ABSTRACT Multi‐modal diagnosis of ophthalmic disease is becoming increasingly important because combining multi‐modal data allows for more accurate diagnosis. Color fundus photograph (CFP) and optical coherence tomography (OCT) are commonly used as two non‐invasive modalities for ophthalmic examination. However, the diagnosis of each modality is not entirely accurate. Compounding the challenge is the difficulty in acquiring multi‐modal data, with existing datasets frequently lacking paired multi‐modal data. To solve these problems, we propose multi‐modal distribution fusion diagnostic algorithm and cross‐modal generation algorithm. The multi‐modal distribution fusion diagnostic algorithm first calculates the mean and variance separately for each modality, and then generates multi‐modal diagnostic results in a distribution fusion manner. In order to generate the absent modality (mainly OCT data), three sub‐networks are designed in the cross‐modal generation algorithm: cross‐modal alignment network, conditional deformable autoencoder and latent consistency diffusion model (LCDM). Finally, we propose multi‐task collaboration strategy where diagnosis and generation tasks are mutually reinforcing to achieve optimal performance. Experimental results demonstrate that our proposed method yield superior results compared to state‐of‐the‐arts.

The development of deep learning has played an increasingly crucial role in assisting medical diagnoses. Lung cancer, as a major disease threatening human health, benefits significantly from the use of auxiliary medical systems to assist in segmenting pulmonary nodules. This approach effectively enhances both the accuracy and speed of diagnosis for physicians, thereby reducing the risk of patient mortality. However, pulmonary nodules are characterized by irregular shapes and a wide range of diameter variations. They often reside amidst blood vessels and various tissue structures, posing significant challenges in designing an automated system for lung nodule segmentation. To address this, we have developed a three-dimensional dual-branch attention-guided network (DAG-Net) for multi-scale information fusion, aimed at segmenting lung nodules of various types and sizes. First, a dual-branch encoding structure is employed to provide the network with prior knowledge about nodule texture information, which aids the network in better identifying different types of lung nodules. Next, we designed a structure to extract global information, which enhances the network's ability to localize lung nodules of different sizes by fusing information from multiple resolutions. Following that, we fused multi-scale information in a parallel structure and used attention mechanisms to guide the network in suppressing the influence of non-nodule regions. Finally, we employed an attention-based structure to guide the network in achieving more accurate segmentation by progressively using high-level semantic information at each layer. Our proposed network achieved a DSC value of 85.6% on the LUNA16 dataset, outperforming state-of-the-art methods, demonstrating the effectiveness of the network.

OBJECTIVES:To evaluate the therapeutic regimen, efficacy and safety of intrathecal or intraventricular (ITH/IVT) administration of polymyxin B for hospital-acquired central nervous system (CNS) infections caused by carbapenem-resistant Acinetobacter baumannii (CRAB). METHODS:A retrospective study was undertaken of patients with CNS infections caused by CRAB treated with ITH/IVT combination therapy. The primary outcome was the clinical efficacy of treatment. The secondary outcomes were the bacterial clearance rate and the safety of therapy. RESULTS:In total, 35 patients who received ITH [n=13 (37.1%)] or IVT [n=22 (62.9%)] polymyxin B as combination therapy were included in this study. The median duration of ITH/IVT polymyxin B therapy was 9 (interquartile range 7-11) days. The overall clinical cure rate and bacterial clearance rate were 77.1% and 85.7%, respectively. No adverse effects considered to be related to ITH/IVT polymyxin B were recorded. Clinical failure was independently associated with an Acute Physiology and Chronic Health Evaluation II score ≥15 [odds ratio (OR) 1.24, 95% confidence interval (CI) 1.05-1.42; P=0.038] and a Glasgow Coma Scale score ≤8 (OR 0.69, 95% CI 0.49-0.88; P=0.029). Early administration (≤4 days of infection onset) of ITH/IVT polymyxin B therapy resulted in a significantly higher clinical cure rate (OR 0.65, 95% CI 0.49-1.12; P<0.001), and may reduce the length of treatment and adverse effects. CONCLUSIONS:ITH/IVT administration of polymyxin B is a valid alternative for the treatment of CNS infections caused by CRAB. Early use of ITH/IVT polymyxin B can result in greater clinical success.

BACKGROUND:RNA N6-methyladenosine (m6A) is the most common type of modification in eukaryotic mRNA. The relationship between m6A modification and disease has been studied extensively, but there have been few studies on chronic heart failure (CHF). This study investigated a possible role for m6A in the diagnosis of CHF.METHODS:Seven candidate m6A regulators (writers: WTAP and ZC3H13; readers: YTHDF3, FMR1, IGFBP1, and ELAVL1; eraser: FTO) were identified using a random forest (RF) model and the GSE5406 dataset from the Gene Expression Omnibus database. A nomogram model was developed to predict the risk of CHF, while consensus clustering methodology assigned CHF samples into two m6A patterns (cluster A and cluster B) according to the 7 candidate m6A regulators. Principal component analysis was used to calculate an m6A score for each sample and to quantify m6A patterns.RESULTS:Decision curve analysis and the nomogram model were used to obtain predictions that may be of clinical use. Patients in cluster B had higher m6A scores than patients in cluster A. Cluster B patients also had higher expression levels (ELs) of IL-4, IL-5, IL-10 and IL-13 than patients in cluster A, whereas cluster A patients had a higher EL for IL-33. The m6A cluster B pattern likely represents the ischemic heart failure (HF) disease group.CONCLUSION:m6A regulators are important in the pathogenesis of CHF associated with ischemic and idiopathic dilated cardiomyopathy, and may prove useful for the diagnosis and treatment of CHF.

This article aims to propose practical solutions that coordinate the conflicting interests between the global community and the pharmaceutical industry on the intellectual property (IP) waiver for COVID-19 vaccines and facilitate a more equitable vaccine supply chain in the post-COVID-19 world. We critically conducted a narrative literature review to identify procedural and practical issues in the current vaccine supply chain. The search was conducted across various academic disciplines, including biomedical science, life science, law and social science, using resources such as PubMed, Web of Science, Scopus and Westlaw. After screening 731 articles, 55 studies were selected for review. The narrative review revealed several critical barriers that hinder vaccine supply in less-developed countries (LDCs) as follows: (1) WTO Trade-Related Aspects of Intellectual Property Rights (TRIPs) waiver requests may not be granted due to its stringent consensus rule; (2) the current compulsory license system may not work due to the complexity of IP rights covering COVID-19 vaccine technologies; (3) only a few LDCs have domestic companies capable of manufacturing vaccines, and (4) political and economic tensions among countries exacerbate existing barriers to vaccine distribution in LDCs. Based on these findings, we proposed a comprehensive compulsory license system, which combines TRIPS's compulsory license system with the third-party beneficiary mechanism under Common Law. This integrated approach offers a balanced solution that ensures fair compensation for vaccine developers while facilitating broader vaccine access.

Retinal vessels have high curvature and diverse morphology, making them difficult to segment, especially tiny vessels. At present, the retinal vessels are mainly annotated manually by experts, which is difficult to meet the vast clinical needs. To solve the above problems, we propose an effective network M3U-CDVAE. It adopts the architecture of a segmentation-refinement network to denoise and optimizes segmentation results. Firstly, we design a lightweight segmentation network M3U with an encoder-decoder structure. Then, the Hierarchical Feature Fusion (HFF) unit combines the intermediate features generated by the segmentation network with the pre-segmentation results and connects them to the corresponding layer in the next sub-model. Finally, Convolutional Denoising Variational Auto-Encoder (CDVAE) is used as the refinement network to remove the background noise and optimize segmentation results. We conduct exhaustive ablation experiments to demonstrate the improvement brought by our contribution. At the same time, we carry out comparison experiments on DRIVE, STARE, and HRF datasets to illustrate the effectiveness of the proposed method. Experimental results exhibit that the proposed method is superior to most state-of-art methods.

BACKGROUND:Pemetrexed plus platinum chemotherapy is the first-line treatment option for lung adenocarcinoma. However, hematological toxicity is major dose-limiting and even life-threatening. The ability to anticipate hematological toxicity is of great value for identifying potential chemotherapy beneficiaries with minimal toxicity and optimizing treatment. The study aimed to develop and validate a prediction model for hematologic toxicity based on real-world data. METHODS:Data from 1754 lung adenocarcinoma patients with pemetrexed plus platinum chemotherapy regimen as first-line therapy were used to establish and calibrate a risk model for hematological toxicity using multivariate and stepwise logistic regression analysis based on real-world data. The predictive performance of the model was tested in a validation cohort of 753 patients. An area under the curve (AUC) of the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis were used to assess the prediction model. RESULTS:5 independent factors (platinum, pre-use vitamin B12, cycle of chemotherapy before hematological toxicity, Hb before first chemotherapy, and PLT before first chemotherapy) identified from multivariate and stepwise logistic regression analysis were included in the prediction model. The hematological toxicity prediction model achieved a sensitivity of 0.840 and a specificity of 0.822. The model showed good discrimination in both cohorts (an AUC of 0.904 and 0.902 for the derivation and validation cohort ROC) at the cut-off value of 0.591. The calibration curve showed good agreement between the actual observations and the predicted results. CONCLUSION:We developed a prediction model for hematologic toxicity with good discrimination and calibration capability in lung adenocarcinoma patients receiving a pemetrexed plus platinum chemotherapy regimen based on real-world data.

Automatic diagnosis of various ophthalmic diseases from ocular medical images is vital to support clinical decisions. Most current methods employ a single imaging modality, especially 2D fundus images. Considering that the diagnosis of ophthalmic diseases can greatly benefit from multiple imaging modalities, this paper further improves the accuracy of diagnosis by effectively utilizing cross-modal data. In this paper, we propose Transformer-based cross-modal multi-contrast network for efficiently fusing color fundus photograph (CFP) and optical coherence tomography (OCT) modality to diagnose ophthalmic diseases. We design multi-contrast learning strategy to extract discriminate features from cross-modal data for diagnosis. Then channel fusion head captures the semantically shared information across different modalities and the similarity features between patients of the same category. Meanwhile, we use a class-balanced training strategy to cope with the situation that medical datasets are usually class-imbalanced. Our method is evaluated on public benchmark datasets for cross-modal ophthalmic disease diagnosis. The experimental results demonstrate that our method outperforms other approaches. The codes and models are available at https://github.com/ecustyy/tcmn.

Background:Patients with atrial septal defect (ASD) exhibit distinctive electrocardiogram (ECG) patterns. However, ASD cannot be diagnosed solely based on these differences. Artificial intelligence (AI) has been widely used for specifically diagnosing cardiovascular diseases other than arrhythmia. Our study aimed to develop an artificial intelligence-enabled 8-lead ECG to detect ASD among adults.Method:In this study, our AI model was trained and validated using 526 ECGs from patients with ASD and 2,124 ECGs from a control group with a normal cardiac structure in our hospital. External testing was conducted at Wuhan Central Hospital, involving 50 ECGs from the ASD group and 46 ECGs from the normal group. The model was based on a convolutional neural network (CNN) with a residual network to classify 8-lead ECG data into either the ASD or normal group. We employed a 10-fold cross-validation approach.Results:Statistically significant differences (p < 0.05) were observed in the cited ECG features between the ASD and normal groups. Our AI model performed well in identifying ECGs in both the ASD group [accuracy of 0.97, precision of 0.90, recall of 0.97, specificity of 0.97, F1 score of 0.93, and area under the curve (AUC) of 0.99] and the normal group within the training and validation datasets from our hospital. Furthermore, these corresponding indices performed impressively in the external test data set with the accuracy of 0.82, precision of 0.90, recall of 0.74, specificity of 0.91, F1 score of 0.81 and the AUC of 0.87. And the series of experiments of subgroups to discuss specific clinic situations associated to this issue was remarkable as well.Conclusion:An ECG-based detection of ASD using an artificial intelligence algorithm can be achieved with high diagnostic performance, and it shows great clinical promise. Our research on AI-enabled 8-lead ECG detection of ASD in adults is expected to provide robust references for early detection of ASD, healthy pregnancies, and related decision-making. A lower number of leads is also more favorable for the application of portable devices, which it is expected that this technology will bring significant economic and societal benefits.

OBJECTIVES:This retrospective study aimed to identify the effectiveness of ceftazidime/avibactam (CAZ/AVI) and its optimisation programs for severe hospital-acquired pulmonary infections (sHAPi) caused by carbapenem-resistant and difficult-to-treat Pseudomonas aeruginosa (CRPA and DTR-P. aeruginosa). METHODS:We retrospectively analysed observational data on treatment and outcomes of CAZ/AVI for sHAPi caused by CRPA or DTR-P. aeruginosa. The primary study outcomes were to evaluate the clinical and microbiology efficacy of CAZ/AVI. RESULTS:The cohort consisted of 84 in-patients with sHAPi caused by CRPA (n = 39) and DTR-P. aeruginosa (n = 45) who received at least 72 h of CAZ/AVI therapy. The clinical cure rate was 63.1% in total. There was no significant difference in study outcomes between patients treated with CAZ/AVI monotherapy and those managed with combination regimens. CAZ/AVI as first-line therapy possessed prominent clinical benefits regarding infections caused by DTR-P. aeruginosa. The clinical cure rate was positively relevant with loading dose for CAZ/AVI (odds ratio [OR] 0.03; 95% confidence interval [CI] 0.004-0.19; P < 0.001) and with CAZ/AVI administration by prolonged infusion (odds ratio 0.15; 95% confidence interval 0.03-0.77; P = 0.002). APACHE II score>15 (P = 0.013), septic shock at infection onset (P = 0.001), and CAZ/AVI dose adjustment for renal dysfunction (P = 0.003) were negative predictors of clinical cure. CONCLUSION:CAZ/AVI is a valid alternative for sHAPi caused by CPRA and DTR-P. aeruginosa, even when used alone. Optimisations of the treatment with CAZ/AVI in critically ill patients, including loading dose, adequate maintenance dose and prolonged infusion, were positively associated with potential clinical benefits.

Modern agricultural irrigation has problems such as low water use efficiency and lack of reasonable guidance methods. Aiming at this problem, this paper proposes an agricultural water-saving irrigation prediction algorithm using Genetic Algorithm (GA) to optimize BP neural network. Factors such as humidity and light intensity are used as inputs to the neural network. The weights and thresholds of BP neural network are optimized by using the global search ability of genetic algorithm, and a GA-BP neural network agricultural water-saving irrigation prediction algorithm is established to predict the water demand of crops. The data of this model comes from the experimental field base of the Labor College of Tianjin University, University of Science and Technology Beijing. Using the Internet of Things technology to build a smart agricultural environment acquisition system, and adding the GA-BP crop water demand prediction intelligent algorithm to the system to achieve a low-power greenhouse environment. It can collect data and realize the function of crop water-saving irrigation. The conclusion proves that the application of GA-BP neural network agricultural water-saving irrigation prediction algorithm has high accuracy in predicting crop water demand, can better achieve the purpose of water-saving irrigation, and has strong adaptability. (Abstract)

Due to the complex morphology and characteristic of retinal vessels, it remains challenging for most of the existing algorithms to accurately detect them. This paper proposes a supervised retinal vessels extraction scheme using constrained-based nonnegative matrix factorization (NMF) and three dimensional (3D) modified attention U-Net architecture. The proposed method detects the retinal vessels by three major steps. First, we perform Gaussian filter and gamma correction on the green channel of retinal images to suppress background noise and adjust the contrast of images. Then, the study develops a new within-class and between-class constrained NMF algorithm to extract neighborhood feature information of every pixel and reduce feature data dimension. By using these constraints, the method can effectively gather similar features within-class and discriminate features between-class to improve feature description ability for each pixel. Next, this study formulates segmentation task as a classification problem and solves it with a more contributing 3D modified attention U-Net as a two-label classifier for reducing computational cost. This proposed network contains an upsampling to raise image resolution before encoding and revert image to its original size with a downsampling after three max-pooling layers. Besides, the attention gate (AG) set in these layers contributes to more accurate segmentation by maintaining details while suppressing noises. Finally, the experimental results on three publicly available datasets DRIVE, STARE, and HRF demonstrate better performance than most existing methods.

Accurate segmentation and classification of pulmonary nodules are of great significance to early detection and diagnosis of lung diseases, which can reduce the risk of developing lung cancer and improve patient survival rate. In this paper, we propose an effective network for pulmonary nodule segmentation and classification at one time based on adversarial training scheme. The segmentation network consists of a High-Resolution network with Multi-scale Progressive Fusion (HR-MPF) and a proposed Progressive Decoding Module (PDM) recovering final pixel-wise prediction results. Specifically, the proposed HR-MPF firstly incorporates boosted module to High-Resolution Network (HRNet) in a progressive feature fusion manner. In this case, feature communication is augmented among all levels in this high-resolution network. Then, downstream classification module would identify benign and malignant pulmonary nodules based on feature map from PDM. In the adversarial training scheme, a discriminator is set to optimize HR-MPF and PDM through back propagation. Meanwhile, a reasonably designed multi-task loss function optimizes performance of segmentation and classification overall. To improve the accuracy of boundary prediction crucial to nodule segmentation, a boundary consistency constraint is designed and incorporated in the segmentation loss function. Experiments on publicly available LUNA16 dataset show that the framework outperforms relevant advanced methods in quantitative evaluation and visual perception.

NOx sensors capable of accurately and quickly detecting oxygen concentration and achieving air-fuel ratio and nitrogen oxide concentration measurement play a key role in controlling exhaust emissions. Due to the coupling relationship between the chambers and the large thermal shock environment, the NOx sensor is difficult to achieve high-precision and rapid measurement under complex conditions. In order to solve this problem, this paper proposes a cascade fuzzy feedforward control algorithm for nitrogenoxygen sensors. On the one hand, the algorithm uses cascade control to complete the control of the main atmosphere influence unit. On the other hand, it is based on the fuzzy control of deviation, and does not need to establish an accurate mathematical model for the controlled object. The analysis shows that the design of the nitrogen-oxygen sensor control system proposed in this paper is feasible, and its response speed is fast and has good control effect.