Objective To validate the accuracy of the colorectal cancer model of the Association of Coloproctology of Great Britain and Ireland (ACPGBI-CCM), and to find out the relationship between clinical risk factors and the predictive value produced by ACPGBI-CCM. Methods The patients diagnosed definitely as colorectal cancer in the department of anal-colorectal surgery, West China hospital from April 2007 to July 2007 were analyzed retrospectively. And the predictive value of mortality for each patient was calculated by ACPGBI-CCM, then the difference of risk factors was compared by classifying the patients into lower risk group and higher risk group by making the median predictive mortality as a cut point. Results From April 2007 to July 2007, a total of 99 patients diagnosed definitely as colorectal cancer accepted treatment, and among which 67 patients included in this study were admitted whose average age was 60.09 years. And there were 34 male and 33 female patients; 15 right hemicolon cancer, 9 left hemicolon cancer, 43 rectal cancer; Dukes staging: A 0 case, B 37 cases, C 24 cases, D 6 cases. The observed mortality 30 days after operation was 0, whereas the predictive mortality was 0.77%-25.75% with a median value of 3.36%. Then the patients whose predictive mortality were ≤3.36% were grouped as lower risk group (34 cases), the others higher risk group (33 cases), and there was strikingly different predictive mortality between two groups 〔(8.86±4.51)% vs (1.76±0.68)%, P＜0.01〕. And between two groups, the age, internal medicine complications, preoperative chemotherapy, ASA grading, cancer resected, and operative time made predominant differences (P＜0.01); and the neoplastic complications, Dukes staging, TNM classification, postoperative pain showed differences, too (P＜0.05); however, the gender, history of abdominal operation, the distance of the neoplasm to anal edge, the cancer location, differentiated degree, postoperative hospitalization time, and total hospitalization time didn’t have any differences (Pgt;0.05). Furthermore, stratification analysis was made for risk factors, and it came out that there were great differences of predictive mortality for different age groups and ASA grading, having internal medicine complications or not, having chemotherapy or not, and for cancer resected or not, and the differences were statistically significant (P＜0.01); also different Dukes staging or differentiation could cause different mortality (P＜0.05); but the difference of mortality didn’t make any sense according to gender, having abdominal operative history or not, having neoplastic complications or not, different TNM staging and cancer location (Pgt;0.05). Conclusion The clinical applicability of the ACPGBI-CCM is ascertained in such a large volume single medical centre, but the ACPGBI-CCM overpredicts the mortality in this study which may be attributed to the different areas, nations, or the different cultures. The complications and the neo-adjuvant or adjuvant therapy are further found out that they may be independent predictive factors of survival, and more research will be needed to prove this.
Arrhythmia is a kind of common cardiac electrical activity abnormalities. Heartbeats classification based on electrocardiogram (ECG) is of great significance for clinical diagnosis of arrhythmia. This paper proposes a feature extraction method based on manifold learning, neighborhood preserving embedding (NPE) algorithm, to achieve the automatic classification of arrhythmia heartbeats. With classification system, we obtained low dimensional manifold structure features of high dimensional ECG signals by NPE algorithm, then we inputted the feature vectors into support vector machine (SVM) classifier for heartbeats diagnosis. Based on MIT-BIH arrhythmia database, we clustered 14 classes of arrhythmia heartbeats in the experiment, which yielded a high overall classification accuracy of 98.51%. Experimental result showed that the proposed method was an effective classification method for arrhythmia heartbeats.
Objective To investigate the awareness and clinical needs of wearable artificial kidney among maintenance hemodialysis (MHD) patients, and to analyze the related influencing factors. Methods MHD patients were recruited from 2 tertiary hospitals in Sichuan province between April and June 2021. The convenient sampling method was used to select patients. The factors influencing the awareness and demand of MHD patients for wearable artificial kidney were analyzed. Results A total of 119 MHD patients were included. The awareness of wearable artificial kidney among the patients was mainly “never heard” (61 cases) and “heard” (58 cases). Most MHD patients (60 cases) were willing to use and participate in clinical trials in the future. The results of logistic regression indicated that the cost on household economy and treatment effect on life quality were the influencing factors for MHD patients’ awareness of wearable artificial kidney (P＜0.05). The average duration of single dialysis and the impact of treatment on working or studying were the influencing factors for MHD patients’ needs of wearable artificial kidney (P＜0.05). Conclusions The awareness of wearable artificial kidney is low among MHD patients. However, most MHD patients showed great interest in the wearable artificial kidney after preliminary understanding, suggesting that the future clinical application of wearable artificial kidney has great demand.