With the emergence of deep learning techniques based on convolutional neural networks, artificial intelligence (AI) has driven transformative developments in the field of medical image analysis. Recently, large language models (LLMs) such as ChatGPT have also started to achieve distinction in this domain. Increasing research shows the undeniable role of AI in reshaping various aspects of medical image analysis, including processes such as image enhancement, segmentation, detection in image preprocessing, and postprocessing related to medical diagnosis and prognosis in clinical settings. However, despite the significant progress in AI research, studies investigating the recent advances in AI technology in the aforementioned aspects, the changes in research hotspot trajectories, and the performance of studies in addressing key clinical challenges in this field are limited. This article provides an overview of recent advances in AI for medical image analysis and discusses the methodological profiles, advantages, disadvantages, and future trends of AI technologies.
Artificial Intelligence ; Humans ; Image Processing, Computer-Assisted/methods* ; Neural Networks, Computer ; Deep Learning ; Diagnostic Imaging/methods*

Gut microbial communities are likely remodeled in tandem with accumulated physiological decline during aging, yet there is limited understanding of gut microbiome variation in advanced age. Here, we performed a metagenomics-based enterotype analysis in a geographically homogeneous cohort of 367 enrolled Chinese individuals between the ages of 60 and 94 years, with the goal of characterizing the gut microbiome of elderly individuals and identifying factors linked to enterotype variations. In addition to two adult-like enterotypes dominated by Bacteroides (ET-Bacteroides) and Prevotella (ET-Prevotella), we identified a novel enterotype dominated by Escherichia (ET-Escherichia), whose prevalence increased in advanced age. Our data demonstrated that age explained more of the variance in the gut microbiome than previously identified factors such as type 2 diabetes mellitus (T2DM) or diet. We characterized the distinct taxonomic and functional profiles of ET-Escherichia, and found the strongest cohesion and highest robustness of the microbial co-occurrence network in this enterotype, as well as the lowest species diversity. In addition, we carried out a series of correlation analyses and co-abundance network analyses, which showed that several factors were likely linked to the overabundance of Escherichia members, including advanced age, vegetable intake, and fruit intake. Overall, our data revealed an enterotype variation characterized by Escherichia enrichment in the elderly population. Considering the different age distribution of each enterotype, these findings provide new insights into the changes that occur in the gut microbiome with age and highlight the importance of microbiome-based stratification of elderly individuals.
Aged ; Aged, 80 and over ; Female ; Humans ; Male ; Middle Aged ; Bacteroides ; China ; Diabetes Mellitus, Type 2/microbiology* ; Escherichia coli/classification* ; Gastrointestinal Microbiome/genetics* ; Metagenomics ; East Asian People

Objective Focusing on gynecological surgery,we constructed a prediction model for surgical duration by extracting features from unstructured surgical planning texts and integrating multimodal data via artificial intelligence technology.Methods The clinical data of 34 614 patients who underwent gynecologic surgeries at West China Second University Hospital,Sichuan University between January 2022 and October 2024 were collected.An embedding-transformer model was constructed to convert surgical planning texts into a one-dimensional numerical feature,referred to as the step feature.The predictive value of the step feature was assessed by comparing the performance improvements of linear regression,random forest,eXtreme Gradient Boosting(XGBoost),support vector regression,K-nearest neighbor regression,and artificial neural network algorithms in two scenarios—with and without the step feature as an input.The out-of-sample prediction accuracy of the models was assessed using mean absolute error(MAE),root mean squared error(RMSE),and R-squared(R2).Furthermore,the model interpretability was examined using SHapley Additive exPlanations(SHAP)values.Results SHAP results showed that the step feature had the highest predictive contribution.Temporal factors in surgical scheduling also influenced gynecological surgery duration.The XGBoost model demonstrated optimal performance on the test set,significantly improving prediction accuracy with a 40.43%increase in R2,while reducing MAE and RMSE by 21.27%and 20.13%,respectively,compared to the baseline model without the step feature.Conclusion The embedding-transformer model developed in this study effectively extracts features from surgical planning texts and enhances the predictive performance of machine learning models.The XGBoost prediction model can assist hospital administrators in implementing more refined management of gynecological surgeries and improving the utilization efficiency of surgical resources.

Objective:To investigate the distribution characteristics of skeletal muscle mass and strength in the older adults over 65 years old in 18 longevity areas in China.Methods:The subjects were selected from the Healthy Aging and Biomarkers Cohort Study conducted in 18 longevity areas of China. A total of 4 662 older adults over 65 years old from a cross- sectional survey in 2021 were included in the study. The information about their sociodemographic characteristics, lifestyle, nutrient intake and other factors were collected through questionnaire surveys and physical examinations. Grip strength was measured by using professional electronic grip dynamometer. Total skeletal muscle mass (TSM) was measured using bioelectrical impedance analysis, and TSM was adjusted by height squared and BMI to obtain TSM Ht2 and TSM BMI. The proportion of individuals with low muscle mass and strength was determined according to the recommended method by the Asian Working Group for Sarcopenia (AWGS). Descriptive analysis was conducted on the population and regional distribution characteristics of people with different muscle mass and grip strength. A generalized additive model was used to analyze the age-related trends of muscle mass and grip strength. Results:The age of 4 662 study subjects was (82.69±10.54) years, men accounted for 46.85% (2 184 cases) and Han Chinese accounted for 96.27% (4 488 cases). The M( Q1, Q3) of TSM, TSM Ht2 and TSM BMI in men were 23.30 (20.50, 26.20) kg, 9.02 (8.13, 9.89) kg/m 2, and 1.01 (0.90, 1.13) kg·(kg/m 2) -1, respectively, which were all higher than those in women [TSM: 18.20 (15.70, 20.70) kg, TSM Ht2: 8.18 (7.42, 9.07) kg/m 2 and TSM BMI: 0.79 (0.69, 0.90) kg·(kg/m 2) -1], the differences were significant (all P<0.001). The grip strength of men [ M( Q1, Q3): 24.50 (17.80, 30.80) kg] was higher than that of women [ M( Q1, Q3): 15.60 (11.10, 19.90) kg], the difference was significant ( P<0.001). Southern elderly men had lower TSM and TSM Ht2 compared with northern elderly men (all P<0.001), while there was no significant regional difference in TSM BMI ( P>0.05). Southern elderly women had higher TSM Ht2 and TSM BMI compared with northern elderly women (all P<0.001), while there was no significant regional difference in TSM ( P>0.05). Furthermore, according to the method recommended by AWGS, the elderly with low muscle mass and grip strength were characterized by older age, illiteracy, being unmarried/divorced/widowed, poor chewing ability, impaired activity of daily living and living in southern region. Conclusion:There were population and regional differences in muscle mass and grip strength in the older adults over 65 years in 18 longevity areas of China, and these differences showed decreasing trends with age.

Background::Carotid intima-media thickness (IMT) and diameter, stiffness, and wave reflections, are independent and important clinical biomarkers and risk predictors for cardiovascular diseases. The purpose of the present study was to establish nationwide reference values of carotid properties for healthy Chinese adults and to explore potential clinical determinants.Methods::A total of 3053 healthy Han Chinese adults (1922 women) aged 18-79 years were enrolled at 28 collaborating tertiary centers throughout China between April 2021 and July 2022. The real-time tracking of common carotid artery walls was achieved by the radio frequency (RF) ultrasound system. The IMT, diameter, compliance coefficient, β stiffness, local pulse wave velocity (PWV), local systolic blood pressure, augmented pressure (AP), and augmentation index (AIx) were then automatically measured and reported. Data were stratified by age groups and sex. The relationships between age and carotid property parameters were analyzed by Jonckheere-Terpstra test and simple linear regressions. The major clinical determinants of carotid properties were identified by Pearson’s correlation, multiple linear regression, and analyses of covariance.Results::All the parameters of carotid properties demonstrated significantly age-related trajectories. Women showed thinner IMT, smaller carotid diameter, larger AP, and AIx than men. The β stiffness and PWV were significantly higher in men than women before forties, but the differences reversed after that. The increase rate of carotid IMT (5.5 μm/year in women and 5.8 μm/year in men) and diameter (0.03 mm/year in both men and women) were similar between men and women. For the stiffness and wave reflections, women showed significantly larger age-related variations than men as demonstrated by steeper regression slopes (all P for age by sex interaction <0.05). The blood pressures, body mass index (BMI), and triglyceride levels were identified as major clinical determinants of carotid properties with adjustment of age and sex. Conclusions::The age- and sex-specific reference values of carotid properties measured by RF ultrasound for healthy Chinese adults were established. The blood pressures, BMI, and triglyceride levels should be considered for clinical application of corresponding reference values.

Objective:To compare image quality and diagnostic parameters of whole-brain CT perfusion scans under different scanning conditions and assess the utility of deep learning image reconstruction algorithm (DLIR) in reducing tube current during low-dose scans.Methods:Method A total of 105 patients with suspected acute ischemic stroke (AIS) were prospectively enrolled in the First Affiliated Hospital of Zhengzhou University from March, 2022 to March, 203 and their baseline information was recorded. All patients underwent head non-contrast CT and CT perfusion (CTP) examinations. CTP scanning was performed at 80 kV in two groups with the tube current of 150 mA (regular dose) and 100 mA (low dose), respectively. The CTP images of 150 mA group were reconstructed using filtered back-projection algorithm as well as adaptive statistical iterative reconstruction-V (ASIR-V) at 40% and 80% strength levels, which were denoted as groups A-C. The CTP images of 100 mA group were reconstructed using ASIR-V80%, DLIR-M, and DLIR-H, which were denoted as groups D-F. Clinical baseline characteristics and radiation doses were compared between the two groups under different scanning conditions. Furthermore, we assessed the subjective and objective image quality, conventional perfusion parameters, and abnormal perfusion parameters of AIS patients across the six groups of reconstructed CTP images.Results:Under the scanning conditions of 150 mA and 100 mA, 47 and 48 patients were diagnosed with AIS, respectively. There were no significant differences in the baseline characteristics between the two groups. However, there was a significant difference in the mean effective radiation dose (5.71 mSv vs. 3.80 mSv, t = 2 768.30, P < 0.001). The standard deviation (SD) of noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) of gray matter (GM) and white matter (WM) were significantly different among the six groups of reconstructed images ( F = 40.58-212.13, P < 0.001). In GM, the SD values in groups C, D, and F were lower than those in other groups ( P < 0.05), and the SNR values in groups C and F were higher than those in other groups ( P < 0.05). In WM, the SD and SNR values in groups C and F were significantly different from those in other groups ( P < 0.05). Additionally, CNR values in groups C and F were higher than those in other groups ( P < 0.05). There was no significant difference in subjective scores among groups B, C, and F ( P > 0.05). Regarding perfusion parameters in the brain GM, groups D and E had lower cerebral blood volume (CBV) values compared to groups A to C ( P < 0.05), and group F had lower CBV values than group B ( P < 0.05). In the brain WM, group D had consistently lower mean transit time (MTT) values compared to the other groups ( P < 0.05). Notably, there were no significant differences in AIS lesion detection rates and relevant diagnostic parameters across the six image groups. Conclusions:Low-tube current CTP scan combined with the DLIR-H algorithm can enhance image quality without affecting perfusion parameters such as CBV and MTT, while reducing radiation dose by 30%. This algorithm can be routinely applied in brain CTP examinations.

Malignant pleural mesothelioma(MPM) is a kind of rare and aggressive malignant neoplasm. Surgery plays one of the most important roles in the treatment of MPM. However, due to the high morbidity and mortality reported, the survival benefit and indication of surgery are still controversial. This article will review the surgical indications, discuss and compare the roles of extrapleural pneumonectomy(EPP) and pleurectomy / decortication(P/D) which aim to achieve macroscopic complete resection(MCR) in the treatment of MPM. Finally, we summarized the progress of other treatment methods including targeted therapy and immunotherapy.

Objective:To investigate the associations of cardiometabolic multimorbidity (CMM) with grip strength and gait speed among older Chinese adults.Methods:This study included participants aged ≥60 years from the China Health and Retirement Longitudinal Survey during 2011-2015. Generalized estimating equation models were employed to estimate the associations of CMM with grip strength and gait speed.Results:A total of 6 357 participants were included to measure grip strength and 6 250 participants to measure gait speed. Compared with no cardiometabolic disease, participants with 1 ( β=-0.018, 95% CI: -0.026--0.010), 2 ( β=-0.029, 95% CI: -0.041- -0.018), and ≥3 ( β=-0.050, 95% CI: -0.063- -0.037) cardiometabolic diseases were associated with a decreased grip strength. The associations between cardiometabolic disease counts (1: β=-0.052, 95% CI: -0.326 -0.222; 2: β=-0.083, 95% CI: -0.506 -0.340; ≥3: β=-0.186, 95% CI: -0.730 -0.358) and gait speed were not statistically significant. The predictive value of gait speed of the participants with 0, 1, 2, and ≥3 cardiometabolic diseases were found to be 1.98 (95% CI: 1.38-2.58), 1.93 (95% CI: 1.34-2.51), 1.89 (95% CI: 1.18-2.61), and 1.79 (95% CI: 1.10-2.48) m/s respectively, which was clinically significant for the magnitude of the decrease. Cardiometabolic combinations with a higher risk of decreased grip strength and gait speed mainly seen in diabetes. Conclusions:Cardiometabolic disease counts and combinations were associated with grip strength and gait speed. Grip strength and gait speed can be used to measure CMM severity.

Objective:To evaluate the clinical effect of anterior clinoid process grinding in the treatment of ophthalmic / superior clinoid process aneurysms and sellar tumors.Methods:The clinical data of 16 patients who underwent anterior clinoid process grinding in Sanbo Brain Hospital, Capital Medical University from January 2015 to July 2021 were analyzed retrospectively. There were 1 patient with recurrent craniopharyngioma, 1 patient with recurrent pituitary adenoma, 13 patients with aneurysms, and 1 patient with suprasellar granulosa cell tumor combined with ophthalmic aneurysm of right internal carotid artery. The Modified Rankin Scale (mRS) score was used to evaluate the situation at discharge and in the medium-and-long term.Results:Sixteen patients underwent anterior clinoidprocess grinding. At discharge and the latest follow-up, the mRS scores of the patients were 0-2. A total of 15 aneurysms were treated, and there were no symptoms of visual loss or visual field defect after operation. No cerebrospinal fluid leakage occurred in all patients.Conclusions:The grinding of anterior clinoid process can effectively and fully stretch the optic nerve and internal carotid artery, and can observe the tumor neck at the lower end of pituitary stalk and the ocular segment/superior clinoid process of internal carotid artery under direct vision. It is one of the important auxiliary methods for the treatment of sellar lesions.