Plausibility of Individual Decisions from Random Forests in Clinical Predictive Modelling Applications.

BACKGROUND: Machine learning algorithms are a promising approach to help physicians to deal with the ever increasing amount of data collected in healthcare each day. However, interpretation of suggestions derived from predictive models can be difficult. OBJECTIVES: The aim of this work was to quantify the influence of a specific feature on an individual decision proposed by a random forest (RF). METHODS: For each decision tree within the RF, the influence of each feature on a specific decision (FID) was quantified. For each feature, changes in outcome value due to the feature were summarized along the path. Results from all the trees in the RF were statistically merged. The ratio of FID to the respective feature's global importance was calculated (FIDrel). RESULTS: Global feature importance, FID and FIDrel significantly differed, depending on the individual input data. Therefore, we suggest to present the most important features as determined for FID and for FIDrel, whenever results of a RF are visualized. CONCLUSION: Feature influence on a specific decision can be quantified in RFs. Further studies will be necessary to evaluate our approach in a real world scenario.

Development of Multivariable Models to Predict and Benchmark Transfusion in Elective Surgery Supporting Patient Blood Management.

BACKGROUND: Blood transfusion is a highly prevalent procedure in hospitalized patients and in some clinical scenarios it has lifesaving potential. However, in most cases transfusion is administered to hemodynamically stable patients with no benefit, but increased odds of adverse patient outcomes and substantial direct and indirect cost. Therefore, the concept of Patient Blood Management has increasingly gained importance to pre-empt and reduce transfusion and to identify the optimal transfusion volume for an individual patient when transfusion is indicated. OBJECTIVES: It was our aim to describe, how predictive modeling and machine learning tools applied on pre-operative data can be used to predict the amount of red blood cells to be transfused during surgery and to prospectively optimize blood ordering schedules. In addition, the data derived from the predictive models should be used to benchmark different hospitals concerning their blood transfusion patterns. METHODS: 6,530 case records obtained for elective surgeries from 16 centers taking part in two studies conducted in 2004-2005 and 2009-2010 were analyzed. Transfused red blood cell volume was predicted using random forests. Separate models were trained for overall data, for each center and for each of the two studies. Important characteristics of different models were compared with one another. RESULTS: Our results indicate that predictive modeling applied prior surgery can predict the transfused volume of red blood cells more accurately (correlation coefficient cc = 0.61) than state of the art algorithms (cc = 0.39). We found significantly different patterns of feature importance a) in different hospitals and b) between study 1 and study 2. CONCLUSION: We conclude that predictive modeling can be used to benchmark the importance of different features on the models derived with data from different hospitals. This might help to optimize crucial processes in a specific hospital, even in other scenarios beyond Patient Blood Management.

Piloting the European Unified Patient Identity Management (EUPID) Concept to Facilitate Secondary Use of Neuroblastoma Data from Clinical Trials and Biobanking.

Data from two contexts, i.e. the European Unresectable Neuroblastoma (EUNB) clinical trial and results from comparative genomic hybridisation (CGH) analyses from corresponding tumour samples shall be provided to existing repositories for secondary use. Utilizing the European Unified Patient IDentity Management (EUPID) as developed in the course of the ENCCA project, the following processes were applied to the data: standardization (providing interoperability), pseudonymization (generating distinct but linkable pseudonyms for both contexts), and linking both data sources. The applied procedures resulted in a joined dataset that did not contain any identifiers that would allow to backtrack the records to either data sources. This provided a high degree of privacy to the involved patients as required by data protection regulations, without preventing proper analysis.

IT Infrastructure for Merging Data from Different Clinical Trials and Across Independent Research Networks.

Opsoclonus Myoclonus Syndrome (OMS) is a rare disease in children which is often associated with neuroblastoma and, therefore, requires treatment by pediatric neurologists and oncologists. The ongoing OMS trial investigates questions related to OMS and potentially underlying neuroblastomas. To support this trial with an adequate IT infrastructure, linkage of neuroblastoma research databases with the OMS electronic data capture (EDC) system was required. Therefore, an EDC system for the OMS trial was developed and integrated into the research infrastructure of the European Network for Cancer Research in Children and Adolescents (ENCCA) project. Application of ENNCA's pseudonymization concept enabled linkage of the OMS trial with neuroblastoma trials from two different scientific societies, while being compliant with current data protection regulations. Linkage of the neurological and the oncological domain could successfully be demonstrated and a promising concept for secondary use of the data of both domains has been developed, proofing the broad potential of the concepts for cross-domain research as promoted in the ENCCA project.

Integrated medication management in mHealth applications.

Continuous medication monitoring is essential for successful management of heart failure patients. Experiences with the recently established heart failure network HerzMobil Tirol show that medication monitoring limited to heart failure specific drugs could be insufficient, in particular for general practitioners. Additionally, some patients are confused about monitoring only part of their prescribed drugs. Sometimes medication will be changed without informing the responsible physician. As part of the upcoming Austrian electronic health record system ELGA, the eMedication system will collect prescription and dispensing data of drugs and these data will be accessible to authorized healthcare professionals on an inter-institutional level. Therefore, we propose two concepts on integrated medication management in mHealth applications that integrate ELGA eMedication and closed-loop mHealth-based telemonitoring. As a next step, we will implement these concepts and analyze--in a feasibility study--usability and practicability as well as legal aspects with respect to automatic data transfer from the ELGA eMedication service.