Search

Abstract Numerous deep-learning models have been developed using task-specific data, but they ignore the inherent connections among different tasks. By jointly learning a wide range of segmentation tasks, we prove that a general medical image segmentation model can improve segmentation performance for computerized tomography (CT) volumes. The proposed general CT image segmentation (gCIS) model utilizes a common transformer-based encoder for all tasks and incorporates automatic pathway modules for task prompt-based decoding. It is trained on one of the largest datasets, comprising 36,419 CT scans and 83 tasks. gCIS can automatically perform various segmentation tasks using automatic pathway modules of decoding networks through text prompt inputs, achieving an average Dice coefficient of 82.84%. Furthermore, the proposed automatic pathway routing mechanism allows for parameter pruning of the network during deployment, and gCIS can also be quickly adapted to unseen tasks with minimal training samples while maintaining great performance.

Abstract Computer-aided diagnosis (CAD) has advanced medical image analysis, while large language models (LLMs) have shown potential in clinical applications. However, LLMs struggle to interpret medical images, which are critical for decision-making. Here we show a strategy integrating LLMs with CAD networks. The framework uses LLMs’ medical knowledge and reasoning to enhance CAD network outputs, such as diagnosis, lesion segmentation, and report generation, by summarizing information in natural language. The generated reports are of higher quality and can improve the performance of vision-based CAD models. In chest X-rays, an LLM using ChatGPT improved diagnosis performance by 16.42 percentage points compared to state-of-the-art models, while GPT-3 provided a 15.00 percentage point F1-score improvement. Our strategy allows accurate report generation and creates a patient-friendly interactive system, unlike conventional CAD systems only understood by professionals. This approach has the potential to revolutionize clinical decision-making and patient communication.

Abstract Recent studies have increasingly highlighted the aberrant expression of SLC16A1-AS1 in a variety of tumor types, where it functions as either an oncogene or a tumor suppressor in the pathogenesis of different cancers. The expression levels of SLC16A1-AS1 have been found to significantly correlate with clinical features and the prognosis of cancer patients. Furthermore, SLC16A1-AS1 modulates a range of cellular functions, including proliferation, migration, and invasion, through its interactions with diverse molecules and signaling pathways. This review examines the latest evidence regarding the role of SLC16A1-AS1 in the progression of various tumors and explores its potential clinical applications as a novel prognostic and diagnostic biomarker. Our comprehensive review aims to deepen the understanding of SLC16A1-AS1’s multifaceted role in oncology, underscoring its potential as a significant biomarker and therapeutic target.

Mechanosensitive ion channel protein 1 (Piezo1) is a large homotrimeric membrane protein. Piezo1 has various effects and plays an important and irreplaceable role in the maintenance of human life activities and homeostasis of the internal environment. In addition, recent studies have shown that Piezo1 plays a vital role in tumorigenesis, progression, malignancy and clinical prognosis. Piezo1 is involved in regulating the malignant behaviors of a variety of tumors, including cellular metabolic reprogramming, unlimited proliferation, inhibition of apoptosis, maintenance of stemness, angiogenesis, invasion and metastasis. Moreover, Piezo1 regulates tumor progression by affecting the recruitment, activation, and differentiation of multiple immune cells. Therefore, Piezo1 has excellent potential as an anti-tumor target. The article reviews the diverse physiological functions of Piezo1 in the human body and its major cellular pathways during disease development, and describes in detail the specific mechanisms by which Piezo1 affects the malignant behavior of tumors and its recent progress as a new target for tumor therapy, providing new perspectives for exploring more potential effects on physiological functions and its application in tumor therapy.

Abstract Background Small intestine cancer (SIC) is difficult to diagnose early and presents a poor prognosis due to distant metastasis. This study aimed to develop nomograms for diagnosing and assessing the prognosis of SIC with distant metastasis. Methods Patients diagnosed with SIC between 2010 and 2015 were included from the Surveillance, Epidemiology and End Results database. Univariate and multifactor analysis determined independent risk factors for distant metastasis and prognostic factors for overall and cancer‐specific survival. We then constructed the corresponding three nomograms and assessed the diagnostic accuracy of the nomograms by net reclassification improvement, receiver operating characteristic curves and calibration curves, assessed the clinical utility by decision curve analysis. RESULTS The cohort consisted of 6697 patients, of whom 1299 had distant metastasis at diagnosis. Tstage, Nstage, age, tumor size, grade, and histological type were independent risk factors for distant metastasis. Age, histological type, T stage, N stage, grade, tumor size, whether receiving surgery, number of lymph nodes removed, and the presence of bone or lung metastases were predictors of both overall survival and cancer‐specific survival. The nomograms showed excellent accuracy in predicting distant metastasis and prognosis. Conclusion Nomograms were developed and validated for SIC patients with distant metastasis, aiding physicians in making rational and personalized clinical decisions.

Power transformer, as a crucial equipment for power transmission, frequently suffers winding deformation defects, owing to the impact of huge external short-circuit current in actual operation. To resist the rapid deterioration of the defects, online monitoring for winding states is imperative, depending on a thorough understanding of the deformation defects. Unfortunately, there are not much related research on the deformation form and location, as two key objects online diagnosed. In this paper, we concentrate on the analysis of winding deformation and investigation of its high-probability location to deeply understand the winding deformation. For the form of the deformation, a three-dimensional finite element model of the transformer is established to obtain the form and extent of the short-circuit impact force, and a buckling model of the low-voltage winding is designed to analyze the mechanism and form of winding deformation under the impact of huge force induced by short-circuit current. For the location of the deformation, the three-dimensional simulation models are established to analyze the force characteristics of different winding structures in three locations (the conductor transposition area, the area with imbalanced amp-turns and the area with uneven circumferential distribution of force). Then we clarify the possibility of winding deformation in these areas, defined as high-probability locations. Finally, the analyzed results are found to be highly consistent with the actual winding deformation, demonstrating the correctness of the research methods and conclusions. Next stage, the results will provide effective theoretical support for online diagnosis of winding deformation and optimal design of winding structure.

Natural products are treasure houses for modern drug discovery. Diphyllin is a natural arylnaphthalene lignan lactone isolated from the leaf of Astilboides tabularis. Studies have found that it possesses plenty of bioactivity characteristics. In this paper, we reviewed the structure, bioactivity, and mechanism of action of diphyllin and its derivatives. The references were obtained from PubMed, Web of Science, and Science Direct databases up to August 2023. Papers without a bio-evaluation were excluded. Diphyllin and its derivatives have demonstrated V-ATPase inhibition, anti-tumor, anti-virus, anti-biofilm, anti-inflammatory, and anti-oxidant activities. The most studied activities of diphyllin and its derivatives are V-ATPase inhibition, anti-tumor activities, and anti-virus activities. Furthermore, V-ATPase inhibition activity is the mechanism of many bioactivities, including anti-tumor, anti-virus, and anti-inflammatory activities. We also found that the galactosylated modification of diphyllin is a common phenomenon in plants, and therefore, galactosylated modification is applied by researchers in the laboratory to obtain more excellent diphyllin derivatives. This review will provide useful information for the development of diphyllin-based anti-tumor and anti-virus compounds.

Cavernous nerve injury (CNI) is the main cause of erectile dysfunction (ED) following pelvic surgery. Our previous studies have demonstrated that transplantation of different sources of mesenchymal stem cells (MSCs) was able to alleviate ED induced by CNI in rat models. However, little is known about the therapeutic effects of human gingiva-derived MSCs (hGMSCs) in CNI ED rats. Herein, we injected the hGMSCs around the bilateral major pelvic ganglia (MPG) in a rat model of CNI and evaluated their efficacy. The results showed that treatment of hGMSCs could significantly promote the recovery of erectile function, enhance smooth muscle and endothelial content, restore neuronal nitric oxide synthase (nNOS) expression, and attenuate cell apoptosis in penile tissue. Moreover, penile fibrosis was significantly alleviated after hGMSC administration. In addition, potential mechanism exploration indicated that hGMSCs might exert its functions via skewed macrophage polarity from M1 toward M2 anti-inflammatory phenotype. In conclusion, this study found that transplantation of hGMSCs significantly improved CNI-related ED, which might provide new clues to evaluate their pre-clinical application.

Multi-task learning (MTL) is a machine learning method to share knowledge for multiple related machine learning tasks via learning those tasks jointly. It has been shown to be capable of effectively improving the generalization capability of each single task (learning just one task at a time). In this paper, we propose a novel MTL architecture that first combines 3D convolutional neural networks (3D CNN) plus the long short-term memory (LSTM) networks together with the MTL mechanism, tailored to information sharing of video inputs. We split each video into several clips and apply the hybrid deep model of 3D CNN and LSTM to extract the sequential features of those video clips. Therefore, our MTL model can share visual knowledge based on those video-clip features among different categories more efficiently. We evaluate our method on three popular public action recognition video datasets. The experimental results show that our novel MTL method can efficiently share detailed information in video clips among multiple action categories and outperforms other multi-task methods.

Abstract Background This study investigated the therapeutic effects of MSC-derived exosomes (MSC-Exos) on erectile function in a rat model of cavernous nerve injury (CNI). Methods MSCs were isolated from rat bone marrow and exosomes were isolated from the supernatants by ultracentrifugation. The tissue explant adherent method was used to isolate and culture corpus cavernosum smooth muscle cells (CCSMCs). MSCs and CCSMCs were identified by flow cytometry, in vitro differentiation or immunofluorescence staining. Thirty-two 10-week-old male Sprague Dawley (SD) rats were divided into four groups: a sham operation group and bilateral CNI groups that received intracavernosal (IC) injection of either PBS, MSCs or MSC-Exos. Four weeks after CNI and treatment, the erectile function of the rats was measured by electrically stimulating the cavernous nerve. The penile tissues were harvested for blinded histologic analysis and western blotting. H2O2 was used to induce apoptosis in the CCSMCs, and a flow cytometer was used to measure the cell viability of the CCSMCs treated with or without exosomes in vitro. Results Recovery of erectile function was observed in the MSC-Exos group. The MSC-Exos treatment significantly enhanced smooth muscle content and neuronal nitric oxide synthase in the corpus cavernosum. The ratio of smooth muscle to collagen in the corpus cavernosum was significantly improved in the MSC-Exos treatment group compared to the PBS vehicle group. WB confirmed these biological changes. Cell viability of the CCSMCs was increased in the MSC-Exos-treated groups, and caspase-3 expression was decreased after the MSC-Exos treatment in vivo and in vitro. Conclusions Exosomes isolated from MSCs culture supernatants by ultracentrifugation could ameliorate CNI-induced ED in rats by inhibiting apoptosis in CCSMCs, with similar potency to that observed in the MSCs-treated group. Therefore, this cell-free therapy has great potential for application in the treatment of CNI-induced ED for replacing cell therapy. Graphical abstract MSC-derived exosomes ameliorate erectile dysfunction in a rat model of cavernous nerve injury

The widely distributed interconnects in printed circuit boards (PCBs) easily couple with high voltage under the action of electromagnetic pulses, which leads to insulation failure. In this study, the dielectric breakdown characteristics of four typical PCBs are studied under continuous square impulse voltage conditions. First, the electric field distribution in the four electrode models is simulated with the ANSYS software (ANSYS Maxwell 17.0). Electric field simulation results show the weak area of electric field distribution. On this basis, the possible breakdown patterns of PCB are analyzed. Second, the influence of factors, such as temperature, pulse duty ratio, interconnect insulation distance, and air pressure, on PCB breakdown voltage is studied through a breakdown test on the PCBs. Results show that the discharge between the single-layer electrodes of the PCBs is surface discharge, and the breakdown is that of a “gas⁻solid composite medium„. Meanwhile, the breakdown of a double-layer PCB is solid breakdown. Finally, scanning electron microscopy (SEM) produced by Tescan (Brno, Czech Republic) is performed to study the carbonization channel after PCB breakdown. SEM results reveal that the PCB carbonization channel is influenced by temperature and pressure in varying degrees.