The objective of this paper is to present an interactive, user-friendly and graphically oriented toolbox for obesity classification. A software tool for data management, design and analysis could help physicians, not only in a preliminary classification of patient typologies, but also in selecting non redundant clinical trials and overall in predicting for each class the effects of therapies. This goal has been set joining the experience of a medical team, expert in obesity treatments and researchers in the field of automatic control and data analysis. Obefix toolbox uses different standard statistical functions and uses PCA (Principal Component Analysis) for reducing data dimensionality. Users can choice between hierarchical clustering and k-means clustering method, for classification of patients. Clustering methods may be applied to data, after a PC separation. After a short description of the software, a clinical example of Obefix effectiveness in medical applications is presented: the case study was a dataset of patients affected by obesity and candidate for gastric bariatric surgery. Data were collected using Minnesota Multiphasic Personality Inventory (MMPI) test; one of depends both on genetic and behavioural factors. Secondary obesity has a lower frequency (< 5%) and it appears as a consequence of a primary pathology (e.g., metabolic and hormonal disorders, diabetes and so on). Since obesity has a multi-factorial aetiology, many different clinical data may be gathered, analysed and integrated for classifying different typologies of essential obesity; unfortunately, due to the complexity of this pathology, a standard classification is far from a general definition and it is difficult to formulate effective treatment strategies. A typical example showing difficulties in predicting the effects of a therapy is the study of the loss of weight for patients submitted to adjustable. gastric banding surgery; a high level of uncertainty does not allow predicting long term effects. This insecure prediction does not indicate a certain the most frequently used personality tests in the mental health fields. Three main homogeneous clusters were identified (patients with working problems, with antisocial personality disorder and with obsessive-compulsive disorder) showing a strict correlation between individuals belonging to each cluster and variations of the body mass index (BMI) 6 months after surgery.
Obefix: a Friendly Toolbox for Obesity Classification
Landi A;Piaggi P;Santini F;
2007-01-01
Abstract
The objective of this paper is to present an interactive, user-friendly and graphically oriented toolbox for obesity classification. A software tool for data management, design and analysis could help physicians, not only in a preliminary classification of patient typologies, but also in selecting non redundant clinical trials and overall in predicting for each class the effects of therapies. This goal has been set joining the experience of a medical team, expert in obesity treatments and researchers in the field of automatic control and data analysis. Obefix toolbox uses different standard statistical functions and uses PCA (Principal Component Analysis) for reducing data dimensionality. Users can choice between hierarchical clustering and k-means clustering method, for classification of patients. Clustering methods may be applied to data, after a PC separation. After a short description of the software, a clinical example of Obefix effectiveness in medical applications is presented: the case study was a dataset of patients affected by obesity and candidate for gastric bariatric surgery. Data were collected using Minnesota Multiphasic Personality Inventory (MMPI) test; one of depends both on genetic and behavioural factors. Secondary obesity has a lower frequency (< 5%) and it appears as a consequence of a primary pathology (e.g., metabolic and hormonal disorders, diabetes and so on). Since obesity has a multi-factorial aetiology, many different clinical data may be gathered, analysed and integrated for classifying different typologies of essential obesity; unfortunately, due to the complexity of this pathology, a standard classification is far from a general definition and it is difficult to formulate effective treatment strategies. A typical example showing difficulties in predicting the effects of a therapy is the study of the loss of weight for patients submitted to adjustable. gastric banding surgery; a high level of uncertainty does not allow predicting long term effects. This insecure prediction does not indicate a certain the most frequently used personality tests in the mental health fields. Three main homogeneous clusters were identified (patients with working problems, with antisocial personality disorder and with obsessive-compulsive disorder) showing a strict correlation between individuals belonging to each cluster and variations of the body mass index (BMI) 6 months after surgery.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.