ABSTRACT
Bipolar Disorder (BD) is a chronic and disabling disease that usually appears around 20 to 30 years old. Patients who suffer with BD may struggle for years to achieve a correct diagnosis, and only 50% of them generally receive adequate treatment. In this work we apply a machine learning technique called Inductive Logic Programming (ILP) in order to model relapse and no-relapse patients in a first attempt in this area to improve diagnosis and optimize psychiatrists' time spent with patients. We use ILP because it is well suited for our multi-relational dataset and because a human can easily interpret the logical rules produced. Our classifiers can predict relapse cases with 92% Recall and no-relapse cases with 73% Recall. The rules and variable theories generated by ILP reproduce some findings from the scientific literature. The generated multi-relational models can be directly interpreted by clinicians and researchers, and also open space to research biological mechanisms and interventions.
Subject(s)
Bipolar Disorder/diagnosis , Decision Support Systems, Clinical , Depression/diagnosis , Diagnosis, Computer-Assisted/methods , Logistic Models , Machine Learning , Algorithms , Bipolar Disorder/complications , Computer Simulation , Depression/etiology , Humans , Recurrence , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
UNLABELLED: de Macedo-Soares MB, Brietzke E, da Silva Dias R, Mendonca T, Moreira C, Lafer B. A comparison of the symptomatic profile between two consecutive depressive episodes in patients with bipolar disorder type I. OBJECTIVE: To compare the variability of patterns of depressive symptoms between two consecutive depressive episodes in patients with bipolar disorder type I. METHODS: Review of prospectively collected data from 136 subjects of an out-patient bipolar unit from 1997 to 2007. Binomial statistics was used for the analysis of Hamilton Depression Rating Scale (HDRS)-31 items of the first and second episodes, and the correlation of the HDRS-31 item scores of both episodes was determined using the Spearman coefficient. RESULTS: Ten depressive symptoms showed a significant correlation between index and subsequent episodes: psychological anxiety, somatic anxiety, somatic symptoms, diurnal variation, paranoid symptoms, obsessive and compulsive symptoms, hypersomnia, loss of appetite and helplessness. Only four symptoms were stable in both statistical tests: paranoid symptoms, obsessive-compulsive symptoms, loss of appetite and hypersomnia. CONCLUSIONS: Paranoid and obsessive-compulsive symptoms, loss of appetite and hypersomnia tended to be found in successive episodes. However, the moderate correlations of the symptoms across two depressive recurrences suggested that clinical presentations in bipolar depression may not be predicted by symptom profiles presented in previous episodes.
ABSTRACT
OBJECTIVE: We aimed to determine the prevalence of obesity and metabolic syndrome (O/MetS) in a sample of Brazilian outpatients with bipolar disorder. METHODS: Eighty-four patients with bipolar disorder were evaluated. We used the definition of MetS established in the Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults, modified by the American Heart Association (AHA). Patients were classified as obese if their body mass index (BMI) was ≥ 30 kg/m2. RESULTS: We found that 28.6% of our sample met the AHA criteria for MetS and 35.7% were obese. The percentage of patients meeting each criterion of the AHA was as follows: 46% for abdominal obesity; 44% for hypertriglyceridemia or cholesterol-lowering medication use; 26% for low high-density lipoprotein cholesterol or being on a lipid-lowering medication; 45% for hypertension; and 20% for high fasting glucose or anti-diabetic medication use. CONCLUSIONS: The prevalence of obesity in our sample of outpatients with bipolar disorder was higher than that observed for the general population of Brazil. The rate of MetS was similar to that observed for the general population. Our data indicate the need for prevention, early detection and treatment of O/MetS in patients with bipolar disorder.
