To review the randomized controlled trials (RCTs) at home and abroad on digital intelligence (DI)-driven rehabilitation in patients of neuromuscular disease, compare the effects of DI-driven rehabilitation with traditional rehabilitation, summarize the special needs and challenges faced by patients in rehabilitation of rare neuromuscular diseases, and provide evidence for the development and quality improvement of rehabilitation for rare neuromuscular diseases.

To provide references to and give suggestions to the development and optimiza-tion of Digital Intelligence (DI) technology in management of non-hospitalized patients by systematical review the application of digital technology in non-hospital settings.

With the rapid advancement of artificial intelligence technology, chatbots have shown great potential in the healthcare sector. From personalized health advice to chronic disease management and psychological support, chatbots have demonstrated significant advantages in improving the efficiency and quality of healthcare services. As the scope of their applications expands, the relationship between technological complexity and practical application scenarios has become increasingly intertwined, necessitating a more comprehensive evaluation of both aspects. This paper, from the perspective of he althcare applications, systematically reviews the technological pathways and development of chatbots in the medical field, providing an in-depth analysis of their performance across various medical scenarios. It thoroughly examines the advantages and limitations of chatbots, aiming to offer theoretical support for future research and propose feasible recommendations for the broader adoption of chatbot technologies in healthcare.

Objective:To investigate the dynamic trend of platelet（PLT）count and mean platelet volume（MPV）in children with severe community-acquired pneumonia（SCAP）in PICU and their correlation with prognosis.Methods:A retrospective study was conducted in 215 SCAP children who were admitted to the PICU of Beijing Children's Hospital Affiliated to Capital Medical University from January 2016 to December 2019.According to the disease outcome，the patients were divided into improvement group ( n=184) and unrecovered group ( n=31).The changes of PLT count and MPV at admission，on the 2nd，3rd，and 7th days of hospitalization and before discharge were observed，and the relationship between changes in PLT parameters and poor prognosis was analyzed. Meanwhile，the correlation between thrombocytopenia on admission and on the 7th day of hospitalization and prognosis was further explored. Results:The PLT count of improvement group at admission，on the 2nd，3rd，and 7th days of hospitalization and at discharge[(328±159, 329±137, 362±159, 439±168, 510±171)×10 9/L] were significantly higher than those of unrecovered group [(210±142, 207±152, 267±143, 260±162, 343±159)×10 9/L]（ P＜0.05）.Although the MPV of improvement group [(10.9±1.9)fL] on admission was significantly lower than that of the unrecovered group[(12.7±2.5) fL]（ P＜0.05），there was no significant difference in MPV between two groups on the 2nd，3rd，7th days of hospitalization and discharge（ P＞0.05）.In addition，compared with the admission，children in improvement group had significantly higher PLT count on the 7th day of hospitalization and before discharge（ P＜0.05），but there was no significant change in unrecovered group（ P＞0.05）.Compared with SCAP patients with thrombocytopenia at admission (PLT＜100×10 9/L)( n=22),those with thrombocytopenia on 7th day of hospitalization had a significant higher rate of non recovery( P＜0.05). Conclusion:The occurrence of thrombocytopenia on admission and after 7 days of hospitalization in children with SCAP is associated with poor prognosis.No significant increase or decrease in PLT count after 7 days of hospitalization is often indicative of poor prognosis.Dynamic monitoring of PLT parameter changes may help to better judge the prognosis of severe pneumonia.

Objective To observe the value of CT radiomics-based nomogram for predicting difference of Lauren types of gastric cancers between endoscopic biopsy and postoperative pathology.Methods Totally 126 patients with gastric cancer diagnosed by surgical pathology were retrospectively analyzed.The patients were divided into concordant group(n=77)and inconsistent group(n=49)according to the concordance between endoscopic biopsy and postoperative pathology results or not,also divided into training set and validation set at the ratio of 2∶1.Clinical predictors were screened,then a clinical prediction model was constructed.Radiomics features were extracted based on venous-phase CT images and screened using L1 regularization.Radiomics models were constructed using 3 machine learning(ML)algorithms,i.e.decision trees,random forests and logistic regression.The nomogram based on clinical and the best ML radiomics model was constructed,and the efficacy and clinical utility of the above models and nomogram for predicting inconsistency of Lauren types of gastric cancers between endoscopic biopsy and postoperative pathology were evaluated.Results Patients'age,platelet count,and arterial-phase CT values of tumors were all independent predictors of inconsistency between endoscopic biopsy and postoperative pathology of Lauren types of gastric cancer.CT radiomics model using random forests algorithm showed better predictive efficacy among 3 ML models,with the area under the curve(AUC)of 0.835 in training set and 0.724 in validation set,respectively.The AUC of clinical model,radiomics model and the nomogram in training set was 0.764,0.835 and 0.884,while was 0.760,0.724 and 0.841 in validation set,respectively.In both training set and validation set,the nomogram showed a good fit and considerable clinical utility.Conclusion CT radiomics-based nomogram had potential clinical application value for predicting inconsistency of Lauren types of gastric cancers between endoscopic biopsy and postoperative pathology.

Objective To observe the value of spectral CT multi-parameter imaging for preoperative predicting lymph node metastasis(LNM)of gastric cancer.Methods Totally 136 patients with gastric adenocarcinoma were retrospectively enrolled.The patients were further divided into LNM group(n=74)and non-LNM group(n=62)according to postoperative pathological findings of lymph nodes status.Clinical data,conventional CT findings and spectral CT parameters were compared between groups.Factors being significant different between groups were included in multivariate logistic regression analysis to screen independent predictors of gastric cancer LNM.Clinical+conventional CT model(model 1),spectrum CT model(model 2)and combined model(model 3)were constructed based on the above independent predictors,respectively.Receiver operating characteristic curve was drawn,and the area under the curve(AUC)was calculated to evaluate the efficacy of each model for preoperative predicting LNM of gastric cancer.Results CT-N stage,CT-T stage,70,100 and 140 keV CT valuestumor at arterial phase(AP),arterial enhancement fraction(AEF)and normalized iodine concentration at venous phase(NICVP)were all independent predictors of gastric cancer LNM(all P＜0.05).AUC of model 3 was 0.846,higher than that of model 1 and model 2(AUC=0.767,0.774,Z=-0.368,-2.373,both P＜0.05)for preoperative predicting LNM of gastric cancer,while the latter two were not significantly different(Z=-0.152,P=0.879).Conclusion Spectral CT multi-parameter imaging could effectively predict LNM of gastric cancer preoperatively.

Purpose/Significance To explore the changes,challenges,key application scenarios and future development directions of generative artificial intelligence(AI)represented by ChatGPT in the field of hospital management,and to provide references for the ap-plication of AI natural language processing(NLP)technology in the field of hospital management in China.Method/Process Through literature review and analysis,the changes and challenges brought about by the rapid development of generative AI in the field of hospital management are sorted out,its key application scenarios and future development directions in the field of hospital management are empha-sized and explored.Result/Conclusion AI has broad application prospects in the field of hospital management,and it should focus on exploring its practical application scenarios and strategic directions to provide reference and guidance for promoting the high-quality de-velopment of public hospitals.

Objective To investigate the value of CT radiomics combined with CT and preoperative pathological features for predicting postoperative early recurrence(ER)of local advanced esophageal squamous cell carcinoma(LAESCC).Methods Data of 334 patients with LAESCC were retrospectively analyzed.The patients were divided into training set(n=234)and verification set(n=100)at the ratio of 7:3 and were followed up to observe ER(recurrence within 12 months after surgery)or not.Univariate and multivariate logistic regression were used to analyze clinical,CT and preoperative pathological features of LAESCC in patients with or without ER in training set.The independent risk factors of ER were screened,and a CT-preoperative pathology model was constructed.Based on venous phase CT in training set,the radiomics features of lesions were extracted and screened to establish radiomics model,and finally a combined model was established based on radiomics model and the independent risk factors.Receiver operating characteristic(ROC)curves were drawn,and the area under the curve(AUC)was calculated to evaluate the diagnostic efficacy of each model.Results Among 334 cases,168 were found with but 166 without ER.In training set,117 cases were found with while the rest 117 without ER,while in verification set,51 were found with but 49 without ER.The length of lesions,cT stage and cN stage shown on CT and tumor differentiation degree displayed with preoperative pathology were all independent risk factors for ER of LAESCC(all P＜0.05).The AUC of CT-preoperative pathology model in training set and validation set was 0.759 and 0.783,respectively.Ten best radiomics features of LAESCC were selected,and AUC of the established radiomics model in training set and validation set was 0.770 and 0.730,respectively.The AUC of combined model in training and validation set was 0.838 and 0.826,respectively.The AUC of CT radiomics combined with CT and preoperative pathological features in training set was higher than that of CT-preoperative pathologymodel and radiomics model(both P＜0.01).Conclusion CT radiomics combined with CT and preoperative pathological features could effectively predict postoperative ER of LAESCC.

Objective To observe the value of dual-energy CT(DECT)radiomics combined with clinical and CT features for predicting differentiation degree of gastric adenocarcinoma(GAC).Methods Totally 254 patients with GAC were prospectively analyzed and divided into high-grade group(low differentiation GAC,n=88)and low-grade group(middle-high differentiation GAC,n=166)according to pathological results.The patients were also divided into training set(n=203,including 70 high-grade and 133 low-grade GAC)and verification set(n=51,including 18 high-grade and 33 low-grade GAC)at the ratio of 8∶2.Radiomics features were extracted and screened based on venous phase single-level(40,70,100 and 140 keV)DECT,and a multi-energy radiomics model was constructed to predict GAC classification.Univariate analysis and multivariate logistic regression were used to analyze clinical and CT features as well as DECT parameters in training set to construct a clinic-CT model.Then a combined model was constructed through combining clinic-CT model with radiomics model.The predictive efficacy of the models were evaluated,and the calibration degree of the combined model was assessed.Results The area under the curve(AUC)of clinic-CT model,multi-energy radiomics model and combined model was 0.74,0.75 and 0.78 in training set,and 0.73,0.77 and 0.78 in verification set,respectively.Except for AUC of combined model was higher than that of clinic-CT model in training set(P＜0.05),no significant difference of AUC was found among models in training set nor verification set(all P＞0.05).The calibration degree of combined model was good in both training set and verification set(both P＞0.05).Conclusion DECT radiomics combined with clinical and CT features could effectively predict differentiation degree of GAC.

Objective To observe the value of spectral CT parameters for evaluating T staging of advanced gastric cancer(AGC).Methods Totally 155 AGC patients were collected and divided into T2 stage(n=40)and T3/4a stage(n=115)according to postoperative pathology.CT values,water concentration(WC)and iodine concentration(IC)of AGC lesions on 40-140 keV arteriovenous phase single energy level images were measured,and the standardized IC(nIC)and spectral curve slopes k1 and k2 were calculated.Clinical variables and spectral quantitative parameters were compared between groups,and receiver operating characteristic curve was plotted,the area under the curve(AUC)was calculated to evaluate the value of each parameter and model for identifying T2 and T3/4a stage AGC.Results Tumor thickness,proportion of low differentiation degree,CT100kev,CT140kev,and WC values in T3/4a group were all significantly higher than those in T2 group(all P＜0.05).CT140keV of AGC lesions on venous phase images presented the highest discrimination efficacy among single parameters,with AUC of 0.782.AUC of clinical-arterial phase-venous phase model was 0.848,higher than that of clinical model and arterial phase model alone(both P＜0.05)but not significantly different compared with AUC of venous phase model(P＞0.05).Conclusion Spectral CT quantitative parameters,especially venous phase parameters could be used to effectively identify T stage of AGC.Multi-parameter combined models had higher diagnostic value.