Motor imagery (MI) is a mental process that can be recognized by electroencephalography (EEG) without actual movement. It has significant research value and application potential in the field of brain-computer interface (BCI) technology. To address the challenges posed by the non-stationary nature and low signal-to-noise ratio of MI-EEG signals, this study proposed a Riemannian spatial filtering and domain adaptation (RSFDA) method for improving the accuracy and efficiency of cross-session MI-BCI classification tasks. The approach addressed the issue of inconsistent data distribution between source and target domains through a multi-module collaborative framework, which enhanced the generalization capability of cross-session MI-EEG classification models. Comparative experiments were conducted on three public datasets to evaluate RSFDA against eight existing methods in terms of classification accuracy and computational efficiency. The experimental results demonstrated that RSFDA achieved an average classification accuracy of 79.37%, outperforming the state-of-the-art deep learning method Tensor-CSPNet (76.46%) by 2.91% ( P < 0.01). Furthermore, the proposed method showed significantly lower computational costs, requiring only approximately 3 minutes of average training time compared to Tensor-CSPNet's 25 minutes, representing a reduction of 22 minutes. These findings indicate that the RSFDA method demonstrates superior performance in cross-session MI-EEG classification tasks by effectively balancing accuracy and efficiency. However, its applicability in complex transfer learning scenarios remains to be further investigated.
Electroencephalography/methods* ; Brain-Computer Interfaces ; Humans ; Imagination/physiology* ; Signal Processing, Computer-Assisted ; Movement/physiology* ; Signal-To-Noise Ratio ; Deep Learning ; Algorithms

The rapid development of artificial intelligence technology is driving profound changes in medical practice, particularly in the field of medical device application. Based on data from the U.S. clinical trials registry, this study analyzes the global registration landscape of clinical trials involving artificial intelligence-based medical devices, aiming to provide a reference for their clinical research and application. A total of 2 494 clinical trials related to artificial intelligence medical devices have been registered worldwide, with participation from 66 countries or regions. The United States leads with 908 trials, while for other countries or regions, including China, each has fewer than 300 trials. Germany, the United States, and Belgium serve as central hubs for international collaboration. Among the sponsors, 63.96% are universities or hospitals, 22.36% are enterprises, and the remainder includes individuals, government agencies and others. Of all trials, 79.99% are interventional studies, 94.67% place no restrictions on participant gender, and 69.69% exclude children. The targeted diseases are primarily neurological and mental disorders. This study systematically reveals the global distribution characteristics and research trends of artificial intelligence medical device clinical trials, offering valuable data support and practical insights for advancing international collaboration, resource allocation, and policy development in this field.
Artificial Intelligence ; Humans ; Clinical Trials as Topic/statistics & numerical data* ; Equipment and Supplies ; Registries ; United States

The choice of biopsy method is critical in diagnosing prostate cancer (PCa). This retrospective cohort study compared systematic biopsy (SB) or cognitive fusion-targeted biopsy combined with SB (CB) in detecting PCa and clinically significant prostate cancer (csPCa). Data from 2572 men who underwent either SB or CB in Fudan University Shanghai Cancer Center (Shanghai, China) between January 2019 and December 2023 were analyzed. Propensity score matching (PSM) was used to balance baseline characteristics, and detection rates were compared before and after PSM. Subgroup analyses based on prostate-specific antigen (PSA) levels and Prostate Imaging-Reporting and Data System (PI-RADS) scores were performed. Primary and secondary outcomes were the detection rates of PCa and csPCa, respectively. Of 2572 men, 1778 were included in the PSM analysis. Before PSM, CB had higher detection rates for both PCa (62.9% vs 52.4%, odds ratio [OR]: 1.54, P < 0.001) and csPCa (54.9% vs 43.3%, OR: 1.60, P < 0.001) compared to SB. After PSM, CB remained superior in detecting PCa (63.1% vs 47.9%, OR: 1.86, P < 0.001) and csPCa (55.0% vs 38.2%, OR: 1.98, P < 0.001). In patients with PSA 4-12 ng ml -1 (>4 ng ml -1 and ≤12 ng ml -1 , which is also applicable to the following text), CB detected more PCa (59.8% vs 40.7%, OR: 2.17, P < 0.001) and csPCa (48.1% vs 27.7%, OR: 2.42, P < 0.001). CB also showed superior csPCa detection in those with PI-RADS 3 lesions (32.1% vs 18.0%, OR: 2.15, P = 0.038). Overall, CB significantly improves PCa and csPCa detection, especially in patients with PSA 4-12 ng ml -1 or PI-RADS 3 lesions.
Humans ; Male ; Prostatic Neoplasms/diagnosis* ; Propensity Score ; Retrospective Studies ; Middle Aged ; Aged ; Image-Guided Biopsy/methods* ; Prostate-Specific Antigen/blood* ; Prostate/diagnostic imaging*

OBJECTIVES: To analyze the disease burden of melanoma among middle-aged and elderly populations in China, and to predict the future trend. METHODS: Data from the Global Burden of Disease (GBD) 2021 were utilized to collect incidence and mortality rates of melanoma, disability-adjusted life years (DALYs), and corresponding age crude rates among the middle-aged and elderly population in China during 1990 and 2021. Additionally, the estimated annual percentage change (EAPC) was employed to assess the temporal trends. Age-period-cohort (APC) and Bayesian age-period-cohort (BAPC) models were utilized to compute age, period, and cohort effects on incidence and mortality rates of melanoma, as well as to predict future trends up to 2035. RESULTS: During 1990-2021, the incidence rate of melanoma for males was higher than that for females among the middle-aged and elderly population in China, and the overall incidence rate increased annually with an EAPC of 2.13 (1.90-2.36), while the overall mortality rate and DALY rate showed a declining trend with an EAPC of －0.28 (－0.41-－0.15) and －0.54 (－0.68-－0.41), respectively. The results of the APC model analysis revealed that age effects on both incidence and mortality rates of melanoma in China's middle-aged and elderly population were significant, with both increasing with age. Period and cohort effects showed an upward trend for incidence rates but a downward trend for mortality rates. Moreover, the period and cohort effects for mortality rates were not significant among females. In the BAPC prediction model, the number of incidences of melanoma in middle-aged and elderly people in China would increase dramatically. By 2035, the number of incidence cases is expected to reach approximately 9600 (males) and 10 300 (females), corresponding to an incidence rate of 2.66/10⁵ and 2.67/10⁵, respectively. The number of deaths is projected to be about 2600 (males) and 3500 (females) by 2035, corresponding to a mortality rate of 0.72/10⁵ and 0.91/10⁵, respectively. CONCLUSIONS The disease burden of melanoma among the middle-aged and elderly population in China remains substantial and is expected to increase over the next decade.
Humans ; Melanoma/mortality* ; China/epidemiology* ; Aged ; Middle Aged ; Male ; Female ; Incidence ; Disability-Adjusted Life Years ; Bayes Theorem ; Cost of Illness ; Skin Neoplasms/epidemiology*

OBJECTIVE: The relationship between benign prostatic hyperplasia (BPH) and the oxidative balance score (OBS) will be discussed in this study. METHODS: The clinical data on 16 dimensions of diet and 4 dimensions of lifestyle from the National Health and Nutrition Examination Survey (NHANES) from 2001 to 2008 were used to calculate OBS. We considered BPH as the outcome and investigated the linear and nonlinear relationships between the two. Additionally, subgroup analyses and interaction tests were conducted as well. Furthermore, the methods of machine learning including XGBoost, support vector machine (SVM) and naive Bayes (NB) were used to establish a predictive model for BPH. RESULTS: Higher OBS was consistently associated with an increased prevalence of BPH, with Restricted Cubic Splines highlighting a significant positive nonlinear association (P=0.015). Subgroup analyses revealed differences and interactive relationships based on alcohol consumption. Among the seven machine learning models that we included the OBS score in, the XGBoost model emerged as the best, with an AUC value of 0.769. CONCLUSION There is a significant association between OBS and the prevalence of BPH in the American population, which provides a valuable insight for further diagnosis and research of the disease.
Humans ; Male ; Prostatic Hyperplasia/epidemiology* ; Machine Learning ; Bayes Theorem ; Nutrition Surveys ; Support Vector Machine ; Life Style ; Oxidative Stress ; Aged ; Diet ; Prevalence

OBJECTIVE: To explore the differences in perioperative clinical and pathological characteristics of patients with different pathological types of prostate cancer undergoing radical prostatectomy, and to analyze the influencing factors that may affect the extraglandular invasion of ductal adenocarcinoma of the prostate. METHODS: Retrospective collection was made of the radical prostatectomy patients who were admitted to Peking University Third Hospital from December 2011 to April 2021. The patients were screened based on inclusion criteria to obtain basic clinical features and postoperative pathological results. According to the pathological results, the patients were divided into ductal adenocarcinoma group (mixed with ductal adenocarcinoma) and acinar adenocarcinoma group, and a 1 ∶1 propensity score matching was performed to compare the differences in clinical characteristics between the two groups. Univariate and multivariate analyses of the factors related to extraglandular invasion were performed in the matched ductal adenocarcinoma groups. RESULTS: A total of 764 patients with prostate cancer were enrolled in this study, of which 62 patients were confirmed to have ductal adenocarcinoma components by postoperative pathology. There was a statistically significant difference in the proportion of the patients with a history of diabetes in baseline characteristics between the two groups before propensity score matching (29.5% vs. 17.7%, P=0.027). A total of 61 patients with simple acinar adenocarcinoma were successfully matched with the patients with ductal adenocarcinoma, and there was no statistically significant difference in baseline characteristics between the two groups after matching (P>0.05). The comparison of perioperative clinical and pathological features showed that International Society of Urology Pathology (ISUP) grade (P=0.003), pT stage (P=0.004), extraglandular invasion rate (P=0.018) and vascular thrombus rate (P=0.019) in ductal adenocarcinoma group were significantly higher than those in simple acinous adenocarcinoma group. Univariate analysis of the influence factors of extraglandular invasion showed that prostate-specific antigen (PSA) level, prostate volume, ISUP grade, seminal vesicle invasion and perineural invasion might be the influencing factors of extraglandular invasion (P < 0.10). Binary Logistic regression analysis showed that perineural invasion was an independent factor of extraglandular invasion (OR=11.78, 95%CI: 1.97-70.56, P=0.007). CONCLUSION Prostatic ductal adenocarcinoma has a worse prognosis than simple acinar adenocarcinoma. Perineural invasion is the influencing factor of extraglandular invasion of ductal adenocarcinoma.
Humans ; Male ; Prostatic Neoplasms/surgery* ; Retrospective Studies ; Prostatectomy ; Neoplasm Invasiveness ; Middle Aged ; Aged ; Carcinoma, Ductal/surgery* ; Propensity Score ; Adenocarcinoma/surgery*

OBJECTIVE: Evaluate the clinical application effect of low-field infant MRI. METHODS: Using literature review, expert consultation, and two rounds of Delphi to determine the evaluation index system. Then retrospectively analyze and compare the data of low-field infant MRI and high-field MRI from January 2023 to December 2024. RESULTS: There is a certain gap between low-field infant MRI and high-field MRI in terms of signal-to-noise ratio, image uniformity, software system reliability, scanning time, user interface friendliness and image result consistency. However, there was no difference in terms of spatial resolution and image quality. The noise, hardware system reliability, mean time between failure and the rate of examination completed without sedation are better than that of high-field MRI. CONCLUSION Low-field infant MRI meets needs of clinical diagnostic and has stable performance. It can be used as a routine screening tool for brain diseases near the bed.
Magnetic Resonance Imaging/methods* ; Humans ; Infant ; Retrospective Studies ; Signal-To-Noise Ratio ; Reproducibility of Results ; Brain Diseases/diagnostic imaging* ; Brain/diagnostic imaging* ; Software

OBJECTIVE: To enhance the reliability of medical equipment, this study aims to develop a failure cause diagnosis model and provide rational suggestions for efficient equipment use. METHODS: Combine fault tree analysis (FTA) to identify basic events causing equipment failure and calculate their prior probabilities. Obtain conditional probability tables for each node through expert assessment. Integrate triangular fuzzy number theory with Bayesian network (BN) to construct a fuzzy Bayesian network (FBN) for posterior probability inference and sensitivity analysis. RESULTS: Using endoscopes as the subject, the analysis shows that the model accurately calculates the endoscope failure probability at 0.385%, and identifies the key causes: improper cleaning ( X5, posterior probability 0.36064), untimely fault detection ( X8, posterior probability 0.23571), irregular transportation ( X6, posterior probability 0.11344), and natural aging ( X10, posterior probability 0.11377). Sensitivity analysis also confirms their influence weights (mutual information values are 0.00749, 0.00591, 0.00202, 0.00174). CONCLUSION The model can accurately perform quantitative analysis and rapid fault location of medical equipment failures, enabling effective preventive measures.
Bayes Theorem ; Fuzzy Logic ; Equipment Failure Analysis/methods* ; Equipment Failure ; Algorithms

OBJECTIVES: Nasopharyngeal carcinoma is often diagnosed at a late stage due to its concealed location and exhibits marked regional clustering, posing a significant public health challenge in China. This study aims to analyze the disease burden of nasopharyngeal carcinoma in China using the latest 2021 Global Burden of Diseases (GBD) database, providing epidemiological evidence for precise prevention and control of nasopharyngeal carcinoma. METHODS: Age-standardized incidence rate (ASIR), mortality rate, and disability-adjusted life year (DALY) rate were used as indicators of disease burden. Stratified analyses were conducted by age, sex, socio-demographic index (SDI), and relevant risk factors. The autoregressive integrated moving average (ARIMA) model and Bayesian age-period-cohort (BAPC) model were employed to project ASIR trends through 2050. RESULTS: In 2021, China's age-standardized incidence, mortality, and DALY rates of nasopharyngeal carcinoma were 3.4/100 000, 1.5/100 000, and 48.7/100 000, respectively, all higher than the global average. Across all age groups, Chinese males exhibited higher ASIR, mortality, and DALY rates than females. From 1990 to 2021, the disease burden of nasopharyngeal carcinoma in China decreased gradually with rising SDI. The proportion of nasopharyngeal carcinoma burden attributed to alcohol consumption, smoking, and occupational formaldehyde exposure in China exceeded global levels, especially among males. Projections from both models indicate a rising trend in ASIR for males, females, and the general population in China and globally from 2022 to 2050. CONCLUSIONS Over the past 30 years, the disease burden of nasopharyngeal carcinoma in China has decreased with the increasing SDI values but remains higher than the global average. Furthermore, ASIR is projected to increase over the next 30 years. It is imperative for China to enhance healthcare resource allocation for nasopharyngeal carcinoma prevention, diagnosis, and treatment, particularly among high-risk male populations.
Humans ; China/epidemiology* ; Male ; Nasopharyngeal Carcinoma/mortality* ; Female ; Middle Aged ; Nasopharyngeal Neoplasms/mortality* ; Adult ; Incidence ; Global Burden of Disease ; Disability-Adjusted Life Years ; Aged ; Risk Factors ; Adolescent ; Databases, Factual ; Young Adult ; Cost of Illness ; Child ; Bayes Theorem

OBJECTIVES: Global epidemiological data indicate that 20% to 30% of intensive care unit (ICU) sepsis patients progress to deep vein thrombosis (DVT) due to coagulopathy, with an associated mortality rate of 25% to 40%. Existing prognostic tools have limitations. This study aims to develop and validate nomogram and machine learning models to predict in-hospital mortality in sepsis patients with DVT and assess their clinical applicability. METHODS: This multicenter retrospective study drew on data from the Medical Information Mart for Intensive Care IV (MIMIC-IV; n=2 235), the eICU Collaborative Research Database (eICU-CRD; n=1 274), and the Patient Admission Dataset from the ICU of Third Xiangya Hospital, Central South University (CSU-XYS-ICU; n=107). MIMIC-IV was split into a training set (n=1 584) and internal validation set (n=651), with the remaining datasets used for external validation. Predictors were selected via least absolute shrinkage and selection operator (LASSO) regression and Bayesian Information Criterion (BIC), and a nomogram model was constructed. An extreme gradient boosting (XGBoost) algorithm was used to build the machine learning model. Model performance was assessed by the concordance index (C-index), calibration curves, Brier score, decision curve analysis (DCA), and net reclassification improvement index (NRI). RESULTS: Five key predictors, age [odds ratio (OR)=1.02, 95% CI 1.01 to 1.03, P<0.001], minimum activated partial thromboplastin (APTT; OR=1.09, 95% CI 1.08 to 1.11, P<0.001), maximum APTT (OR=1.01, 95% CI 1.00 to 1.01, P<0.001), maximum lactate (OR=1.56, 95% CI 1.39 to 1.75, P<0.001), and maximum serum creatinine (OR=2.03, 95% CI 1.79 to 2.30, P<0.001), were included in the nomogram. The model showed robust performance in internal validation (C-index=0.845, 95% CI 0.811 to 0.879) and external validation (eICU-CRD: C-index=0.827, 95% CI 0.800 to 0.854; CSU-XYS-ICU: C-index=0.779, 95% CI 0.687 to 0.871). Calibration curves indicated good agreement between predicted and observed outcomes (Brier score<0.25), and DCA confirmed clinical benefit. The XGBoost model achieved an area under the receiver operating characteristic curve (AUC) of 0.982 (95% CI 0.969 to 0.985) in the training set, but performance declined in external validation (eICU-CRD, AUC=0.825, 95% CI 0.817 to 0.861; CSU-XYS-ICU, AUC=0.766, 95% CI 0.700 to 0.873), though it remained above clinical thresholds. Net reclassification improvement was slightly lower for XGBoost compared with the nomogram (NRI=0.58). CONCLUSIONS Both the nomogram and XGBoost models effectively predict in-hospital mortality in sepsis patients with DVT. However, the nomogram offers superior generalizability and clinical usability. Its visual scoring system provides a quantitative tool for identifying high-risk patients and implementing individualized interventions.
Humans ; Sepsis/complications* ; Machine Learning ; Nomograms ; Venous Thrombosis/complications* ; Retrospective Studies ; Hospital Mortality ; Male ; Female ; Middle Aged ; Aged ; Intensive Care Units ; Prognosis ; Bayes Theorem