vivaGen - a survival data set generator for software testing.

An amendment to this paper has been published and can be accessed via the original article.

Methods for a similarity measure for clinical attributes based on survival data analysis.

BACKGROUND: Case-based reasoning is a proven method that relies on learned cases from the past for decision support of a new case. The accuracy of such a system depends on the applied similarity measure, which quantifies the similarity between two cases. This work proposes a collection of methods for similarity measures especially for comparison of clinical cases based on survival data, as they are available for example from clinical trials. METHODS: Our approach is intended to be used in scenarios, where it is of interest to use longitudinal data, such as survival data, for a case-based reasoning approach. This might be especially important, where uncertainty about the ideal therapy decision exists. The collection of methods consists of definitions of the local similarity of nominal as well as numeric attributes, a calculation of attribute weights, a feature selection method and finally a global similarity measure. All of them use survival time (consisting of survival status and overall survival) as a reference of similarity. As a baseline, we calculate a survival function for each value of any given clinical attribute. RESULTS: We define the similarity between values of the same attribute by putting the estimated survival functions in relation to each other. Finally, we quantify the similarity by determining the area between corresponding curves of survival functions. The proposed global similarity measure is designed especially for cases from randomized clinical trials or other collections of clinical data with survival information. Overall survival can be considered as an eligible and alternative solution for similarity calculations. It is especially useful, when similarity measures that depend on the classic solution-describing attribute "applied therapy" are not applicable. This is often the case for data from clinical trials containing randomized arms. CONCLUSIONS: In silico evaluation scenarios showed that the mean accuracy of biomarker detection in k = 10 most similar cases is higher (0.909-0.998) than for competing similarity measures, such as Heterogeneous Euclidian-Overlap Metric (0.657-0.831) and Discretized Value Difference Metric (0.535-0.671). The weight calculation method showed a more than six times (6.59-6.95) higher weight for biomarker attributes over non-biomarker attributes. These results suggest that the similarity measure described here is suitable for applications based on survival data.

Tocolysis with the ß2-sympathomimetic fenoterol does not increase the occurrence of infantile hemangioma in preterm and term infants.

PURPOSE: ß2-sympathomimetics are used in obstetrics as tocolytic agents, despite a remarkable profile of side effects. Recently, the ß2-sympathomimetic tocolytic drug hexoprenaline was identified as an independent risk factor for the development of infantile hemangioma (IH) in preterm infants. The aim of this study was to evaluate whether this observed effect was applicable to other ß2-mimetic tocolytic agents like fenoterol. METHODS: Clinical prospectively collected data of all infants born between 2001 and 2012 and admitted to the neonatal intensive care unit (NICU) at Heidelberg University Hospital and respective maternal data were merged. For the current retrospective cohort study, cases (IH) were matched to controls (no IH) at a ratio of 1:4, adjusting for birth weight, gestational age, gender and multiple gestations. Prenatal exposure to fenoterol and perinatal outcome were analyzed in the total cohort and in subgroups. RESULTS: N = 5070 infants were admitted to our neonatal department, out of which n = 172 infants with IH were identified and compared to n = 596 matched controls. Exposure to fenoterol was not associated with a higher rate of IH in the total matched population (OR 0.926, 95% CI 0.619-1.384) or in a subgroup of neonates < 32 weeks of gestation or with a birth weight < 1500 g (OR 1.127, 95% CI 0.709-1.791). In the total matched population, prenatal exposure to glucocorticoids was associated with a reduced occurrence of IH (OR 0.566, 95% CI 0.332-0.964) and neonates with IH showed a prolonged total hospital stay compared to controls (69 vs. 57 days, p = 0.0033). Known risk factors for IH were confirmed by our large study cohort and included female gender, low birth weight, preterm birth and multiple gestations (all p < 0.005). CONCLUSIONS: Exposure to fenoterol during pregnancy does not increase the occurrence of IH. Further studies are needed to explore differences in the risk profiles of different ß2-sympathomimetic tocolytic drugs.

Implementing Systems Medicine: A Medical Informatics Perspective.

Systems medicine is a paradigm for translating in silico methods developed for modelling biological systems into the field of medicine. Such approaches rely on the integration of as many data sources as possible, both in the dimension of disease knowledge and patient data. This is a challenging task that can only be implemented in clinical routine with the help of suitable information technology from the field of Medical Informatics. For the research project "Clinically-applicable, omics-based assessment of survival, side effects, and targets in multiple myeloma" (CLIOMMICS) we developed a prototypical systems medicine application system. It is based on a three-level-architecture covering data representation, decision support, and user interface. The core decision support component is implemented as a case-based reasoning engine. However, the architecture follows a modular design that allows to replace individual components as needed.

Systems Medicine for Multiple Myeloma: A Review on Decision Support Systems.

Systems medicine is a current approach trying to improve treatment for patients with complex diseases by analyzing as much phenotype and genotype data as possible for the disease in question. For individualized treatment decisions in clinical practice, this task has to be supported by an application system with decision support component. For a research project on systems medicine we reviewed methods for decision support. Criteria for selecting a method are derived from characteristics of the data and the diseases. They include, among others: dimensionality of data and existence of a priori models for diseases. As a result we decided to implement a prototype system with a case-based reasoning component for systems medicine on multiple myeloma.

An IT Architecture for Systems Medicine.

Systems medicine aims to support treatment of complex diseases like cancer by integrating all available data for the disease. To provide such a decision support in clinical practice, a suitable IT architecture is necessary. We suggest a generic architecture comprised of the following three layers: data representation, decision support, and user interface. For the systems medicine research project "Clinically-applicable, omics-based assessment of survival, side effects, and targets in multiple myeloma" (CLIOMMICS) we developed a concrete instance of the generic architecture. We use i2b2 for representing the harmonized data. Since no deterministic model exists for multiple myeloma we use case-based reasoning for decision support. For clinical practice, visualizations of the results must be intuitive and clear. At the same time, they must communicate the uncertainty immanent in stochastic processes. Thus, we develop a specific user interface for systems medicine based on the web portal software Liferay.

Screening Internet forum participants for depression symptoms by assembling and enhancing multiple NLP methods.

Depression is a disease that can dramatically lower quality of life. Symptoms of depression can range from temporary sadness to suicide. Embarrassment, shyness, and the stigma of depression are some of the factors preventing people from getting help for their problems. Contemporary social media technologies like Internet forums or micro-blogs give people the opportunity to talk about their feelings in a confidential anonymous environment. However, many participants in such networks may not recognize the severity of their depression and their need for professional help. Our approach is to develop a method that detects symptoms of depression in free text, such as posts in Internet forums, chat rooms and the like. This could help people appreciate the significance of their depression and realize they need to seek help. In this work Natural Language Processing methods are used to break the textual information into its grammatical units. Further analysis involves detection of depression symptoms and their frequency with the help of words known as indicators of depression and their synonyms. Finally, similar to common paper-based depression scales, e.g., the CES-D, that information is incorporated into a single depression score. In this evaluation study, our depressive mood detection system, DepreSD (Depression Symptom Detection), had an average precision of 0.84 (range 0.72-1.0 depending on the specific measure) and an average F measure of 0.79 (range 0.72-0.9).

A framework for integrating heterogeneous clinical data for a disease area into a central data warehouse.

Structured collection of clinical facts is a common approach in clinical research. Especially in the analysis of rare diseases it is often necessary to aggregate study data from several sites in order to achieve a statistically significant cohort size. In this paper we describe a framework how to approach an integration of heterogeneous clinical data into a central register. This enables site-spanning queries for the occurrence of specific clinical facts and thus supports clinical research. The framework consists of three sequential steps, starting from a formal data harmonization process, to the data transformation methods and finally the integration into a proper data warehouse. We implemented reusable software templates that are based on our best practices in several projects in integrating heterogeneous clinical data. Our methods potentially increase the efficiency and quality for future data integration projects by reducing the implementation effort as well as the project management effort by usage of our approaches as a guideline.