irGSEA: the integration of single-cell rank-based gene set enrichment analysis.

irGSEA is an R package designed to assess the outcomes of various gene set scoring methods when applied to single-cell RNA sequencing data. This package incorporates six distinct scoring methods that rely on the expression ranks of genes, emphasizing relative expression levels over absolute values. The implemented methods include AUCell, UCell, singscore, ssGSEA, JASMINE and Viper. Previous studies have demonstrated the robustness of these methods to variations in dataset size and composition, generating enrichment scores based solely on the relative gene expression of individual cells. By employing the robust rank aggregation algorithm, irGSEA amalgamates results from all six methods to ascertain the statistical significance of target gene sets across diverse scoring methods. The package prioritizes user-friendliness, allowing direct input of expression matrices or seamless interaction with Seurat objects. Furthermore, it facilitates a comprehensive visualization of results. The irGSEA package and its accompanying documentation are accessible on GitHub (https://github.com/chuiqin/irGSEA).

Ligand-Nondestructive Direct Photolithography Assisted by Semiconductor Polymer Cross-Linking for High-Resolution Quantum Dot Light-Emitting Diodes.

The photolithographic patterning of fine quantum dot (QD) films is of great significance for the construction of QD optoelectronic device arrays. However, the photolithography methods reported so far either introduce insulating photoresist or manipulate the surface ligands of QDs, each of which has negative effects on device performance. Here, we report a direct photolithography strategy without photoresist and without engineering the QD surface ligands. Through cross-linking of the surrounding semiconductor polymer, QDs are spatially confined to the network frame of the polymer to form high-quality patterns. More importantly, the wrapped polymer incidentally regulates the energy levels of the emitting layer, which is conducive to improving the hole injection capacity while weakening the electron injection level, to achieve balanced injection of carriers. The patterned QD light-emitting diodes (with a pixel size of 1.5 µm) achieve a high external quantum efficiency of 16.25% and a brightness of >1.4 × 105 cd/m2. This work paves the way for efficient high-resolution QD light-emitting devices.

Abnormal phosphorylation / dephosphorylation and Ca2+ dysfunction in heart failure.

Heart failure (HF) can be caused by a variety of causes characterized by abnormal myocardial systole and diastole. Ca2+ current through the L-type calcium channel (LTCC) on the membrane is the initial trigger signal for a cardiac cycle. Declined systole and diastole in HF are associated with dysfunction of myocardial Ca2+ function. This disorder can be correlated with unbalanced levels of phosphorylation / dephosphorylation of LTCC, endoplasmic reticulum (ER), and myofilament. Kinase and phosphatase activity changes along with HF progress, resulting in phased changes in the degree of phosphorylation / dephosphorylation. It is important to realize the phosphorylation / dephosphorylation differences between a normal and a failing heart. This review focuses on phosphorylation / dephosphorylation changes in the progression of HF and summarizes the effects of phosphorylation / dephosphorylation of LTCC, ER function, and myofilament function in normal conditions and HF based on previous experiments and clinical research. Also, we summarize current therapeutic methods based on abnormal phosphorylation / dephosphorylation and clarify potential therapeutic directions.

Discrete wavelet transform assisted convolutional neural network equalizer for PAM VLC system.

With the deepening of research and the further differentiation of damage types, and to compensate for both linear and nonlinear damage in visible light communication systems (VLCs), we propose a novel discrete wavelet transform-assisted convolutional neural network (DWTCNN) equalizer that combines the advantages of wavelet transform and deep learning methods. More specifically, wavelet transform is used in DWTCNN to decompose the signal into diverse coefficient series and employ an adaptive soft-threshold method to eliminate redundant information in the signal. The coefficients are then reconstructed to achieve complete signal compensation. The experimental results show that the proposed DWTCNN equalizer can significantly reduce nonlinear impairment and improve system performance with the bit error rate (BER) under the 7% hard-decision forward error correction (HD-FEC) limit of 3.8 × 10-3. We also experimentally compared DWTCNN with the Long Short-Term Memory (LSTM) and entity extraction neural network (EXNN) equalizer, the Q factor has been improved by 0.76 and 0.53 dB, and the operating ranges of the direct current (DC) bias have increased by 4.76% and 23.5%, respectively.

A crazy trio in Parkinson's disease: metabolism alteration, α-synuclein aggregation, and oxidative stress.

Parkinson's disease (PD) is an aging-associated neurodegenerative disorder, characterized by the progressive loss of dopaminergic neurons in the pars compacta of the substantia nigra and the presence of Lewy bodies containing α-synuclein within these neurons. Oligomeric α-synuclein exerts neurotoxic effects through mitochondrial dysfunction, glial cell inflammatory response, lysosomal dysfunction and so on. α-synuclein aggregation, often accompanied by oxidative stress, is generally considered to be a key factor in PD pathology. At present, emerging evidences suggest that metabolism alteration is closely associated with α-synuclein aggregation and PD progression, and improvement of key molecules in metabolism might be potentially beneficial in PD treatment. In this review, we highlight the tripartite relationship among metabolic changes, α-synuclein aggregation, and oxidative stress in PD, and offer updated insights into the treatments of PD, aiming to deepen our understanding of PD pathogenesis and explore new therapeutic strategies for the disease.

An Adaptive Sampling Algorithm with Dynamic Iterative Probability Adjustment Incorporating Positional Information.

Physics-informed neural networks (PINNs) have garnered widespread use for solving a variety of complex partial differential equations (PDEs). Nevertheless, when addressing certain specific problem types, traditional sampling algorithms still reveal deficiencies in efficiency and precision. In response, this paper builds upon the progress of adaptive sampling techniques, addressing the inadequacy of existing algorithms to fully leverage the spatial location information of sample points, and introduces an innovative adaptive sampling method. This approach incorporates the Dual Inverse Distance Weighting (DIDW) algorithm, embedding the spatial characteristics of sampling points within the probability sampling process. Furthermore, it introduces reward factors derived from reinforcement learning principles to dynamically refine the probability sampling formula. This strategy more effectively captures the essential characteristics of PDEs with each iteration. We utilize sparsely connected networks and have adjusted the sampling process, which has proven to effectively reduce the training time. In numerical experiments on fluid mechanics problems, such as the two-dimensional Burgers' equation with sharp solutions, pipe flow, flow around a circular cylinder, lid-driven cavity flow, and Kovasznay flow, our proposed adaptive sampling algorithm markedly enhances accuracy over conventional PINN methods, validating the algorithm's efficacy.

Urban development and resource endowments shape natural resource utilization efficiency in Chinese cities.

Improving the efficiency with which natural resources are utilized is an indispensable for achieving sustainable development and carbon neutrality. By analyzing the utilization efficiency of energy, water, and land, we established a comprehensive natural resource utilization efficiency index (NRUEI). We then calculated the 2016 index for 165 cities in China, and investigated what caused it to vary. We found three main results: (1) the NRUEI varied greatly across China and there were significant positive correlations between urban energy utilization efficiency (EUE), water utilization efficiency (WUE) and land utilization efficiency (LUE); (2) the NRUEI showed a significant positive correlation with a city's population size, economy (Gross Domestic Product (GDP)), and the level of urban social development (GDP per capita); (3) cities in East China have the highest natural resource utilization efficiency, while cities in Northeast China have the lowest. These results indicate that China's increasing urban development is associated with rising natural resource utilization efficiency and that the city's endowment of natural resources is an important factor affecting that efficiency. Further, the results showed that the determinants of a city's NRUEI differed in large and small cities. Lastly, our results suggest that improving EUE is key for improving NRUEI in urban China, and different efficiencies can be improved intertwined. A major takeaway of this study is that there is great potential for improving natural resource utilization efficiency in Chinese cities and we include city-specific suggestions for efficiency improvements.

Utilizing deep neural networks to extract non-linearity as entities in PAM visible light communication with noise.

Herein, we propose a novel entity extraction neural network (EXNN) with a newly designed sampling convolution kernel approach and a deep learning-based structure to differentiate noise in visible light communication (VLC) systems. In particular, EXNN is used to extract linear and nonlinear distortion in a received signal as an entity and compensate for the signal by removing it. First, we designed a deep learning structure tailored for VLC systems, used experimentation to validate our algorithm's usability, and determined an appropriate range for the hyper-parameters that govern the PAM-8 system. Second, we compared our approach with existing finite impulse response (FIR) linear and Volterra nonlinear compensation algorithms via experiments. Below the hard-decision forward error correction (HD-FEC) threshold limit of 3.8×10-3, experimental results show that the use of the EXNN increased the operating range of the direct current (DC) bias and the voltage by ∼33.3% and by ∼50% under optimal conditions, respectively. Furthermore, under corresponding optimal power conditions, the proposed approach improved the Q factor of the VLC system by 0.36 and 1.57 dB compared with the aforementioned linear and nonlinear equalizers, respectively. To the best of our knowledge, this is the first time that a deep learning operator has been custom-designed for the VLC system and we have named the completely redesigned network with this sampling convolution kernel operator as EXNN.

A Pan-Cancer Analysis of the BIRC Gene Family and Its Association with Prognosis, Tumor Microenvironment, and Therapeutic Targets.

The inhibitors of apoptosis protein (IAP)/baculoviral IAP repeat containing (BIRC) gene families are necessary for cell protection, and most of these genes act as endogenous inhibitors of apoptosis. In some cancers, the over-expression of the BIRC gene is associated with cancer progression, multidrug resistance, poor prognosis and short-term survival. In this study, we aimed to assess the effect of the BIRC family in pan-cancer. We downloaded transcriptome and clinical data from 33 types of TCGA tumor samples and adjacent tissues. Then, the expression characteristics of IAP family members BIRC2, BIRC3, BIRC5, BIRC6 and BIRC7 in pan-cancer were analyzed. R packet and Cox regression were used to analyze the clinical correlation. In addition, the transcription level of BIRC and immune subtypes, stem cells, immune tumor microenvironment (TME) and drug sensitivity were analyzed by multidimensional correlation. Our studies have shown that the expression of IAP family members BIRC2, BIRC3, BIRC5, BIRC6, and BIRC7 is different in different tumor types, and the heterogeneity is obvious in cancers. Overall, our analysis showed that BIRC2, BIRC3, BIRC6, and BIRC7 were mainly down-regulated in tumors, whereas BIRC5 was mainly up-regulated in tumors. The expression of IAP family members is related to the overall survival of patients. However, the direction of the association varies depending on specific IAP subtypes and specific types of cancer. More specifically, BIRC5 is mainly related to poor prognosis. The rest of the IAP family showed either a survival advantage or a survival disadvantage, depending on the type of cancer. In addition, BIRC2, BIRC3, BIRC5, BIRC6 and BIRC7 were significantly correlated with immune infiltration subtypes and had different degrees of correlation with the degree of interstitial cell infiltration and tumor cell dryness. Finally, our study revealed that BIRC2, BIRC5, and BIRC7 genes may be related to drug resistance of tumor cells. Our systematic analysis of (IAP) gene expression and its relationship with immune infiltration, TME, cancer stem cells, drug sensitivity and prognosis of cancer patients highlights the need to study IAP family members as separate entities in each specific cancer type. In addition, our study confirmed that IAP family genes are promising therapeutic targets for cancer and potential prognostic indicators for clinical application, although further laboratory verification is needed.

Laboratory indicators predict axillary nodal pathologic complete response after neoadjuvant therapy in breast cancer.

The purpose was to integrate clinicopathological and laboratory indicators to predict axillary nodal pathologic complete response (apCR) after neoadjuvant therapy (NAT). The pretreatment clinicopathological and laboratory indicators of 416 clinical nodal-positive breast cancer patients who underwent surgery after NAT were analyzed from April 2015 to 2020. Predictive factors of apCR were examined by logistic analysis. A nomogram was built according to logistic analysis. Among the 416 patients, 37.3% achieved apCR. Multivariate analysis showed that age, pathological grading, molecular subtype and neutrophil-to-lymphocyte ratio were independent predictors of apCR. A nomogram was established based on these four factors. The area under the curve (AUC) was 0.758 in the training set. The validation set showed good discrimination, with AUC of 0.732. In subtype analysis, apCR was 23.8, 47.1 and 50.8% in hormone receptor-positive/HER2-, HER2+ and triple-negative subgroups, respectively. According to the results of the multivariate analysis, pathological grade and fibrinogen level were independent predictors of apCR after NAT in HER2+ patients. Except for traditional clinicopathological factors, laboratory indicators could also be identified as predictive factors of apCR after NAT. The nomogram integrating pretreatment indicators demonstrated its distinguishing capability, with a high AUC, and could help to guide individualized treatment options.

Lay abstract The purpose of this study was to integrate more pretreatment indicators, including clinicopathological factors and simple laboratory indicators, to predict axillary nodal pathologic complete response (apCR) after neoadjuvant therapy for breast cancer. The authors collected the pretreatment clinicopathological factors and laboratory indexes of 416 nodal-positive patients with breast cancer. The authors then built a nomogram to predict the therapeutic effect in axillary lymph nodes. Among 416 patients, 37.3% (155 of 416) achieved apCR. The results showed that age, pathological grading, molecular subtype and neutrophil-to-lymphocyte ratio were independent predictors of apCR. Based on these four factors, a nomogram was then built. This nomogram helped to predict apCR. In addition to traditional clinicopathological factors, laboratory indicators were also identified as predictive factors of apCR after neoadjuvant therapy. Integrating pretreatment indicators might help to predict apCR and guide individualized treatment options.

Efficacy and safety analysis of paclitaxel, docetaxel and liposomal paclitaxel after neoadjuvant therapy in breast cancer.

BACKGROUND AND PURPOSE: Paclitaxel-based regimens are widely used in the neoadjuvant therapy (NAT) of breast cancer. The purpose is to analysis the efficacy and adverse events (AEs) among common paclitaxel (PTX), docetaxel and liposomal paclitaxel. At the same time, we want to analysis the axillary nodal pathologic complete response (apCR) after NAT among the three groups. METHODS: From April 2014 to 2020, 647 breast cancer patients underwent operation after NAT were included in this study. Patients received full course of anthracycline- and paclitaxel-based chemotherapy before surgery. The paclitaxel-based regimens included PTX, docetaxel and liposomal paclitaxel. The therapy efficacy and AEs of the three groups were evaluated. At the same time, the apCR was also analyzed. RESULTS: In general, 30.6% (198/647) of patients achieved breast pathologic complete response (bpCR), which was 28.6%, 28.3% and 39.3% among PTX, docetaxel and liposomal paclitaxel group, respectively (p = 0.067). The total pathologic complete response (tpCR) (achieving both bpCR and apCR) was 21.6% (140/647). The tpCR was 13.3%, 19.4% and 34.4% among PTX, docetaxel and liposomal paclitaxel group, respectively (p = 0.026). The multivariate logistic analysis result showed that clinical tumor stage and molecular subtype were significantly associated with tpCR (all p < 0.05). Among 592 clinical positive patients (cN+), the apCR was 39.0% (231/592). The multivariate logistic analysis showed that paclitaxel- based regimens and molecular subtype were indicated as independent predictors for apCR of NAT. The apCR was significantly higher in liposomal paclitaxel group (63.5%) than in PTX (24.6%) and docetaxel group (34.8%) (p < 0.001). The subgroup analysis among different molecular subtypes found that in triple negative (TN) and HER-2 positive (HER2+) subgroup, the apCR in liposomal paclitaxel group was significantly higher than those in PTX and docetaxel group (all p < 0.05). But no significant result was found in the subgroup analysis in hormone receptor positive/HER-2 negative subgroup (p = 0.050). Safety analysis indicated that the incidence of neutropenia (grade III-IV) and peripheral neurotoxicity (grade I-II) was significantly lower in the liposomal paclitaxel group than in the PTX and docetaxel group. The incidence of oral mucositis, anaphylaxis and palmar-plantar erythrodysesthesia syndrome was also much lower in liposomal paclitaxel than other two groups (all p < 0.05). And there was no significant difference in other AEs among the three groups (all p > 0.05). CONCLUSION: Liposome paclitaxel had similar tumor suppressor effect compared with PTX and docetaxel in NAT setting. Moreover, it had a better axillary lymph node (ALN) response after NAT than PTX and docetaxel. These patients who received liposome paclitaxel had more chance to avoid ALN dissection after NAT. At the same time, the application of liposome enables liposome paclitaxel could significantly reduce AEs caused by chemotherapy. Therefore, we suggested the application of liposome paclitaxel in the NAT setting, especially for cN+ patients with TN and HER2 + disease.

Internal Mammary Sentinel Lymph Node Biopsy in Clinically Axillary Lymph Node-Positive Breast Cancer: Diagnosis and Implications for Patient Management.

BACKGROUND: Routine performance of internal mammary sentinel lymph node biopsy (IM-SLNB) remains a subject of debate due to no clinical relevance in breast cancer, because it was performed only in clinically axillary lymph node (ALN)-negative patients. In this study, IM-SLNB was performed in clinically ALN-positive patients, and its impact on nodal staging and therapeutic strategy were subsequently analyzed. METHODS: Clinically ALN-positive patients who underwent IM-SLNB were enrolled in this prospective study. Statistical analysis was performed using Chi square test, Mann-Whitney U and logistic regression models with a significance level of 0.05. RESULTS: Among the 352 recruited patients, the internal mammary sentinel lymph node (IMSLN) visualization rate of patients who received initial surgery and neoadjuvant systemic therapy (NST) was 71.9% (123/171) and 33.1% (60/181), respectively. The 183 patients who underwent IM-SLNB successfully had the average time duration of 7 min and the median IMSLN number of 2. There were 87 positive IMSLNs in all the 347 removed IMSLNs, which were mainly concentrated in the second (50.6%) and third (34.5%) intercostal space. The IMSLN metastasis rate was 39.8% (initial surgery) and 13.3% (NST), respectively. All of the 183 IM-SLNB patients received more accurate nodal staging, 57 of whom had stage elevated, which might have prompted modifications to the therapeutic strategy. CONCLUSIONS: IM-SLNB should be routinely performed in clinically ALN-positive patients, and thus more accurate nodal staging and perfect pathologic complete response definition could be put forward. The identification of IMLN metastases by IM-SLNB might potentially influence therapeutic strategies.

Correction to: Internal Mammary Sentinel Lymph Node Biopsy in Clinically Axillary Lymph Node-Positive Breast Cancer: Diagnosis and Implications for Patient Management.

The authors' affiliations are corrected as reflected here.

Edge Computing and Blockchain for Quick Fake News Detection in IoV.

The dissemination of false messages in Internet of Vehicles (IoV) has a negative impact on road safety and traffic efficiency. Therefore, it is critical to quickly detect fake news considering news timeliness in IoV. We propose a network computing framework Quick Fake News Detection (QcFND) in this paper, which exploits the technologies from Software-Defined Networking (SDN), edge computing, blockchain, and Bayesian networks. QcFND consists of two tiers: edge and vehicles. The edge is composed of Software-Defined Road Side Units (SDRSUs), which is extended from traditional Road Side Units (RSUs) and hosts virtual machines such as SDN controllers and blockchain servers. The SDN controllers help to implement the load balancing on IoV. The blockchain servers accommodate the reports submitted by vehicles and calculate the probability of the presence of a traffic event, providing time-sensitive services to the passing vehicles. Specifically, we exploit Bayesian Network to infer whether to trust the received traffic reports. We test the performance of QcFND with three platforms, i.e., Veins, Hyperledger Fabric, and Netica. Extensive simulations and experiments show that QcFND achieves good performance compared with other solutions.

Neo-adjuvant chemotherapy and axillary de-escalation management for patients with clinically node-negative breast cancer.

This study aimed to explore the optimal time of sentinel lymph node biopsy (SLNB) and neo-adjuvant chemotherapy (NAC) and to assess the feasibility of selective elimination of axillary surgery after NAC in clinically node-negative (cN0) patients. From April 2010 to August 2018, 845 patients undergoing surgery after NAC were included in this retrospective study to analyze the correlation between different clinicopathological characteristics of cN0 patients and negative axillary lymph node after NAC (ypN0). Among the 148 cN0 patients, 83.1% (123/148) were ypN0. The rates of ypN0 in patients with hormone receptor positive (HR+)/HER2-, HR+/HER2+, HR-/HER2+, and triple-negative (TN) breast cancer were 75.4% (46/61), 82.6% (19/23), 85.2% (23/27), and 94.6% (35/37), respectively (P < 0.001). The rates of ypN0 in TN and HER2+ patients were 94.6% and 95.5%, which were significantly higher than that in HR+/HER2- patients (P < 0.05). Molecular subtypes, clinical stage, radiologic complete response, and pathologic complete response (bpCR) of the breast tumor correlated with ypN0 after full-course NAC (P < 0.05). Molecular subtypes (OR = 2.374, P = 0.033), clinical stage (OR = 0.320, P = 0.029), and bpCR (OR = 0.454, P = 0.012) were independent predictors for ypN0. The optimal time of SLNB and NAC in cN0 patients might be different among different molecular subtypes: it would be preferable to perform SLNB prior to NAC for HR+/HER2- patients, and SLNB after NAC for TN and HER2+ patients to reduce the risk of axillary lymph node dissection. In view of the high ypN0 rate in cN0 patients, axillary surgical staging might be selectively eliminated, especially for HER2+ and TN patients.

[Internal mammary sentinel lymph node biopsy in breast cancer: accurate staging and individualized treatment].

OBJECTIVE: The aim of this study was to determine the impact of routinely performed internal mammary sentinel lymph node biopsy (IM-SLNB) on the staging and treatment, and to analyze the success rate, complications and learning curve. METHODS: All patients with biopsy-proven breast cancer who underwent sentinel lymph node biopsy between 2012 and 2014 were included in a prospective analysis. Internal mammary sentinel lymph node biopsy (IM-SLNB) was performed in all patients with IM-SLN visualized on preoperative lymphoscintigraphy and/or detected by intraoperative gamma probe detection. The adjuvant treatment plan was adjusted according to the current guidelines. RESULTS: In a total of 349 patients, 249 patients (71.1%) showed internal mammary drainage. IM-SLNB was performed in 153 patients with internal mammary drainage, with a success rate of IM-SLNB of 97.4% (149/153). Pleural lesion and internal mammary artery bleeding were found in 7.2% and 5.2% patients, respectively. In 8.1% of patients (12/149) the IM-SLN was tumor positive. In the group of patients who underwent IM-SLNB, lymph node staging was changed in 8.1% of patients, and IMLNs radiotherapy was guided by these results, however, systemic treatment was changed in only 0.7% of the patients. CONCLUSIONS: IM-SLNB has a high successful rate and good safety. Identification of internal mammary metastases through IM-SLNB may provide more accurate staging and guide the tailored internal mammary radiotherapy. TRIAL REGISTRATION: ClinicalTrials. gov, NCT01642511.

[Accuracy of MRI for estimating residual tumor size after neoadjuvant chemotherapy in breast cancer with three-dimensional reconstruction technique].

OBJECTIVE: To evaluate the accuracy of MRI for estimating residual tumor size after neoadjuvant chemotherapy (NAC) with three-dimensional (3D) reconstruction technique. METHODS: This was a prospective study. The data of 61 patients with pathologically proven solitary invasive ductal carcinoma (IIA-IIIC) who had received 6 to 8 cycles of NAC from July 2010 to August 2013 was analyzed. All the patients were female, aging from 31 to 70 years with a median of 49 years. Breast specimen after surgery was prepared with part-mount sub-serial section, and residual tumors were microscopically outlined, scanned and registered by Photoshop software. The 3D model of pathological and MRI residual tumors was reconstructed with 3D-DOCTOR software. The longest diameter, maximum cross-section area and volume of the residual tumors determined using 3D MRI were compared with 3D pathological findings, and the associations between MRI and pathology were analyzed by Spearman rank correlation and Bland-Altman analysis. RESULTS: The longest diameter, maximum cross-section area and volume of the residual tumors after NAC measured by MRI and pathology was highly correlated (r=0.942, 0.941, 0.903, all P=0.00). MRI appears to underestimate pathology in the longest diameter, maximum cross-section area, but slightly overestimate in volume, and two methods had a good consistence (MD=0.3 cm, 95% CI: -1.43 to 1.9 cm; MD=1.39 cm², 95% CI: -9.55 to 12.34 cm²; MD=-0.433 cm³, 95% CI: -7.065 to 6.199 cm³). CONCLUSION: 3D MRI reconstruction after NAC could accurately detects the residual tumors after neoadjuvant chemotherapy, and contribute to select patients who received breast conserving therapy after NAC with tumor downstaging.

Prediction of recurrence risk factors in patients with early-stage cervical cancers by nomogram based on MRI handcrafted radiomics features and deep learning features: a dual-center study.

PURPOSE: To establish and validate a deep learning radiomics nomogram (DLRN) based on intratumoral and peritumoral regions of MR images and clinical characteristics to predict recurrence risk factors in early-stage cervical cancer and to clarify whether DLRN could be applied for risk stratification. METHODS: Two hundred and twenty five pathologically confirmed early-stage cervical cancers were enrolled and made up the training cohort and internal validation cohort, and 40 patients from another center were enrolled into the external validation cohort. On the basis of region of interest (ROI) of intratumoral and different peritumoral regions, two sets of features representing deep learning and handcrafted radiomics features were created using combined images of T2-weighted MRI (T2WI) and diffusion-weighted imaging (DWI). The signature subset with the best discriminant features was chosen, and deep learning and handcrafted signatures were created using logistic regression. Integrated with independent clinical factors, a DLRN was built. The discrimination and calibration of DLNR were applied to assess its therapeutic utility. RESULTS: The DLRN demonstrated satisfactory performance for predicting recurrence risk factors, with AUCs of 0.944 (95% confidence interval 0.896-0.992) and 0.885 (95% confidence interval 0.834-0.937) in the internal and external validation cohorts. Furthermore, decision curve analysis revealed that the DLRN outperformed the clinical model, deep learning signature, and radiomics signature in terms of net benefit. CONCLUSION: A DLRN based on intratumoral and peritumoral regions had the potential to predict and stratify recurrence risk factors for early-stage cervical cancers and enhance the value of individualized precision treatment.

Multi-level feature extraction and reconstruction for 3D MRI image super-resolution.

Magnetic resonance imaging (MRI) is an essential radiology technique in clinical diagnosis, but its spatial resolution may not suffice to meet the growing need for precise diagnosis due to hardware limitations and thicker slice thickness. Therefore, it is crucial to explore suitable methods to increase the resolution of MRI images. Recently, deep learning has yielded many impressive results in MRI image super-resolution (SR) reconstruction. However, current SR networks mainly use convolutions to extract relatively single image features, which may not be optimal for further enhancing the quality of image reconstruction. In this work, we propose a multi-level feature extraction and reconstruction (MFER) method to restore the degraded high-resolution details of MRI images. Specifically, to comprehensively extract different types of features, we design the triple-mixed convolution by leveraging the strengths and uniqueness of different filter operations. For the features of each level, we then apply deconvolutions to upsample them separately at the tail of the network, followed by the feature calibration of spatial and channel attention. Besides, we also use a soft cross-scale residual operation to improve the effectiveness of parameter optimization. Experiments on lesion-free and glioma datasets indicate that our method obtains superior quantitative performance and visual effects when compared with state-of-the-art MRI image SR methods.

Improving water efficiency is more effective in mitigating water stress than water transfer in Chinese cities.

The interactions between human and natural systems and their effects have unforeseen results, particularly in the management of water resources. Using water stress mitigation as an example, a water resources management effect index (WRMEI) was created to quantitatively evaluate the trends of water management effects. This revealed that the WRMEI was decreasing due to the impact of the water resources management process. The findings demonstrate that water resources management has unintended effects: there was a gap between the expectation of water stress to be mitigated and the actual results of water stress increasing. That is caused by human activities in water utilization: (1) increasing available water resources from water transfer was not utilized sparingly in the receiving cities-increased water transfers from external sources increase domestic water consumption per capita; (2) improving water efficiency has a positive effect on mitigating water stress, but the population growth decreased the efficiency. It was concluded that much greater attention needs to be paid to water conservation in residential and living use to counter these unintended water management effects.