Predicting Outcome of Patients With Cerebral Hemorrhage Using a Computed Tomography-Based Interpretable Radiomics Model: A Multicenter Study.

OBJECTIVE: The aim of this study was to develop and validate an interpretable and highly generalizable multimodal radiomics model for predicting the prognosis of patients with cerebral hemorrhage. METHODS: This retrospective study involved 237 patients with cerebral hemorrhage from 3 medical centers, of which a training cohort of 186 patients (medical center 1) was selected and 51 patients from medical center 2 and medical center 3 were used as an external testing cohort. A total of 1762 radiomics features were extracted from nonenhanced computed tomography using Pyradiomics, and the relevant macroscopic imaging features and clinical factors were evaluated by 2 experienced radiologists. A radiomics model was established based on radiomics features using the random forest algorithm, and a radiomics-clinical model was further trained by combining radiomics features, clinical factors, and macroscopic imaging features. The performance of the models was evaluated using area under the curve (AUC), sensitivity, specificity, and calibration curves. Additionally, a novel SHAP (SHAPley Additive exPlanations) method was used to provide quantitative interpretability analysis for the optimal model. RESULTS: The radiomics-clinical model demonstrated superior predictive performance overall, with an AUC of 0.88 (95% confidence interval, 0.76-0.95; P < 0.01). Compared with the radiomics model (AUC, 0.85; 95% confidence interval, 0.72-0.94; P < 0.01), there was a 0.03 improvement in AUC. Furthermore, SHAP analysis revealed that the fusion features, rad score and clinical rad score, made significant contributions to the model's decision-making process. CONCLUSION: Both proposed prognostic models for cerebral hemorrhage demonstrated high predictive levels, and the addition of macroscopic imaging features effectively improved the prognostic ability of the radiomics-clinical model. The radiomics-clinical model provides a higher level of predictive performance and model decision-making basis for the risk prognosis of cerebral hemorrhage.

[Incidence of comorbidities in hospitalized chronic obstructive pulmonary disease patients and correlated risk factors].

OBJECTIVE: To explore the incidence of comorbidities in hospitalized chronic obstructive pulmonary disease (COPD) patients and influencing factors. METHODS: A retrospective review of medical records was performed for 495 hospitalized COPD patients in Peking University Third Hospital from January 2003 to December 2008. Their comorbidities were identified and ranked in prevalence. The risk factors of comorbidities were analyzed by multivariable Logistic regression. RESULTS: The most frequent comorbidities were: hypertension (60.0%), ischemic heart disease (16.0%), malignant tumor (10.9%), diabetes mellitus (10.5%), chronic heart failure (9.9%) and dyslipidemia (9.9%). The percentage of patients with ischemic heart disease in people with FEV(1)% Pred < 30%, 30% ≤ FEV(1)%Pred < 50%, 50% ≤ FEV(1)%Pred < 80% and FEV(1)%Pred ≥ 80% were 9.2%, 12.7%, 21.3% and 16.5% respectively (P = 0.052). The percentage of patients with dyslipidemia in people with FEV(1)%Pred < 30%, 30% ≤ FEV(1)%Pred < 50%, 50% ≤ FEV(1)%Pred < 80%, FEV(1)%Pred ≥ 80% were 5.3%, 5.7%, 10.9% and 20.3% respectively (P = 0.002). The percentage of patients with ischemic heart disease in people with body mass index (BMI) < 18.5, 18.5 - 23.9, 24.0 - 27.9, ≥ 28.0 kg/m(2) were 13.6%, 14.2%, 14.6% and 28.3% respectively (P = 0.051). The percentage of patients with dyslipidemia in people with BMI < 18.5, 18.5 - 23.9, 24.0 - 27.9, ≥ 28.0 kg/m(2) were 5.5%, 5.7%, 14.6% and 25.0% respectively (P = 0.000). The levels of C-reactive protein (CRP) were higher than the 75(th) percentile (OR = 2.371, P = 0.014), and complications with metabolic syndrome (OR = 2.694, P = 0.003) were independent risk factors of ischemic heart disease in COPD patients. CONCLUSIONS: The incidence of comorbidities varies in different COPD populations. Higher levels of CRP and complications with metabolic syndrome are independent risk factors of ischemic heart disease in COPD patients.

[Causes of chronic obstructive pulmonary disease-related death and influencing factors of survival time].

OBJECTIVE: To study the causes of chronic obstructive pulmonary disease (COPD)-related death and influencing factors of survival time from first hospitalization due to acute exacerbation of chronic obstructive pulmonary disease (AECOPD). METHODS: All patients with a primary or secondary diagnosis of COPD (International Classification of Diseases (ICD)-10 Codes J40-J47) were continuously enrolled at our hospital from January 2006 to December 2008. A retrospective review was performed on the medical records of COPD patients who died during hospitalization. The causes of death were coded and analyzed according to the International Classification of Diseases (ICD)-10. The underlying causes of death were identified and ranked in order of prevalence. The medical records of first hospitalization due to AECOPD were reviewed. Retrospective analysis was performed for the clinical data. And Cox regression analysis was used to select the independent risk factors of influencing the survival time. RESULTS: Sixty-seven patients died during hospitalization. The median FEV(1) (forced expiratory volume in 1 second percentage) was 34%. The causes of death in the COPD inpatients were as follows: respiratory diseases (n = 39), cardio-cerebrovascular diseases (n = 16), malignant tumors (n = 10), diabetes mellitus (n = 1) and suicide (n = 1). Among them, 54 patients had a history of hospitalization due to AECOPD. Thirteen deceased patients were never hospitalized because of AECOPD. The mean interval between the first admission with AECOPD as the primary diagnosis and death was 38 months (range: 1 - 159). The independent risk factors of influencing the survival time were as follows: complications with coronary artery disease & severe pulmonary hypertension, age, body mass index (BMI) and serum level of C-reactive protein (CRP) (all P < 0.05). CONCLUSION: The major cause of death in moderate-severe COPD patients is respiratory disease. complicated with coronary artery disease & severe pulmonary hypertension, age, BMI and serum level of CRP are the independent risk factors of affecting the survival time from first hospitalization due to AECOPD.