ABSTRACT
Acute respiratory distress syndrome (ARDS) is a serious threat to human life and health disease, with acute onset and high mortality. The current diagnosis of the disease depends on blood gas analysis results, while calculating the oxygenation index. However, blood gas analysis is an invasive operation, and can't continuously monitor the development of the disease. In response to the above problems, in this study, we proposed a new algorithm for identifying the severity of ARDS disease. Based on a variety of non-invasive physiological parameters of patients, combined with feature selection techniques, this paper sorts the importance of various physiological parameters. The cross-validation technique was used to evaluate the identification performance. The classification results of four supervised learning algorithms using neural network, logistic regression, AdaBoost and Bagging were compared under different feature subsets. The optimal feature subset and classification algorithm are comprehensively selected by the sensitivity, specificity, accuracy and area under curve (AUC) of different algorithms under different feature subsets. We use four supervised learning algorithms to distinguish the severity of ARDS (P/F ≤ 300). The performance of the algorithm is evaluated according to AUC. When AdaBoost uses 20 features, AUC = 0.832 1, the accuracy is 74.82%, and the optimal AUC is obtained. The performance of the algorithm is evaluated according to the number of features. When using 2 features, Bagging has AUC = 0.819 4 and the accuracy is 73.01%. Compared with traditional methods, this method has the advantage of continuously monitoring the development of patients with ARDS and providing medical staff with auxiliary diagnosis suggestions.
Subject(s)
Humans , Algorithms , Area Under Curve , Blood Gas Analysis , Machine Learning , Monitoring, Physiologic , Methods , ROC Curve , Respiratory Distress Syndrome , Diagnosis , Sensitivity and SpecificityABSTRACT
Antecedentes. La psoriasis es una enfermedad inflamatoria crónica multisistémica que implica un riesgo cardiovascular aumentado, incluidos enfermedad coronaria, infarto de miocardio y muerte de causa cardiovascular, sobre todo en pacientes jóvenes y con psoriasis más graves. Esto se debe a una aterogénesis precoz y en ocasiones subclínica, que podría ser identificada de manera no invasiva mediante estudios vasculares de vasos periféricos. Objetivos. a) identificar la presencia de ateroesclerosis subclínica en pacientes con psoriasis; b) establecer su valor predictivo independientemente de otros factores de riesgo cardiovascular. Diseño. Estudio de cohorte, prospectivo, controlado. Métodos. Se incluyeron 175 pacientes consecutivos, un grupo con psoriasis (GP, n:35) y un grupo control (GC, n:140), a quienes se les efectuó estudios de identificación de placas de ateroesclerosis y de elasticidad arterial en vasos carotídeos extracraneales y femorales. Resultados. El score de Framingham en los pacientes de ambos grupos fue bajo (7 ± 2,3% vs. 5,7 ± 1,8%; GP vs. GC respectivamente; p = .003), mientras que el score de riesgo vascular determinado según la alteración de parámetros de ateroesclerosis subclínica evaluados fue mayor en el GP (2,9 ± 1,2 vs. 2,2 ± 0.08; p = .002). La rigidez de la pared arterial se halló significativamente incrementada en el GP (EIM 0,7 ± 0,2 vs. 0,63 ± 0,1 mm, p< 0.001), donde se observó una prevalencia mayor de placas ateroescleróticas (94% vs. 62,5%, p = .001) la mayoría de alta vulnerabilidad. Conclusiones. El uso de técnicas no invasivas que faciliten la detección precoz de pacientes psoriásicos con enfermedad subclínica de la pared arterial, permitiría evaluar correctamente el riesgo, que puede ser subestimado si sólo se realizan estudios clínicos de los factores de riesgoconvencionales.
Background. Psoriasis is a chronic multisystem inflammatory disease that involves an increasedcardiovascular risk as heart disease, myocardial infarction and cardiovascular death, especially inyounger patients and severe psoriasis. This is due to early atherogenesis, and sometimes to a subclinicalcourse, which could be identified by non invasive vascular studies of peripheral vessels.Objectives. a) Identify the presence of subclinical atherosclerosis in patients with psoriasis; b)Determine its predictive value independently of other cardiovascular risk factors.Design. Cohort, prospective, controlled trial. Methods. 175 consecutive patients were included, a Psoriatic Group (PG, n:35) and a Control Group(CG, n:140) where identified, and submited to studies for identification of atherosclerotic plaquesand arterial elasticity in extracranial carotids and femoral vessels.Results. The Framingham scores in patients of both groups was low (7 ± 2.3% vs. 5.7 ± 1.8%;GP vs. GC respectively, p = .003) while the vascular risk score, determined by the alteration ofsubclinical atherosclerosis parameters, was higher in the GP (2.9 ± 1.2 vs. 2.2 ± 0.08, p = .002).The arterial wall stiffness was found significantly increased in the GP (EIM 0.7 ± 0.2 vs 0.63 ± 0.1mm, p< 0.001), where a higher prevalence of atherosclerotic plaques (94% vs. 62.5%, p = .001),mostly of high vulnerability, was also observed.Conclusions. The use of non invasive techniques that facilitate early detection of psoriaticindividuals with subclinical arterial wall disease, would allow a proper assessment of risk whichmay be underestimates by only clinical assessment of conventional risk factors.