of cognitive impairment and other behavioral changes. However, there is a lack of research on the topological characteristics and connectivity changes of the functional networks in HM patients. In this study, we employed graph theoretical analysis to investigate the topological structure and regional connectivity of the brain function network in HM patients. METHODS: We conducted rs-fMRI scans on 82 individuals with HM and 59 healthy controls (HC), ensuring that the two groups were matched for age and education level. Through graph theoretical analysis, we studied the topological structure of whole-brain functional networks among participants, exploring the topological properties and differences between the two groups. RESULTS: In the range of 0.05 to 0.50 of sparsity, both groups demonstrated a small-world architecture of the brain network. Compared to the control group, HM patients showed significantly lower values of normalized clustering coefficient (γ) (P = 0.0101) and small-worldness (σ) (P = 0.0168). Additionally, the HM group showed lower nodal centrality in the right Amygdala (P < 0.001, Bonferroni-corrected). Notably, there is an increase in functional connectivity (FC) between the saliency network (SN) and Sensorimotor Network (SMN) in the HM group, while the strength of FC between the basal ganglia is relatively weaker (P < 0.01). CONCLUSION: HM Patients exhibit reduced small-world characteristics in their brain networks, with significant drops in γ and σ values indicating weakened global interregional information transfer ability. Not only that, the topological properties of the amygdala nodes in HM patients significantly decline, indicating dysfunction within the brain network. In addition, there are abnormalities in the FC between the SN, SMN, and basal ganglia networks in HM patients, which is related to attention regulation, motor impairment, emotions, and cognitive performance. These findings may provide a new mechanism for central pathology in HM patients.

Effects of photoperiod and temperature on the developmental duration and diapause in Dolycoris baccarum (Heteroptera: Pentatomidae) from Hohhot, Inner Mongolia, China.

The shield bug, Dolycoris baccarum (L.) (Heteroptera: Pentatomidae), is widely distributed across Asia and Europe. At high latitudes, it overwinters, as adult in diapause, which then becomes the insect source for the following year. To fully understand the developmental duration and diapause characteristics of D. baccarum, the effects of photoperiod and temperature were studied in a population from Hohhot, Inner Mongolia, China. The results indicated that the developmental duration was significantly prolonged at temperatures of 20 or 25 °C, with a prolonged light period; however, when the light period was prolonged to 16L:8D and 18L:6D, the developmental duration was shortened significantly. Furthermore, the developmental duration was also shortened significantly with increasing temperature, when the photoperiod was 12L:12D for short days and 16L:8D for long days. All individuals entered diapause under short-day conditions of 10L:14D and 12L:12D at a temperature of 20 °C; however, the diapause rate decreased significantly under 14L:10D and 16L:8D photoperiods, and the diapause rate decreased significantly at a temperature of 25 °C with prolonged photoperiod. Interestingly, when the photoperiod was fixed at 12L:12D, the diapause rates at different temperatures (20, 25, 28, and 30 °C) exceeded 95%; while the effect of temperature on diapauses was nonsignificant under this photoperiod, it was still sensitive to the photoperiod; at a photoperiod of 16L:8D, the effect of temperature on the diapause rate was noticeable, and the diapause rate decreased significantly with increasing temperature.

Experimental Research on Shipborne SINS Rapid Mooring Alignment with Variance-Constraint Kalman Filter and GNSS Position Updates.

Analytical coarse alignment and Kalman filter fine alignment based on zero-velocity are typically used to obtain initial attitude for inertial navigation systems (SINS) on a static base. However, in the shipboard mooring state, the static observation condition is corrupted. This paper presents a rapid alignment method for SINS on swaying bases. The proposed method begins with a coarse alignment technique in the inertial frame to obtain an initial rough attitude. Subsequently, a Kalman filter with position updates is employed to estimate the remaining misalignment error. To enhance the filter estimation performance, an appropriate lower boundary is set to the target states' variances according to a carefully designed relative convergence index. The variance-constraint Kalman filter (VCKF) approach is proposed in this paper, and the shipborne experiments validate its effectiveness. The results demonstrate that the VCKF approach significantly reduces the time requirement for fine alignment to achieve the same accuracy on a swaying base, from 90 min in the classic Kalman filter to 30 min. Additionally, the parameter estimation performance in the Kalman filter is also improved, particularly in situations where unpredicted external interference is involved during fine alignment.

Predictive value of metabolism and its heterogeneity parameters measured by preoperative 18 F-FDG PET/CT for mediastinal occult lymph node metastasis in cN0 lung invasive adenocarcinoma.

OBJECTIVE: To explore the potential of intratumoral metabolism and its heterogeneous parameters, as measured by preoperative fluorine-18-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) imaging, to predict mediastinal occult lymph node metastasis in cN0 lung invasive adenocarcinoma. SUBJECTS AND METHODS: Seventy five patients were consecutively enrolled from January 2018 to December 2022. All patients underwent 18F-FDG PET/CT scans within two weeks before surgery, and had mediastinal lymph node metastasis confirmed by pathologic diagnosis after surgery. Metabolic parameters including the maximum standardized uptake value (SUVmax), mean SUV (SUVmean), maximum average SUV (SUVpeak), tumor metabolic volume (MTV), and metabolic heterogeneity (HF) were measured. The relationship between primary focal metabolism, its heterogeneity parameters, and occult mediastinal lymph node metastasis was analyzed using an independent-sample t-test, analysis of covariance, and Mann-Whitney U test. A multivariate logistic regression model was used to analyze independent risk factors for mediastinal lymph node metastasis, while the receiver operating characteristic (ROC) curve assessed the predictive value of metabolic heterogeneity parameters for mediastinal occult lymph node metastasis. RESULTS: A total of 20 out of 75 patients (26.7%) were pathologically confirmed to have mediastinal lymph node metastasis. Analysis of covariance showed that the SUVmax, SUVmean, SUVpeak and MTV were significantly higher in patients with metastasis than in those without (all P<0.05). The metabolic heterogeneity parameters HF2 and HF3 were significantly higher in patients with mediastinal lymph node metastasis than in those without (P=0.013, 0.001), but not HF1. Multivariate Logistic regression analysis identified that tumor size, SUVmax, SUVpeak, lymph node SUVmax, and HF2 of the primary tumor as independent risk factors for mediastinal lymph node metastasis. Metabolic heterogeneity 3 demonstrated high predictive value for mediastinal occult lymph node metastasis (AUC=0.720, P=0.004). CONCLUSION: Metabolism and heterogeneity, as measured by preoperative 18F-FDG PET/CT in lung invasive adenocarcinoma, potentially have clinical value for predicting mediastinal occult lymph node metastasis.

Performance of the Risk Scores for Predicting In-Hospital Mortality in Patients with Acute Coronary Syndrome in a Chinese Cohort.

Background: The prognosis of patients with acute coronary syndrome (ACS) varies greatly, and risk assessment models can help clinicians to identify and manage high-risk patients. While the Global Registry of Acute Coronary Events (GRACE) model is widely used, the clinical pathways for acute coronary syndromes (CPACS), which was constructed based on the Chinese population, and acute coronary treatment and intervention outcomes network (ACTION) have not yet been validated in the Chinese population. Methods: Patients with ACS who underwent coronary angiography or percutaneous coronary intervention from 2011 to 2020, were retrospectively recruited and the appropriate corresponding clinical indicators was obtained. The primary endpoint was in-hospital mortality. The performance of the GRACE, GRACE 2.0, ACTION, thrombolysis in myocardial infarction (TIMI) and CPACS risk models was evaluated and compared. Results: A total of 19,237 patients with ACS were included. Overall, in-hospital mortality was 2.2%. ACTION showed the highest accuracy in predicting discriminated risk (c-index 0.945, 95% confidence interval [CI] 0.922-0.955), but the calibration was not satisfactory. GRACE and GRACE 2.0 did not significantly differ in discrimination (p = 0.1480). GRACE showed the most accurate calibration in all patients and in the subgroup analysis of all models. CPACS (c-index 0.841, 95% CI 0.821-0.861) and TIMI (c-index 0.811, 95% CI 0.787-0.835) did not outperform (c-index 0.926, 95% CI 0.911-0.940). Conclusions: In contemporary Chinese ACS patients, the ACTION risk model's calibration is not satisfactory, although outperformed the gold standard GRACE model in predicting hospital mortality. The CPACS model developed for Chinese patients did not show better predictive performance than the GRACE model.

Assembly of ligation chain reaction and DNA triangular prism for miRNA diagnostics.

Controllable assembly of DNA nanostructure provides a powerful way for quantitative analysis of various targets including nucleic acid molecules. In this study, we have designed detachable DNA nanostructures at electrochemical sensing interface and constructed a ligation chain reaction (LCR) strategy for amplified detection of miRNA. A three-dimensional DNA triangular prism nanostructure is fabricated to provide suitable molecule recognition environment, which can be further regenerated for additional tests via convenient pH adjustment. Target triggered LCR is highly efficient and specific towards target miRNA. Under optimal experimental conditions, this approach enables ultrasensitive exploration in a wide linear range with a single-base resolution. Moreover, it shows excellent performances for the analysis of cell samples and clinical serum samples.

A decentralized federated learning-based cancer survival prediction method with privacy protection.

Background: Survival prediction is one of the crucial goals in precision medicine, as accurate survival assessment can aid physicians in selecting appropriate treatment for individual patients. To achieve this aim, extensive data must be utilized to train the prediction model and prevent overfitting. However, the collection of patient data for disease prediction is challenging due to potential variations in data sources across institutions and concerns regarding privacy and ownership issues in data sharing. To facilitate the integration of cancer data from different institutions without violating privacy laws, we developed a federated learning-based data integration framework called AdFed, which can be used to evaluate patients' survival while considering the privacy protection problem by utilizing the decentralized federated learning technology and regularization method. Results: AdFed was tested on different cancer datasets that contain the patients' information from different institutions. The experimental results show that AdFed using distributed data can achieve better performance in cancer survival prediction (AUC = 0.605) than the compared federated-learning-based methods (average AUC = 0.554). Additionally, to assess the biological interpretability of our method, in the case study we list 10 identified genes related to liver cancer selected by AdFed, among which 5 genes have been proved by literature review. Conclusions: The results indicate that AdFed outperforms better than other federated-learning-based methods, and the interpretable algorithm can select biologically significant genes and pathways while ensuring the confidentiality and integrity of data.

Mechanisms with network pharmacology approach of Ginsenosides in Alzheimer's disease.

Alzheimer's disease (AD) is a progressive neurodegenerative disease characterized by memory loss, cognitive disorder, language dysfunction, and mental disability. The main neuropathological changes in AD mainly include amyloid plaque deposition, neurofibrillary tangles, synapse loss, and neuron reduction. However, the current anti-AD drugs do not demonstrate a favorable effect in altering the pathological course of AD. Moreover, long-term use of these drugs is usually accompanied with various side effects. Ginsenosides are the major active constituents of ginseng and have protective effects on AD through various mechanisms in both in vivo and in vitro studies. In this review, we focused on discussing the therapeutic potential effects and the mechanisms of pharmacological activities of ginsenosides in AD, to provide new insight for further research and clinical application of ginsenosides in the future. Recent studies on the pharmacological effects and mechanisms of ginsenosides were retrieved from Chinese National Knowledge Infrastructure, National Science and Technology Library, Wanfang Data, Elsevier, ScienceDirect, PubMed, SpringerLink, and the Web of Science database up to April 2023 using relevant keywords. Network pharmacology and bioinformatics analysis were used to predict the therapeutic effects and mechanisms of ginsenosides against AD. Ginsenosides presented a wide range of therapeutic and biological activities, including alleviating Aß deposition, decreasing tau hyperphosphorylation, regulating the cholinergic system, resisting oxidative stress, modulating Ca2+ homeostasis, as well as anti-inflammation and anti-apoptosis in neurons, respectively. For further developing the therapeutic potential as well as clinical applications, the network pharmacology approach was combined with a summary of published studies.

High-risk event prone driver identification considering driving behavior temporal covariate shift.

Drivers who perform frequent high-risk events (e.g., hard braking maneuvers) pose a significant threat to traffic safety. Existing studies commonly estimated high-risk event occurrence probabilities based upon the assumption that data collected from different time periods are independent and identically distributed (referred to as i.i.d. assumption). Such approach ignored the issue of driving behavior temporal covariate shift, where the distributions of driving behavior factors vary over time. To fill the gap, this study targets at obtaining time-invariant driving behavior features and establishing their relationships with high-risk event occurrence probability. Specifically, a generalized modeling framework consisting of distribution characterization (DC) and distribution matching (DM) modules was proposed. The DC module split the whole dataset into several segments with the largest distribution gaps, while the DM module identified time-invariant driving behavior features through learning common knowledge among different segments. Then, gated recurrent unit (GRU) was employed to conduct time-invariant driving behavior feature mining for high-risk event occurrence probability estimation. Moreover, modified loss functions were introduced for imbalanced data learning caused by the rarity of high-risk events. The empirical analyses were conducted utilizing online ride-hailing services data. Experiment results showed that the proposed generalized modeling framework provided a 7.2% higher average precision compared to the traditional i.i.d. assumption based approach. The modified loss functions further improved the model performance by 3.8%. Finally, benefits for the driver management program improvement have been explored by a case study, demonstrating a 33.34% enhancement in the identification precision of high-risk event prone drivers.

Uncovering the Pre-Deterioration State during Disease Progression Based on Sample-Specific Causality Network Entropy (SCNE).

Complex diseases do not always follow gradual progressions. Instead, they may experience sudden shifts known as critical states or tipping points, where a marked qualitative change occurs. Detecting such a pivotal transition or pre-deterioration state holds paramount importance due to its association with severe disease deterioration. Nevertheless, the task of pinpointing the pre-deterioration state for complex diseases remains an obstacle, especially in scenarios involving high-dimensional data with limited samples, where conventional statistical methods frequently prove inadequate. In this study, we introduce an innovative quantitative approach termed sample-specific causality network entropy (SCNE), which infers a sample-specific causality network for each individual and effectively quantifies the dynamic alterations in causal relations among molecules, thereby capturing critical points or pre-deterioration states of complex diseases. We substantiated the accuracy and efficacy of our approach via numerical simulations and by examining various real-world datasets, including single-cell data of epithelial cell deterioration (EPCD) in colorectal cancer, influenza infection data, and three different tumor cases from The Cancer Genome Atlas (TCGA) repositories. Compared to other existing six single-sample methods, our proposed approach exhibits superior performance in identifying critical signals or pre-deterioration states. Additionally, the efficacy of computational findings is underscored by analyzing the functionality of signaling biomarkers.

Mycorrhizal mediation of soil carbon in permafrost regions depends on soil nutrient stoichiometry and physical protection.

Mycorrhizal associations are considered as one of the key drivers for soil carbon (C) accumulation and stability. However, how mycorrhizal associations influence soil organic C (SOC) and its fractions (i.e., particulate organic C [POC] and mineral-associated organic C [MAOC]) remain unclear. In this study, we examined effects of plant mycorrhizal associations with arbuscular mycorrhiza (AM), ectomycorrhiza (ECM), and their mixture (Mixed) on SOC and its fractions as well as soil stoichiometric ratios across 800-km transect in permafrost regions. Our results showed that soil with only ECM-associated trees had significantly higher SOC and POC compared to only AM-associated tree species, while soil in Mixed plots with both AM- and ECM- associated trees tend to be somewhat in the middle. Using structural equation models, we found that mycorrhizal association significantly influenced SOC and its fraction (i.e., POC, MAOC) indirectly through soil stoichiometric ratios (C:N, C:P, and N:P). These results suggest that selecting ECM tree species, characterized by a "slow cycling" nutrient uptake strategy, can effectively enhance accumulation of SOC and its fractions in permafrost forest ecosystems. Our findings provide novel insights for quantitatively assessing the influence of mycorrhiza-associated tree species on the management of soil C pool and biogeochemical cycling.

Core decompression combined with bone marrow mononuclear cells in the treatment of femoral head necrosis: a systematic review and meta-analysis.

BACKGROUND: The effectiveness of bone marrow mononuclear cells combined with core decompression in the treatment of femoral head necrosis is controversial. The purpose of this study was to conduct a meta-analysis and systematic review of the evaluation of bone marrow mononuclear cells combined with core decompression in the treatment of femoral head necrosis, and to compare the therapeutic effect of this method with that of core decompression alone, so as to provide a basis for subsequent research and clinical treatment. METHODS: We conducted detailed searches across four databases in Embase, PubMed, Web of Science, and the Cochrane Library (up to October 2023), including eight studies with a total of 370 participants and 491 hip cases. This meta-analysis followed the Preferred Reporting Project (PRISMA) guidelines. Review Manager 5.4 was used to summarize and analyze the outcome indicators and the quality and reliability of the MAs were graded against a Measurement Tool to Assess Systematic Reviews 2 (AMSTAR 2). RESULTS: Eight studies were included inclusion criteria. The results of meta-analysis showed that the therapeutic effect of CD combined with BMMC on VAS was better than that of CD alone (MD =-5.32, 95%CI: -9.90, -0.74, P=0.02, I²=98%), and there was no statistically significant difference between CD combined with BMMC and CD alone in the treatment of HHS (MD =2.73, 95%CI: -2.63,8.09, P=0.32, I²=82%). We conducted sensitivity analysis, the results showed that CD joint BMMC treatment effect on the HHS is superior to the single CD (MD = 5.57, 95% CI: 1.94, 9.20, P = 0.003, I squared = 0%), both no significant differences in VAS (MD = 0.47, 95% CI: -1.74, 0.79, P=0.46, I²=83%). CONCLUSION: In this study, we found that core decompression combined with bone marrow monocyte therapy improved femoral head necrosis better than core decompression alone.

Extracorporeal shock wave therapy with imaging examination for early osteonecrosis of the femoral head: a systematic review.

BACKGROUND: Extracorporeal shockwave therapy (ESWT) is a traditional non-invasive therapy to treat osteonecrosis of the femur head (ONFH). This systematic review aims to investigate whether ESWT can improve the clinical function of ONFH and whether differences in improvement can be observed in radiographic outcomes. MATERIALS AND METHODS: Two authors independently searched PubMed, Embase, Cochrane Library, and Web of Science for English articles until October 21, 2023. After screening and reading the literature, the two authors independently used corresponding scales to evaluate the quality of the included articles and extracted data. The key data extracted included the Harris Hip Score (HHS), Visual Analog Scale (VAS), changes in lesion size, the change in the Association Research Circulation Osseous (ARCO) stage, and bone marrow edema stage. RESULTS: Nine articles included 468 males and 248 females. The average age was 43.29 years and the mean follow-up time was 15.19 months. After receiving ESWT, five studies involving 146 hips showed a higher HHS (MD=-33.38; 95%CI, -46.31, -20.45), and the difference was statistically significant (P<0.00001). The average VAS before treatment was above 5, but it dropped to 1.2 after ESWT (MD=4.64; 95%CI, 3.63, 5.64), and the difference was statistically significant (P<0.00001). Three studies found no significant differences in the areas of femoral head necrosis before and after treatment with ESWT(MD=9.66; 95%CI, -0.36, 19.67; P=0.06; I2=84%). Two articles showed that the use of ESWT had no significant effect on the change in the ARCO stage (MD=1.11; 95%CI, 0.76, 1.62; P=0.60; I2=0%). Three studies indicated that using ESWT could improve the bone marrow edema symptom in the early stage of ONFH (MD=4.35; 95%CI, 1.32, 14.37; P=0.02; I2=62%). CONCLUSION: Based on the current evidence, ESWT shows promise as a therapy to enhance hip function and alleviate pain in the early stage of ONFH. With the advancement of more precise imaging techniques, ESWT can potentially reduce the area affected by ONFH. However, such reduction was not found to be statistically significant at the imaging level. Additionally, ESWT could improve symptoms of bone marrow edema in the early stage. However, no significant change in ARCO grade was observed with ESWT treatment.

Protection efficacy and the safety of the synergy between modified Bazhen powder and PRRSV modified-live virus vaccine against HP-PRRSV in piglets.

The highly pathogenic porcine reproductive and respiratory syndrome virus (HP-PRRSV) poses a significant threat to the global swine industry. Vaccination is a preventive measure against viral infections. However, the use of vaccines in livestock healthcare programs faces the challenge of safety and delayed immune responses. Earlier studies have shown the potential of modified Bazhen powder as an immunomodulator with significant biological properties, but its effect on vaccines against HP-PRRSV is yet to be studied. This study elucidated how modified Bazhen powder could improve the safety and efficacy of the conventional PRRSV vaccine by evaluating T-cell responses, antibody levels, clinical symptoms, levels of viremia, organ health, and cytokine production. The results revealed that the oral application of modified Bazhen powder in combination with PRRS vaccination improved both cellular and humoral immunity, accelerated viremia clearance, improved lung injury scores, and reduced viral load in the tonsils. The modified Bazhen powder also effectively reduced inflammatory responses following a PRRSV challenge. These findings further highlight the pharmacological properties of modified Bazhen powder as a potential oral immunomodulatory agent that could enhance vaccine efficacy and ensure broad-spectrum protection against HP-PRRSV in pigs.

A transition from arbuscular to ectomycorrhizal forests halts soil carbon sequestration during subtropical forest rewilding.

Ecological succession and restoration rapidly promote multiple dimensions of ecosystem functions and mitigate global climate change. However, the factors governing the limited capacity to sequester soil organic carbon (SOC) in old forests are poorly understood. Ecological theory predicts that plants and microorganisms jointly evolve into a more mutualistic relationship to accelerate detritus decomposition and nutrient regeneration in old than young forests, likely explaining the changes in C sinks across forest succession or rewilding. To test this hypothesis, we conducted a field experiment of root-mycorrhizal exclusion in successional subtropical forests to investigate plant-decomposer interactions and their effects on SOC sequestration. Our results showed that SOC accrual rate at the 0-10 cm soil layer was 1.26 mg g-1 yr-1 in early-successional arbuscular mycorrhizal (AM) forests, which was higher than that in the late-successional ectomycorrhizal (EcM) forests with non-significant change. A transition from early-successional AM to late-successional EcM forests increase fungal diversity, especially EcM fungi. In the late-successional forests, the presence of ectomycorrhizal hyphae promotes SOC decomposition and nutrient cycle by increasing soil nitrogen and phosphorus degrading enzyme activity as well as saprotrophic microbial richness. Across early- to late-successional forests, mycorrhizal priming effects on SOC decomposition explain a slow-down in the capacity of older forests to sequester soil C. Our findings suggest that a transition from AM to EcM forests supporting greater C decomposition can halt the capacity of forests to provide nature-based global climate change solutions.

MRI-based deep learning and radiomics for prediction of occult cervical lymph node metastasis and prognosis in early-stage oral and oropharyngeal squamous cell carcinoma: a diagnostic study.

INTRODUCTION: The incidence of occult cervical lymph node metastases (OCLNM) is reported to be 20-30% in early-stage oral cancer and oropharyngeal cancer. There is a lack of an accurate diagnostic method to predict occult lymph node metastasis and to help surgeons make precise treatment decisions. AIM: To construct and evaluate a preoperative diagnostic method to predict OCLNM in early-stage oral and oropharyngeal squamous cell carcinoma (OC and OP SCC) based on deep learning features (DLFs) and radiomics features. METHODS: A total of 319 patients diagnosed with early-stage OC or OP SCC were retrospectively enrolled and divided into training, test and external validation sets. Traditional radiomics features and DLFs were extracted from their MRI images. The least absolute shrinkage and selection operator (LASSO) analysis was employed to identify the most valuable features. Prediction models for OCLNM were developed using radiomics features and DLFs. The effectiveness of the models and their clinical applicability were evaluated using the area under the curve (AUC), decision curve analysis (DCA), and survival analysis. RESULTS: Seventeen prediction models were constructed. The Resnet50 deep learning (DL) model based on the combination of radiomics and DL features achieves the optimal performance, with AUC values of 0.928 (95% CI: 0.881-0.975), 0.878 (95% CI: 0.766-0.990), 0.796 (95% CI: 0.666-0.927), and 0.834 (95% CI: 0.721-0.947) in the training, test, external validation set1, and external validation set2, respectively. Moreover, the Resnet50 model has great prediction value of prognosis in patients with early-stage OC and OP SCC. CONCLUSION: The proposed MRI-based Resnet50 DL model demonstrated high capability in diagnosis of OCLNM and prognosis prediction in the early-stage OC and OP SCC. The Resnet50 model could help refine the clinical diagnosis and treatment of the early-stage OC and OP SCC.

Effects of plant density on alfalfa (Medicago sativa L.) seed yield in western Heilongjiang areas.

Alfalfa (Medicago sativa L.) is known as the "king of forages". The aim of the current study is to determine the optimum planting density as the key cultivation technique for high yield of alfalfa seed. Alfalfa variety (Longmu 801) was planted in experimental fields from 2014 to 2017. In the planting density test, the row spacing was 65, 80, and 95 cm, and the plant spacing was 30, 45, 60, 75, and 90 cm. The seed yield and yield components in the row spacing and plant spacing tests were measured. On the basis of 3 years average of the experimental data, the highest seed yield of 225.49 kg ha-1 was obtained with row spacing vs plant spacing of 65 and 60 cm, respectively. Correlation analysis showed a significant positive correlation between the racemes per stem, pods per raceme, pods per stem, seeds per pod, and the seed yield. These results suggested that Longmu 801 should be cultivated with 65 cm row spacing and 60 cm plant spacing to maximize seed yields in western Heilongjiang areas.