Artificial intelligence(AI)large models are becoming the core driving force in the field of healthcare,providing new solutions to address challenges such as population aging,chronic disease burden and uneven distribution of medical resources.The AI model based on Transformer architecture significantly improved multi task generalization ability through massive data self-supervised learning and fine-tuning techniques,enabling collaborative optimization of precise diagnosis and treatment processes,acceleration of drug development cycles and innovation of medical education models.However,this technology faces bottlenecks such as lack of evaluation criteria,risks of data privacy exposure,limitations in model reliability and computing power resources.The application status of AI large models in the medical and health field,including precision diagnosis and treatment,drug development,and medical education were reviewed,and the technological challenges and future development trends were analyzed in this article.

Artificial intelligence(AI)large models are becoming the core driving force in the field of healthcare,providing new solutions to address challenges such as population aging,chronic disease burden and uneven distribution of medical resources.The AI model based on Transformer architecture significantly improved multi task generalization ability through massive data self-supervised learning and fine-tuning techniques,enabling collaborative optimization of precise diagnosis and treatment processes,acceleration of drug development cycles and innovation of medical education models.However,this technology faces bottlenecks such as lack of evaluation criteria,risks of data privacy exposure,limitations in model reliability and computing power resources.The application status of AI large models in the medical and health field,including precision diagnosis and treatment,drug development,and medical education were reviewed,and the technological challenges and future development trends were analyzed in this article.

Purpose To analyze the multimodal imaging manifestations of phosphaturic mesenchymal tumor.Materials and Methods A retrospective analysis was performed on ten patients with phosphaturic mesenchymal tumor confirmed by pathology and molecular imaging from November 2012 to June 2022 in Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology.The imaging features of CT in four cases,of MRI in seven cases,and of nuclear medicine in ten cases were reviewed,along with the clinical characteristics of all patients.Results The main clinical symptoms of ten patients,eight cases had pain,four cases had weakness,and nine cases had multiple fractures.Laboratory tests results:ten cases showed decreased blood phosphorus,six cases showed decreased blood 25-hydroxyvitamin D and ten cases showed elevated blood alkaline phosphatase.The lesions were heterogeneous in four cases on CT scans and in six cases on MRI scans.Calcification was common,and some had cysts and fatty components.After enhancement,seven cases showed moderate to obvious uneven enhancement.99Tcm-MDP SPECT showed increased uptake of multiple bones and joints in eight cases,18F-FDG PET/CT showed no or slight increase in lesion metabolism in eight cases,68Ga-DOTATATE PET/CT showed significantly high uptake at the lesion site in ten cases.Conclusion Phosphaturic mesenchymal tumor patients typically present with pain,fractures and hypophosphatemia.The conventional imaging features are characterized by small and concealed lesions and complex components.Particularly,68Ga-DOTATATE PET/CT imaging exhibits high sensitivity for the tumor detection.

Objective:To investigate the value of radiomics based on three-dimensional high resolution MR vessel wall imaging (3D HRMR-VWI) for identifying culprit plaques in symptomatic patients with middle cerebral atherosclerosis.Methods:The clinical and imaging features of 117 patients (139 middle cerebral artery plaques) with cerebrovascular diseases in Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology from October 2018 to October 2020 were respectively reviewed. Stratified random sampling was used to divide 139 plaques into training set (97 plaques) and validation set (42 plaque) at the ratio of 7∶3. The plaques were divided into 69 culprit plaques and 70 non-culprit plaques based on plaque MR features and clinical symptoms. The clinical and imaging characteristics of culprit plaques and non-culprit plaques were compared by independent sample t-test, Mann-Whitney U test and χ 2 test, and factors with significant difference between two groups in univariate analysis were further analyzed by multivariate logistic regression to find out the independent predictors of culprit plaques. Radiomics features were extracted, screened and radiomics model was constructed using pre-and post-contrast 3D HRMR-VWI based on the training set. The combined model was constructed by combining the independent predictors and radiomics model. Receiver operating characteristic curve and area under curve (AUC) were used to evaluate the efficacy of each model, and DeLong test was used to compare the efficacy of different models. Results:Significant difference was found in intraplaque hemorrhage, lumen area of stenosis, stenosis diameter, stenosis rate, plaque burden and enhancement rate between culprit and non-culprit plaques (all P<0.05). Multivariate logistic regression analysis confirmed that only intraplaque hemorrhage was the independent predictor for culprit plaques (OR=7.045,95%CI 1.402-35.397, P=0.018). In the validation set, the AUC of the pre-contrast 3D HRMR-VWI model was lower than that of the post-contrast 3D HRMR-VWI model ( Z=-2.01, P=0.044). The AUC of pre+post-contrast 3D HRMR-VWI model was not significantly different from that of post-contrast 3D HRMR-VWI model ( Z=0.79, P=0.427). The AUC showed no significant difference between combined model and pre+post-contrast 3D HRMR-VWI model ( Z=-0.59, P>0.05). The combined model showed the best performance in predicting culprit plaques of middle cerebral artery (AUC=0.939), with the sensitivity, specificity and accuracy of 95.24%, 76.19% and 85.71%. Conclusion:Radiomics based on 3D HRMR-VWI has potential values in identifying culprit plaques in symptomatic patients with middle cerebral atherosclerosis.

Objective: To develop a model incorporating radiomic features and clinical factors to accurately predict acute ischemic stroke (AIS) outcomes. Materials and Methods: Data from 522 AIS patients (382 male [73.2%]; mean age ± standard deviation, 58.9 ± 11.5 years) were randomly divided into the training (n = 311) and validation cohorts (n = 211). According to the modified Rankin Scale (mRS) at 6 months after hospital discharge, prognosis was dichotomized into good (mRS ≤ 2) and poor (mRS > 2); 1310 radiomics features were extracted from diffusion-weighted imaging and apparent diffusion coefficient maps. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator logistic regression method were implemented to select the features and establish a radiomics model. Univariable and multivariable logistic regression analyses were performed to identify the clinical factors and construct a clinical model. Ultimately, a multivariable logistic regression analysis incorporating independent clinical factors and radiomics score was implemented to establish the final combined prediction model using a backward step-down selection procedure, and a clinical-radiomics nomogram was developed. The models were evaluated using calibration, receiver operating characteristic (ROC), and decision curve analyses. Results: Age, sex, stroke history, diabetes, baseline mRS, baseline National Institutes of Health Stroke Scale score, and radiomics score were independent predictors of AIS outcomes. The area under the ROC curve of the clinical-radiomics model was 0.868 (95% confidence interval, 0.825–0.910) in the training cohort and 0.890 (0.844–0.936) in the validation cohort, which was significantly larger than that of the clinical or radiomics models. The clinical radiomics nomogram was well calibrated (p > 0.05). The decision curve analysis indicated its clinical usefulness. Conclusion The clinical-radiomics model outperformed individual clinical or radiomics models and achieved satisfactory performance in predicting AIS outcomes.

The purpose of this study is to systematically determine an optimal percentile cutoff in histogram analysis for calculating the mean parameters obtained from a nonGaussian continuous-time random-walk (CTRW) diffusion model for differentiating individual glioma grades. This retrospective study included 90 patients with histopathologically proven gliomas (42 grade II, 19 grade III, and 29 grade IV). We performed diffusion-weighted imaging using 17 b-values (0-4000 s/mm²) at 3T, and analyzed the images with the CTRW model to produce an anomalous diffusion coefficient (D m ) along with temporal (α) and spatial (β) diffusion heterogeneity parameters. Given the tumor ROIs, we created a histogram of each parameter; computed the P-values (using a Student’s t-test) for the statistical differences in the mean D m , α, or β for differentiating grade II vs. grade III gliomas and grade III vs. grade IV gliomas at different percentiles (1% to 100%); and selected the highest percentile with P < 0.05 as the optimal percentile. We used the mean parameter values calculated from the optimal percentile cut-offs to do a receiver operating characteristic (ROC) analysis based on individual parameters or their combinations.We compared the results with those obtained by averaging data over the entire region of interest (i.e., 100th percentile). We found the optimal percentiles for D m , α, and β to be 68%, 75%, and 100% for differentiating grade II vs. III and 58%, 19%, and 100% for differentiating grade III vs. IV gliomas, respectively. The optimal percentile cut-offs outperformed the entire-ROI-based analysis in sensitivity (0.761 vs. 0.690), specificity (0.578 vs. 0.526), accuracy (0.704 vs. 0.639), and AUC (0.671 vs.0.599) for grade II vs. III differentiations and in sensitivity (0.789 vs. 0.578) and AUC (0.637 vs. 0.620) for grade III vs. IV differentiations, respectively. Percentile-based histogram analysis, coupled with the multi-parametric approach enabled by the CTRW diffusion model using high b-values, can improve glioma grading.

Grey matter (GM) alterations may contribute to cognitive decline in individuals with white matter hyperintensities (WMH) but no consensus has yet emerged. Here, we investigated cortical thickness and grey matter volume in 23 WMH patients with mild cognitive impairment (WMH-MCI), 43 WMH patients without cognitive impairment, and 55 healthy controls. Both WMH groups showed GM atrophy in the bilateral thalamus, fronto-insular cortices, and several parietal-temporal regions, and the WMH-MCI group showed more extensive and severe GM atrophy. The GM atrophy in the thalamus and fronto-insular cortices was associated with cognitive decline in the WMH-MCI patients and may mediate the relationship between WMH and cognition in WMH patients. Furthermore, the main results were well replicated in an independent dataset from the Alzheimer's Disease Neuroimaging Initiative database and in other control analyses. These comprehensive results provide robust evidence of specific GM alterations underlying WMH and subsequent cognitive impairment.

OBJECTIVE: To evaluate the secretory function of parotid glands by dynamic magnetic resonance (MR) sialography and determine the clinical performance of this technique in diagnosing and evaluating Sjögren's syndrome (SS) patients. MATERIALS AND METHODS: This study enrolled 29 healthy volunteers (25 women and 4 men; mean age, 34.8 ± 6.3 years; age range, 26–47 years) and 25 primary SS (pSS) patients (23 women and 2 men; mean age, 37.7 ± 7.9 years; age range, 25–50 years) with decreased secretory function. The volume of the parotid gland ducts was precisely measured for both groups at single pre- and 6 post-gustatory-stimulated phases. Time-dependent volume change ratio curves were generated, four parameters were derived from the curves: the slope of the increase in the first post-stimulation phase (slope(1st)), the peak value, the time-to-peak, the total saliva secretion post-stimulation. All values were used to quantitatively evaluate the secretory function of the parotid gland. The repeated measurement analysis, Mann-Whitney U test and receiver operating characteristic curve were applied. RESULTS: Time-dependent volume change ratio curves demonstrated that there is a statistically significant difference between the two groups (F = 8.750; p = 0.005). A quickly increasing curve was shown in the volunteer group, whereas a slowly increasing curve was shown in the pSS patient group. The slope(1st), peak value and total saliva secretion post-stimulation of the patient group were significantly lower than those of the volunteer group (p = 0.005, p = 0.003, and p = 0.002, respectively). The time-to-peak between the two groups was not significantly different (p = 0.383). The slope(1st) can be used as a discriminator to diagnose SS patients (p = 0.015; odds ratio = 4.234; area under the curve = 0.726). CONCLUSION: Dynamic MR sialography is proven to be an effective method in evaluating salivary gland function and has a great potential in diagnosing and evaluating pSS patients.
Autoimmune Diseases ; Female ; Healthy Volunteers ; Humans ; Male ; Methods ; Odds Ratio ; Parotid Gland ; ROC Curve ; Saliva ; Salivary Glands ; Salivation ; Sialadenitis ; Sialography ; Volunteers

Objective: To explore the application and effect of the PDCA cycle nursing management model in the treatment of peri-implant mucositis. Methods : Thirty patients with peri-implant mucositis were treated nonsurgically. Before treatment, the 30 patients had no history of systemic diseases, drug allergies, or bad habits. According to the principle of single-blind randomized control, the patients were divided into two groups: 15 patients were assigned to the control group and received routine clinical nursing and oral hygiene education according to the doctor′s prescription; and 15 patients were assigned to the intervention group, in which the PDCA cycle nursing management model was adopted. The four steps of “plan, do, check and act” were carried out. The plaque index (PL), gingival index (GI) and probe depth (PD) in the two groups were recorded before treatment and 3 and 6 months after treatment. Results: There was no significant difference in the PL, GI or PD between the intervention group and the control group before treatment (P > 0.05). Three months after treatment, the PL in the intervention group was 1.25 ± 0.44, while the PL in the control group was 1.49 ± 0.39, with a significant difference (t=2.56, P=0.008); the GI in the intervention group was 1.21 ± 0.43, while the GI in the control group was 1.56 ± 0.37, with significant difference (t=2.94, P=0.006); and the PD in the intervention group was 4.39 ± 0.41 while the PD in the control group was 4.47 ± 0.52 mm, with no significant difference (t=2.24, P=0.062). Six months after treatment, the PL in the intervention group was 1.26 ± 0.48, while the PL in the control group was 1.51 ± 0.42, with a significant difference (t=2.66, P=0.007); the GI in the intervention group was 1.34 ± 0.28, while the GI in the control group was 1.74 ± 0.48 (t=2.98, P=0.008); and the PD in the intervention group was 4.46 ± 0.52 mm, while the PD in the control group was 4.54 ± 0.66, with no significant difference (t=2.28, P=0.077). Conclusion The PDCA cycle nursing management model can enhance patients′ awareness of oral health maintenance, reduce gingival plaque accumulation, and effectively improve the health status of peri-implant tissues.