Abstract Diabetes mellitus type 2 (DMT2) is a severe and complex health problem. It is the most common type of diabetes. DMT2 is a chronic metabolic disorder that affects the way your body metabolizes sugar. With DMT2, your body either resists the effects of insulin or does not produce sufficient insulin to continue normal glucose levels. DMT2 is a disease that requires a multifactorial approach of controlling that includes lifestyle change and pharmacotherapy. Less than ideal management increases the risk of developing complications and comorbidities such as cardiovascular disease and numerous social and economic penalties. That is why the studies dedicated to the pathophysiological mechanisms and the treatment of DMT2 are extremely numerous and diverse. In this study, exploratory data analysis approaches are applied for the treatment of clinical and anthropometric readings of patients with DMT2. Since multivariate statistics is a well-known method for classification, modeling and interpretation of large collections of data, the major aim of the present study was to reveal latent relations between the objects of the investigation (group of patients and control group) and the variables describing the objects (clinical and anthropometric parameters). In the proposed method by the application of hierarchical cluster analysis and principal component analysis it is possible to identify reduced number of parameters which appear to be the most significant discriminant parameters to distinguish between four patterns of patients with DMT2. However, there is still lack of multivariate statistical studies using DMT2 data sets to assess different aspects of the problem like optimal rapid monitoring of the patients or specific separation of patients into patterns of similarity related to their health status which could be of help in preparation of data bases for DMT2 patients. The outcome from the study could be of custom for the selection of significant tests for rapid monitoring of patients and more detailed approach to the health status of DMT2 patients.

中文翻译：

---

2型糖尿病：基于临床读数的探索性数据分析

摘要 2 型糖尿病 (DMT2) 是一种严重且复杂的健康问题。它是最常见的糖尿病类型。DMT2 是一种慢性代谢紊乱，会影响您的身体代谢糖的方式。使用 DMT2，您的身体要么抵抗胰岛素的作用，要么无法产生足够的胰岛素来维持正常的血糖水平。DMT2 是一种需要多因素控制方法的疾病，包括生活方式改变和药物治疗。不理想的管理会增加出现并发症和合并症的风险，例如心血管疾病以及许多社会和经济惩罚。这就是为什么致力于病理生理学机制和 DMT2 治疗的研究极其众多且多种多样。在这项研究中，探索性数据分析方法用于治疗 DMT2 患者的临床和人体测量读数。由于多元统计是对大量数据进行分类、建模和解释的众所周知的方法，因此本研究的主要目的是揭示调查对象（患者组和对照组）与变量之间的潜在关系描述对象（临床和人体测量参数）。在所提出的方法中，通过应用层次聚类分析和主成分分析，可以识别数量减少的参数，这些参数似乎是区分 DMT2 患者的四种模式的最重要的判别参数。然而，仍然缺乏使用 DMT2 数据集来评估问题的不同方面的多变量统计研究，例如对患者进行最佳快速监测或将患者具体分为与其健康状况相关的相似模式，这可能有助于准备数据库用于 DMT2 患者。该研究的结果可以用于选择重要的测试以快速监测患者和更详细地了解 DMT2 患者的健康状况。