An enhanced machine learning approach for effective prediction of IgA nephropathy patients with severe proteinuria based on clinical data.

IgA Nephropathy (IgAN) is a disease of the glomeruli that may eventually lead to chronic kidney disease or kidney failure. The signs and symptoms of IgAN nephropathy are usually not specific enough and are similar to those of other glomerular or inflammatory diseases. This makes a correct diagnosis more difficult. This study collected data from a sample of adult patients diagnosed with primary IgAN at the First Affiliated Hospital of Wenzhou Medical University, with proteinuria ≥1 g/d at the time of diagnosis. Based on these samples, we propose a machine learning framework based on weIghted meaN oF vectOrs (INFO). An enhanced COINFO algorithm is proposed by merging INFO, Cauchy Mutation (CM) and Oppositional-based Learning (OBL) strategies. At the same time, COINFO and Support Vector Machine (SVM) were integrated to construct the BCOINFO-SVM framework for IgAN diagnosis and prediction. Initially, the proposed enhanced COINFO is evaluated using the IEEE CEC2017 benchmark problems, with the outcomes demonstrating its efficient optimization capability and accuracy in convergence. Furthermore, the feature selection capability of the proposed method is verified on the public medical datasets. Finally, the auxiliary diagnostic experiment was carried out through IgAN real sample data. The results demonstrate that the proposed BCOINFO-SVM can screen out essential features such as High-Density Lipoprotein (HDL), Uric Acid (UA), Cardiovascular Disease (CVD), Hypertension and Diabetes. Simultaneously, the BCOINFO-SVM model achieves an accuracy of 98.56%, with sensitivity at 96.08% and specificity at 97.73%, making it a potential auxiliary diagnostic model for IgAN.

An optimized machine learning model for predicting hospitalization for COVID-19 infection in the maintenance dialysis population.

COVID-19 has a high rate of infection in dialysis patients and poses a serious risk to human health. Currently, there are no dialysis centers in China that have analyzed the prevalence of COVID-19 infection in dialysis patients and the mortality rate. Although machine learning-based disease prediction methods have proven to be effective, redundant attributes in the data and the interpretability of the predictive models are still worth investigating. Therefore, this paper proposed a wrapper feature selection classification model to achieve the prediction of the risk of COVID-19 infection in dialysis patients. The method was used to optimize the feature set of the sample through an enhanced JAYA optimization algorithm based on the dispersed foraging strategy and the greedy levy mutation strategy. Then, the proposed method combines fuzzy K-nearest neighbor for classification prediction. IEEE CEC2014 benchmark function experiments as well as prediction experiments on the uremia dataset are used to validate the proposed model. The experimental results showed that the proposed method has a high prediction accuracy of 95.61% for the prevalence risk of COVID-19 infection in dialysis patients. Furthermore, it was shown that proalbumin, CRP, direct bilirubin, hemoglobin, albumin, and phosphorus are of great value for clinical diagnosis. Therefore, the proposed method can be considered as a promising method.

Meta-Resolve of Risk Factors for Nosocomial Infection in Patients Undergoing Thoracic Surgery.

As we all know, various complications may occur after surgery, and postoperative bleeding and infection are the most common in clinical practice. Postoperative infection mainly manifests as abdominal abscess, peritonitis, and fungal infection. Thoracic surgery is a very common clinical operation. It can directly deal with the relevant lesions, so a better curative effect can usually be obtained. However, patients undergoing thoracic surgery are generally more severely ill, with low immune resistance, long duration, and complicated surgical treatment process. Therefore, the probability of nosocomial infection is high, and there are many risk factors for infection. After the occurrence of HAI, it not only increases the suffering and economic burden of patients and the workload of medical staff but also prolongs the hospitalization time of patients, reduces the turnover rate of hospital beds, causes unnecessary economic losses, and affects the social and economic benefits of hospitals. Based on this, this paper proposes to analyze the risk factors of nosocomial infection in patients undergoing thoracic surgery, so as to provide a reference for the prevention or control of nosocomial infection. This paper analyzes the actual situation of nosocomial infection in a city hospital and then uses meta-analysis to determine the factors of nosocomial infection from the perspective of relevant research literature. Meta-analysis results show that patients older than 60 years have twice the risk of postoperative infection compared with patients younger than 60 years.