To construct a predictive model of cerebral hemorrhage prognosis prognosis using multimodal medical data and machine learning, and to explore its predictive value.

A total of 400 cerebral hemorrhage data was retrospective analyzed, included 98 cerebral hemorrhage patients in the Neurology Department of Peking Union Medical College Shenzhen Hospital, Huazhong University of Science and Technology from January to December 2020 and 302 cerebral hemorrhage patients from the National Multicenter Intracranial Hemorrhage Database established by Peking Union Medical College Hospital from January to December 2020 to build an imaging database of cerebral hemorrhage, extract the labels of imaging omics and clinical related factors, and build a model to predict patients' prognosis. Another 100 patients with intracerebral hemorrhage in the Neurology Department of Peking Union Medical College Shenzhen Hospital, Huazhong University of Science and Technology from January to December 2021 were selected for prospective verification of the model.

The incidence of poor prognosis in 400 patients was 19.00%. The results of univariate analysis and multivariate logistic regression analysis showed that GCS score, systolic blood pressure, diastolic blood pressure, blood glucose, hematoma volume, peripheral edema volume, fibrinogen were all the influencing factors of poor prognosis (P<0.05). The sensitivity, specificity, accuracy and area under curve (AUC) [95% confidence interval (95%CI)] of self coding image feature clinical data model for predicting prognosis in patients with cerebral hemorrhage were 100.00%, 99.38%, 99.50% and 0.994 (0.935-0.998) respectively, which were higher than those of self coding image feature model and traditional model (P<0.05), and those of the self coding image feature model were higher than those of traditional model (P<0.05). After verification, the sensitivity, specificity, accuracy and AUC (95%CI) of the self coded image feature clinical data model for predicting prognosis in patients with cerebral hemorrhage were 100.00%, 97.47%, 98.00% and 0.974 (0.922-0.996) respectively.

The self coding image feature clinical data model based on multi-modal medical data and machine learning has high efficiency in predicting cerebral hemorrhage prognosis.

To systematically review the efficacy of hyperbaric oxygen therapy (HBOT) on cognitive function in patients with traumatic brain injury (TBI).

PubMed, Embase, Web of Science, The Cochrane Library, VIP, WanFang Data, and CNKI databases were electronically searched to collect studies of random control trails of HBOT treatment for TBI patients from inception to September 9^th, 2022. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies. Finally, meta-analysis was performed using RevMan 5.4 software.

A total of 47 RCTs involving 4321 TBI patients were included. The results of the meta-analysis showed that the scores of GCS (MD=2.22, 95%CI: 1.91-2.53, P<0.001), mini-mental status examination (MD=2.62, 95%CI: 1.72-3.52, P<0.001), Montreal cognitive assessment (MD=3.31, 95%CI: 0.86-5.76, P<0.001), Barthel index (MD=13.36, 95%CI: 9.56-17.15, P<0.001), modified Barthel index (MD=9.91, 95%CI: 4.33-15.48, P<0.001), functional independence measure (MD=15.76, 95%CI: 7.83-23.68, P<0.001) in HBOT group were superior to the control group.

Current evidence shows that HBOT may have positive influence on cognitive function and activities of daily life in patients with TBI.

To investigate the clinical efficacy and health economic evaluation of hematoma drainage for spontaneous intracerebral hemorrhage (SICH) with the assistance of neurosurgical navigation and positioning planning system (referred to as robot).

The clinical data and economic data of 199 SICH patients who underwent surgical treatment from March 2019 to March 2022 were collected. All patients were divided into 4 groups according to surgical methods: the robotic surgery group consisted of 77 patients who underwent robot assisted stereotactic hematoma drainage surgery; 65 cases in the neuro-endoscopy groups underwent hematoma removal surgery under neuroendoscopy; 21 cases in the small bone window craniotomy group underwent microscopic hematoma removal with small bone window open; 36 cases in the rigid catheter group underwent CT image-guided free-hand rigid catheter technique. Clinical follow-up outcomes and medical cost from the immediate post-operative to one year period were compared and analyzed. Data from both the robotic surgery and neuro-endoscopy groups were sampled and matched using propensity scoring methods, and health economics were evaluated using modified Rankin scale (mRS) scores and quality adjusted life years (QALYs) indicators.

At the time of discharge, there were differences between the groups in terms of operation time, hematoma residual volume, total hospital costs, surgery-related costs, number of days in the ICU, and duration of ventilator use (P<0.05). In the robotic surgery group complication rate of rebleeding was 2.6%, intracranial infection was 1.3%, average hospital stay was 15.45 d, and average hospital cost was ¥46 077.90. There was difference in the proportion of mRS≤3 points between the 4 groups (P<0.05), in which the proportion of patients with mRS≤3 points at 3 months and 1 year after surgery in the robotic surgery group was 55.8% and 74.0%, respectively. Propensity score sampling was matched, resulting in 37 patients each in the balanced and comparable robotic surgery and neuro-endoscopy groups. One year after surgery, the robotic surgery group was able to save ¥36 862.14 per capita and gain 0.062 more QALYs.

Based on our model of SICH calculations suggest that robotic-assisted stereotactic drainage has the health economic advantage of being less costly and more effective, further results await multicenter, prospective randomized controlled trials with expanded sample size.

To analyze the registration status and development trend of clinical trials for vascular cognitive impairment (VCI) based on data from the WHO International Clinical Trials Registry Platform (ICTRP), China Clinical Trials Registry (ChiCTR) and National Institutes of Health Clinical Trial Registry Platform (ClinicalTrials.gov) registry platform.

The clinical trial information of VCI on ICTRP, ChiCTR, and ClinicalTrials.gov platforms since the creation of the databases until December 31, 2022. After the literature was independently screened and data extracted by 2 investigators, descriptive analyses were conducted in terms of enrollment profiles, study design, interventions, and observational indicators of the included trials.

After screening, 177 VCI-related clinical trials were finally included, and the overall enrolment showed a fluctuating growth trend. ChiCTR (67 trials) and ClinicalTrials.gov (87 trials) were the main enrolment platforms. China (110 trials) dominated the applications for VCI-related clinical trials, with Beijing (38 trials), Shanghai (11 trials), Hubei (8 trials) and Taiwan (8 trials) as the main hotspot provinces and cities for VCI research in China. Research in this field is dominated by interventional studies (132 trials). Randomised and blinded methods were more commonly used. The design mode was mainly randomised parallel control (110 trials). The sample size varied widely among studies, and most of them did not specify the clinical trial stage (81 trials). Interventions were predominantly pharmacological therapies versus placebo, with a relatively small proportion of non-pharmacological therapy studies. The main observational indicators were cognitive performance scales and quality of daily life scales.

At present, research on VCI is on the rise, and there is much room for research on VCI not dementia, vascular mild cognitive impairment disease types and non-pharmacological therapies. China should continue to strengthen international and domestic inter-regional cooperation to produce more high-quality evidence-based evidence for VCI.

To study the relationship between the first CT signs after mechanical thrombectomy and prognosis of patients based on different neutrophil-lymphocyte ratios (NLR).

The clinical data of 201 patients who underwent mechanical thrombectomy from January 2017 to December 2021 in the Neurology Department of The First People's Hospital of Zhangjiagang City were retrospectively analyzed. According to the first postoperative CT imaging, they were divided into high-density CT imaging group and non high-density CT imaging group; At the same time, they were divided into low NLR group and high NLR group according to preoperative NLR values, and then subgroup analysis was conducted based on whether there were high-density shadows on the first postoperative CT. The 90-day good prognosis rate, the incidence of symptomatic intracranial hemorrhage and the mortality rate within 90 d were compared among the subgroups.

There were 100 patients with high-density CT imaging after operation (high-density CT imaging group), and the remaining 101 patients were included in the non high-density CT imaging group. The rate of symptomatic intracranial hemorrhage in the non high-density CT imaging group was lower than that in high-density CT imaging group (OR=0.207, 95%CI: 0.082-0.521, P<0.001), but there was no statistically significant difference in 90-day good prognosis and mortality between the two groups (P>0.05). There were 137 patients in the low NLR group. There was no significant difference in the rate of 90-day good prognosis, mortality and symptomatic intracranial hemorrhage at 90 d between the subgroup without and with high-density CT imaging (P>0.05). There were 64 patients in the high NLR group. Compared with the subgroup with high-density CT imaging, the subgroup without high-density CT imaging had a higher 90-day good prognosis (OR=3.515, 95%CI: 1.051-11.748, P=0.041), and a lower 90-day mortality (OR=0.262, 95%CI: 0.071-0.977, P=0.046), symptomatic intracranial hemorrhage rate was low (OR=0.132, 95%CI: 0.026-0.679, P=0.015).

After mechanical thrombectomy, CT with high-density imaging has a higher rate of symptomatic intracranial hemorrhage than without high-density imaging, but it has no significant effect on prognosis. In patients with high NLR, the first high-density CT imaging after thrombectomy affects the clinical prognosis; In patients with low NLR, the presence or absence of high-density has no significant effect on prognosis.

To investigate the effect of extracorporeal shock wave therapy (ESWT) combined with electromyography biofeedback therapy (EMG-BFT) on foot drop after stroke.

Ninety patients with poststroke foot drop admitted to the Rehabilitation Medicine Department of Enshi Tujia and Miao Autonomous Prefecture Central Hospital from May 2019 to May 2022 were randomly divided into 3 group using a random number table method: observation group, BFT group, and control group, 30 patients in each group. The control group received routine rehabilitation treatment, the BFT group received EMG-BFT on the basis of the control group, and the observation group received ESWT combined with EMG-BFT on the basis of the control group. All 3 groups were treated for 4 weeks. The ankle dorsiflexion mobility, comprehensive spasticity scale (CSS) and Fugl-Meyer motor function scale lower extremity (FMA-LE) and surface electromyography were performed before and after 4 weeks of treatment for all patients. The improvement of ankle range of motion, ankle spasms, lower limb motor function, and key muscle surface electromyography were observed.

The differences in ankle mobility, CSS score, FMA-LE score, anterior tibial muscle iEMG, gastrocnemius muscle iEMG and foot dorsiflexion CR before treatment in the 3 groups were not statistically significant (P>0.05). After treatment, ankle mobility, CSS score, and FMA-LE score in all three groups improved significantly compared with those before treatment (P<0.05); the results of tibialis anterior muscle iEMG, gastrocnemius muscle iEMG and foot dorsiflexion CR in patients in the observation group and BFT group improved significantly compared with those before treatment (P<0.05), and after treatment, the ankle mobility, CSS score, anterior tibialis iEMG, and foot dorsiflexion CR were better in the observation group than in the BFT group and the control group (P<0.05).

The combined use of ESWT and EMG-BFT was significantly more effective in improving the active and passive mobility of the affected ankle joint, improving spasticity, and improving motor function of the lower extremity than the absence of both treatments or the use of EMG-BFT alone.

To evaluate the effectiveness of building cynomolgus monkey middle cerebral artery occlusion (MCAO) model by taking supraorbital keyhole approach operation.

Four ordinary grade experimental monkeys were randomly divided into two groups. Three experimental monkeys of operation group underwent left supraorbital keyhole approach craniotomy, initial part of M1 segment of left middle cerebral artery was exposed and then coagulated and cut to make MCAO model. One experimental monkey of control group underwent left supraorbital keyhole approach craniotomy, initial part of M1 segment of left middle cerebral artery was exposed but not coagulated and cut to make negative control. All experimental monkeys were evaluated using the non-human primate stroke scale (NHPSS) at 24 h, 3 d, 7 d, 14 d, and 28 d postoperatively. Brain MRI axial T2 weighted scans were performed on all experimental monkeys at 7 d postoperatively.

All experimental monkeys survived after operation without any significant postoperative complications, and lived well during experimental period. The NHPSS scores of the operation group at postoperative time of 24 h, 3 d, 7 d, 14 d, 28 d was (19.67±1.15), (22.00±1.00), (19.33±1.15), (18.67±1.53), (18.00±1.00) points respectively. The NHPSS scores of the control group at postoperative time of 24 h, 3 d, 7 d, 14 d, 28 d were all 0 points. Brain MRI axial T2 weighted scan at 7 d after operation showed that brain infarction focus appeared in the blood supply area of left middle cerebral artery of all experimental monkeys of operation group, scattered in the left frontal lobe, parietal lobe, temporal lobe and basal ganglion area, and accompanied with mild cerebral edema, whereas no brain infarction focus and brain edema was appeared in the experimental monkey of control group.

The application of the supraorbital keyhole approach surgical method has the advantage of minimally invasion, reliable model effect, good repeatability. The approach method has good repeatability and can make mass production of cynomolgus monkey MCAO models.

To investigate the clinical efficacy of minimally invasive puncture and drainage combined with pilacetan and urapidil in the treatment of hypertensive intracerebral hemorrhage (HICH) in the basal ganglia and its effects on the levels of serum markers of brain injury, matrix metalloproteinases (MMPs), interleukin-1β (IL-1β) and high mobility group protein B1 (HMGB1).

Eighty-six patients with HICH in basal ganglia region admitted to the Neurosurgery Department of Beijing Luhe Hospital Affiliated to Capital Medical University from January 2019 to December 2021 were randomly divided into a control group and an observation group, with 43 patients in each group. The control group was treated with minimally invasive puncture and drainage combined with urapidil, and the observation group was treated with piracetam based on the control group. After 14 d of continuous treatment, the clinical efficacy of the two groups was compared. The National Institutes of Health stroke scale (NIHSS) score was used to evaluate the neurological function of the subjects before and after treatment, GCS score was used to evaluate the consciousness status of the subjects, and mini-mental state examination (MMSE) was used to evaluate the cognitive function of the patients. Serum brain injury markers [GFAP, neuron-specific enolase (NSE), S100β], MMPs (MMP-2, MMP-3, MMP-9), IL-1β and HMGB1 were detected by enzyme-linked immunoassay before and after treatment. The adverse drug reactions of the two groups were recorded.

The total effective rate of the observation group was 90.70% (39/43), which was significantly higher than that of the control group [74.42% (32/43)] (P<0.05). After treatment, NIHSS score and levels of serum NSE, GFAP, S100β protein, MMP-2, MMP-3, MMP-9, IL-1β and HMGB1 were significantly lower than before treatment, and GCS and MMSE scores were significantly higher than before treatment, and the changes were more significant in observation group (P<0.05). There was no obvious drug side reaction in all subjects.

Minimally invasive puncture and drainage combined with piracetam and urapidil in the treatment of BASAL ganglia HICH can safely and effectively reduce the levels of serum markers of brain injury, MMPs, IL-1β and HMGB1, promote the relief of inflammatory response, and alleviate neurological deficits.

To explore the risk factors of cerebral hernia after interventional surgery for ruptured intracranial aneurysms, and to establish a nomogram model to predict the risk of cerebral hernia after interventional surgery accordingly.

The clinical data of 398 patients with ruptured intracranial aneurysms who underwent interventional surgery therapy in Department of Neurology of Huishan District People's Hospital of Wuxi City from January 2017 to October 2022 were analyzed retrospectively, and they were divided into concurrent group and non-concurrent group according to the complications of postoperative cerebral hernia. The clinical data of the two groups were compared, and the risk factors of cerebral hernia after interventional surgery for ruptured intracranial aneurysms were analyzed by multivariate logistic regression analysis method, and a nomogram model for predicting the risk of cerebral hernia after interventional surgery for ruptured intracranial aneurysms was constructed according to the results of multivariate analysis, and Bootstrap method was used to internally verify the nomogram model and receiver operating characteristic (ROC) curve was used to evaluate the prediction efficiency of the nomogram model.

A total of 48 cases of 398 patients developed cerebral hernia after operation (the concurrent group), and the complication rate of cerebral hernia was 12.06%, and the rest were the non-concurrent group (350 cases). The proportions of hypertension history, Hunt-Hess grade Ⅳ-Ⅴ at admission, CT Fisher grade Ⅲ-Ⅳ at admission, World Federation of Neurosurgical Societies (WFNS) grade Ⅲ-Ⅴ at admission, operation time>2 h, hydrocephalus, brain edema, wide necked aneurysms and re rupture of aneurysms in the concurrent group were higher than those in the non-concurrent group, and the differences were statistically significant (P<0.05). Multivariate Logistic regression analysis showed that the history of hypertension, Hunt-Hess grade Ⅳ-Ⅴ at admission, CT Fisher grade Ⅲ-Ⅳ at admission, WFNS grade Ⅲ-Ⅴ at admission, operation time>2 h, wide necked aneurysms, brain edema and re rupture of aneurysms were the risk factors of cerebral hernia after interventional surgery for ruptured intracranial aneurysms (P<0.05). The nomogram model was constructed based on the above results of multivariate analysis, and the consistency index of it was 0.841, and ROC curve showed that the area under the curve of the nomogram model to predict the risk of cerebral hernia after interventional surgery for ruptured intracranial aneurysm was 0.836 (95%CI: 0.785-0.879, P<0.001), and the sensitivity was 83.33% and the specificity was 75.93%.

History of hypertension, Hunt-Hess grade at admission, CT Fisher grade at admission, WFNS grade at admission, wide necked aneurysms, brain edema and re rupture of aneurysms are the risk factors of brain hernia after interventional surgery for ruptured intracranial aneurysms, and the prediction nomogram model based on this has good discrimination, and the prediction efficiency is high, which can provide a reference for early individualized prediction of the risk of cerebral hernia and identification of high-risk patients.

To observe the clinical efficacy of virtual scenario interaction technology combined with shoulder blade motor control training on upper limb motor function and daily living activities in hemiplegia.

From May 2019 to October 2021, 105 patients with stroke hemiplegia admitted to the Department of Rehabilitation Medicine of Hefei Second People's Hospital were selected as the research subjects, and were divided into virtual group, scapula group and joint group by random number table method, with 35 cases in each group. All three groups of patients were given conventional rehabilitation training treatment, the virtual group was supplemented with virtual scenario interactive training, the scapula group was supplemented with scapula motion control intensive training, and the joint group was given virtual situational interaction technology combined with shoulder blade motion control intensive training. Before treatment and 4 weeks after treatment, Brunnstrom staged upper limb part, simplified Fugl-Meyer assessment upper extremity (FMA-UE), Wolf motor function test (WMFT), Hong Kong edition of functional test for the hemiplegic upper extremities (FTHUE-HK) were used to evaluate the patient's upper limb recovery, and the modified Barthel index (MBI) was used to evaluate the patient's daily living activities.

After 4 weeks of treatment, the FMA-UE score, WMFT score, MBI score, Brunnstrom stage upper limb part and FTHUE-HK grade of the three groups were all improved to a certain extent compared with the pre-treatment treatment of this group, and the outcomes of the joint group were better than those in the virtual group and the scapular group, and the differences were statistically significant (P<0.05).

Virtual scenario interactive technology combined with scapular motor control intensive training can significantly improve upper limb motor function and daily living activities of patients with hemiplegia after stroke.