Outcome Prediction of Spontaneous Supratentorial Intracerebral Hemorrhage after Surgical Treatment Based on Non-Contrast Computed Tomography: A Multicenter Study.

This study aims to explore the value of a machine learning (ML) model based on radiomics features and clinical features in predicting the outcome of spontaneous supratentorial intracerebral hemorrhage (sICH) 90 days after surgery. A total of 348 patients with sICH underwent craniotomy evacuation of hematoma from three medical centers. One hundred and eight radiomics features were extracted from sICH lesions on baseline CT. Radiomics features were screened using 12 feature selection algorithms. Clinical features included age, gender, admission Glasgow Coma Scale (GCS), intraventricular hemorrhage (IVH), midline shift (MLS), and deep ICH. Nine ML models were constructed based on clinical feature, and clinical features + radiomics features, respectively. Grid search was performed on different combinations of feature selection and ML model for parameter tuning. The averaged receiver operating characteristics (ROC) area under curve (AUC) was calculated and the model with the largest AUC was selected. It was then tested using multicenter data. The combination of lasso regression feature selection and logistic regression model based on clinical features + radiomics features had the best performance (AUC: 0.87). The best model predicted an AUC of 0.85 (95%CI, 0.75-0.94) on the internal test set and 0.81 (95%CI, 0.64-0.99) and 0.83 (95%CI, 0.68-0.97) on the two external test sets, respectively. Twenty-two radiomics features were selected by lasso regression. The second-order feature gray level non-uniformity normalized was the most important radiomics feature. Age is the feature with the greatest contribution to prediction. The combination of clinical features and radiomics features using logistic regression models can improve the outcome prediction of patients with sICH 90 days after surgery.

Research on nonstroke dementia screening and cognitive function prediction model for older people based on brain atrophy characteristics.

BACKGROUND: Brain atrophy is an important feature in dementia and is meaningful to explore a brain atrophy model to predict dementia. Using machine learning algorithm to establish a dementia model and cognitive function model based on brain atrophy characteristics is unstoppable. METHOD: We acquired 157 dementia and 156 normal old people.s clinical information and MRI data, which contains 44 brain atrophy features, including visual scale assessment of brain atrophy and multiple linear measurement indexes and brain atrophy index. Five machine learning models were used to establish prediction models for dementia, general cognition, and subcognitive domains. RESULTS: The extreme Gradient Boosting (XGBoost) model had the best effect in predicting dementia, with a sensitivity of 0.645, a specificity of 0.839, and the area under curve (AUC) of 0.784. In this model, the important brain atrophy features for predicting dementia were temporal horn ratio, cella media index, suprasellar cistern ratio, and the thickness of the corpus callosum genu. CONCLUSION: For nonstroke elderly people, the machine learning model based on clinical head MRI brain atrophy features had good predictive value for dementia, general cognitive impairment, immediate memory impairment, word fluency disorder, executive dysfunction, and visualspatial disorder.

Machine Learning Decomposition of the Anatomy of Neuropsychological Deficit in Alzheimer's Disease and Mild Cognitive Impairment.

Objective: Alzheimer's Disease (AD) is a progressive condition characterized by cognitive decline. AD is often preceded by mild cognitive impairment (MCI), though the diagnosis of both conditions remains a challenge. Early diagnosis of AD, and prediction of MCI progression require data-driven approaches to improve patient selection for treatment. We used a machine learning tool to predict performance in neuropsychological tests in AD and MCI based on functional connectivity using a whole-brain connectome, in an attempt to identify network substrates of cognitive deficits in AD. Methods: Neuropsychological tests, baseline anatomical T1 magnetic resonance imaging (MRI), resting-state functional MRI, and diffusion weighted imaging scans were obtained from 149 MCI, and 85 AD patients; and 140 cognitively unimpaired geriatric participants. A novel machine learning tool, Hollow Tree Super (HoTS) was utilized to extract feature importance from each machine learning model to identify brain regions that were associated with deficit and absence of deficit for 11 neuropsychological tests. Results: 11 models attained an area under the receiver operating curve (AUC-ROC) greater than 0.65, while five models had an AUC-ROC ≥ 0.7. 20 parcels of the Human Connectome Project Multimodal Parcelation Atlas matched to poor performance in at least two neuropsychological tests, while 14 parcels were associated with good performance in at least two tests. At a network level, most parcels predictive of both presence and absence of deficit were affiliated with the Central Executive Network, Default Mode Network, and the Sensorimotor Networks. Segregating predictors by the cognitive domain associated with each test revealed areas of coherent overlap between cognitive domains, with the parcels providing possible markers to screen for cognitive impairment. Conclusion: Approaches such as ours which incorporate whole-brain functional connectivity and harness feature importance in machine learning models may aid in identifying diagnostic and therapeutic targets in AD.

Identification and validation of critical genes with prognostic value in gastric cancer.

Background: Gastric cancer (GC) is a digestive system tumor with high morbidity and mortality rates. Molecular targeted therapies, including those targeting human epidermal factor receptor 2 (HER2), have proven to be effective in clinical treatment. However, better identification and description of tumor-promoting genes in GC is still necessary for antitumor therapy. Methods: Gene expression and clinical data of GC patients were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Last absolute shrinkage and selection operator (LASSO) Cox regression were applied to build a prognostic model, the Prognosis Score. Functional enrichment and single-sample gene set enrichment analysis (ssGSEA) were used to explore potential mechanisms. Western blotting, RNA interference, cell migration, and wound healing assays were used to detect the expression and function of myosin light chain 9 (MYL9) in GC. Results: A four-gene prognostic model was constructed and GC patients from TCGA and meta-GEO cohorts were stratified into high-prognosis score groups or low-prognosis score groups. GC patients in the high-prognosis score group had significantly poorer overall survival (OS) than those in the low-prognosis score groups. The GC prognostic model was formulated as PrognosisScore = (0.06 × expression of BGN) - (0.008 × expression of ATP4A) + (0.12 × expression of MYL9) - (0.01 × expression of ALDH3A1). The prognosis score was identified as an independent predictor of OS. High expression of MYL9, the highest weighted gene in the prognosis score, was correlated with worse clinical outcomes. Functional analysis revealed that MYL9 is mainly associated with the biological function of epithelial-mesenchymal transition (EMT). Knockdown of MYL9 expression inhibits migration of GC cells in vitro. Conclusion: We found that PrognosisScore is potential reliable prognostic marker and verified that MYL9 promotes the migration and metastasis of GC cells.

Chimeric antigen receptor T-cell therapy: a promising treatment modality for relapsed/refractory mantle cell lymphoma.

Mantle cell lymphoma (MCL) is a distinct histological type of B-cell lymphoma with a poor prognosis. Several agents, such as proteasome inhibitors, immunomodulatory drugs, and inhibitors of B cell lymphoma-2 and Bruton's tyrosine kinase have shown efficacy for relapsed or refractory (r/r) MCL but often have short-term responses. Chimeric antigen receptor (CAR) T-cell therapy has emerged as a novel treatment modality for r/r non-Hodgkin's lymphoma. However, long-term safety and tolerability associated with CAR T-cell therapy are not defined well, especially in MCL. In this report, we described a 70-year-old patient with r/r MCL with 48-month duration of follow-up who achieved long-term remission after CAR T-cell therapy. CAR T-cell-related toxicities were also mild and tolerated well even in this elderly patient. This report suggested that CAR T-cell therapy is a promising treatment modality for patients with MCL, who are generally elderly and have comorbid conditions.

[The effect of continuous low doses X-ray on the proliferation of pyramidal cell in hippocampus CA1 in prenatal rats].

OBJECTIVE: To explore the effect of low doses X-ray on proliferation of hippocampal pyramidal cell in the area of CA1 in prenatal rat and its relevant mechanism. METHODS: A total of 25 pregnant rats were randomly divided into four experimental groups and one control group. The experimental groups, in a duration of consistent 18 days, respectively received different doses as follows: 0.015 mGy/d, 0.03 mGy/d, 0.06 mGy/d and 0.09 mGy/d. The control group received sham radiation. To observe the density and width of hippocampal pyramidal cell in the area of CA1 by HE stained and observe the expression of the ERK1/2 by IHM. RESULTS: (1) Except C group, all other groups presented increment in width of the level of hippocampal pyramidal cell, compared with C group; H group, M group, L1 group and L2 group were higher than that (F value respectively were 8.475, 33.42, 14.395, 44.955; P value respectively were 0.002, 0.048, 0.030, 0.012). But the phenomenon of inhomogeneity in width in H group was observed, at the same time, the density of cell in H group became looser (F = 4.466, P = 0.017). (2) The expression of ERK1/2 in the hippocampus CA1 was seen in cytoplasm of every group, the average optical density of positive ERK1/2 protein significantly increased in L1 group and L2 group, compared with control group respectively (F value respectively were 4.561, 4.103, P value respectively were 0.044, 0.035). CONCLUSION: Low doses X-ray could promote proliferation of hippocampus CA1 cell in prenatal. The reason could be the increment of the ERK1/2 protein induced by X-ray. When the doses reached 0.09 mGy/d, the excesses proliferation phenomenon was observed.

[CT and pathological manifestations of intraductal papillary mucinous neoplasm of pancreas: a comparative study].

OBJECTIVE: To evaluate the diagnostic value of CT in pancreas intraductal papillary mucinous neoplasm (IPMN) by analyzing its CT feature and pathological findings. METHODS: The clinical and CT data was analyzed among 39 patients with IPMN whose diagnosis was confirmed by pathology. The CT manifestations were classified into 3 types: simple main pancreatic duct enlargement; main pancreatic duct enlargement combined with pancreatic cystic lesion; and simple pancreatic cystic lesion. The correlation between the CT types and Takada pathological types (main duct type, branch type, and mixed type) was analyzed. All the cases were pathologically classified into benign and malignant/boundary groups. Statistical tests of the difference of CT features (mural nodule, septa, size, caliber of main pancreatic duct and common bile duct) between the 2 groups were performed. RESULTS: The CT type I matched the main duct type, the CT type II mainly matched the branch type and mixed type, and the CT type III matched the branch type (P < 0.001). The probability of benign lesion was 92% with no mural nodule in the lesion, while the probability of benign lesion was only 42% with mural nodule presented (P = 0.003). In terms of the septa, there was no significant difference between benign and malignant lesions (P = 0.793). The size of malignant/boundary lesions exceeded that of benign lesions (P = 0.016). There were no significant difference in calibers of main pancreatic duct and common bile duct between the benign and malignant/ boundary groups. Without considering pathological grouping the caliber of main pancreatic duct exceeded that of the common bile duct in all the cases (P = 0.02). CONCLUSION: CT typing of IPMN well matches the pathological typing which benefits the CT diagnosis of IPMN. The caliber of main pancreatic duct usually exceeds that of common bile duct in IPMN. This feature contributes to its diagnosis.