Predictive modeling for identification of older adults with high utilization of health and social services.

AIM: Machine learning techniques have demonstrated success in predictive modeling across various clinical cases. However, few studies have considered predicting the use of multisectoral health and social services among older adults. This research aims to utilize machine learning models to detect high-risk groups of excessive health and social services utilization at early stage, facilitating the implementation of preventive interventions. METHODS: We used pseudonymized data covering a four-year period and including information on a total of 33,374 senior citizens from Southern Finland. The endpoint was defined based on the occurrence of unplanned healthcare visits and the total number of different services used. Input features included individual's basic demographics, health status and past usage of healthcare resources. Logistic regression and eXtreme Gradient Boosting (XGBoost) methods were used for binary classification, with the dataset split into 70% training and 30% testing sets. RESULTS: Subgroup-based results mirrored trends observed in the full cohort, with age and certain health issues, e.g. mental health, emerging as positive predictors for high service utilization. Conversely, hospital stay and urban residence were associated with decreased risk. The models achieved a classification performance (AUC) of 0.61 for the full cohort and varying in the range of 0.55-0.62 for the subgroups. CONCLUSIONS: Predictive models offer potential for predicting future high service utilization in the older adult population. Achieving high classification performance remains challenging due to diverse contributing factors. We anticipate that classification performance could be increased by including features based on additional data categories such as socio-economic data.

Integrating data from multiple Finnish biobanks and national health-care registers for retrospective studies: Practical experiences.

Aim: This case study aimed to investigate the process of integrating resources of multiple biobanks and health-care registers, especially addressing data permit application, time schedules, co-operation of stakeholders, data exchange and data quality. Methods: We investigated the process in the context of a retrospective study: Pharmacogenomics of antithrombotic drugs (PreMed study). The study involved linking the genotype data of three Finnish biobanks (Auria Biobank, Helsinki Biobank and THL Biobank) with register data on medicine dispensations, health-care encounters and laboratory results. Results: We managed to collect a cohort of 7005 genotyped individuals, thereby achieving the statistical power requirements of the study. The data collection process took 16 months, exceeding our original estimate by seven months. The main delays were caused by the congested data permit approval service to access national register data on health-care encounters. Comparison of hospital data lakes and national registers revealed differences, especially concerning medication data. Genetic variant frequencies were in line with earlier data reported for the European population. The yearly number of international normalised ratio (INR) tests showed stable behaviour over time. Conclusions: A large cohort, consisting of versatile individual-level phenotype and genotype data, can be constructed by integrating data from several biobanks and health data registers in Finland. Co-operation with biobanks is straightforward. However, long time periods need to be reserved when biobank resources are linked with national register data. There is a need for efforts to define general, harmonised co-operation practices and data exchange methods for enabling efficient collection of data from multiple sources.

Functional brain segmentation using inter-subject correlation in fMRI.

The human brain continuously processes massive amounts of rich sensory information. To better understand such highly complex brain processes, modern neuroimaging studies are increasingly utilizing experimental setups that better mimic daily-life situations. A new exploratory data-analysis approach, functional segmentation inter-subject correlation analysis (FuSeISC), was proposed to facilitate the analysis of functional magnetic resonance (fMRI) data sets collected in these experiments. The method provides a new type of functional segmentation of brain areas, not only characterizing areas that display similar processing across subjects but also areas in which processing across subjects is highly variable. FuSeISC was tested using fMRI data sets collected during traditional block-design stimuli (37 subjects) as well as naturalistic auditory narratives (19 subjects). The method identified spatially local and/or bilaterally symmetric clusters in several cortical areas, many of which are known to be processing the types of stimuli used in the experiments. The method is not only useful for spatial exploration of large fMRI data sets obtained using naturalistic stimuli, but also has other potential applications, such as generation of a functional brain atlases including both lower- and higher-order processing areas. Finally, as a part of FuSeISC, a criterion-based sparsification of the shared nearest-neighbor graph was proposed for detecting clusters in noisy data. In the tests with synthetic data, this technique was superior to well-known clustering methods, such as Ward's method, affinity propagation, and K-means ++. Hum Brain Mapp 38:2643-2665, 2017. © 2017 Wiley Periodicals, Inc.

Multimodal Prediction of Five-Year Breast Cancer Recurrence in Women Who Receive Neoadjuvant Chemotherapy.

In current clinical practice, it is difficult to predict whether a patient receiving neoadjuvant chemotherapy (NAC) for breast cancer is likely to encounter recurrence after treatment and have the cancer recur locally in the breast or in other areas of the body. We explore the use of clinical history, immunohistochemical markers, and multiparametric magnetic resonance imaging (DCE, ADC, Dixon) to predict the risk of post-treatment recurrence within five years. We performed a retrospective study on a cohort of 1738 patients from Institut Curie and analyzed the data using classical machine learning, image processing, and deep learning. Our results demonstrate the ability to predict recurrence prior to NAC treatment initiation using each modality alone, and the possible improvement achieved by combining the modalities. When evaluated on holdout data, the multimodal model achieved an AUC of 0.75 (CI: 0.70, 0.80) and 0.57 specificity at 0.90 sensitivity. We then stratified the data based on known prognostic biomarkers. We found that our models can provide accurate recurrence predictions (AUC > 0.89) for specific groups of women under 50 years old with poor prognoses. A version of our method won second place at the BMMR2 Challenge, with a very small margin from being first, and was a standout from the other challenge entries.

System Architecture of a European Platform for Health Policy Decision Making: MIDAS.

Background: Healthcare data is a rich yet underutilized resource due to its disconnected, heterogeneous nature. A means of connecting healthcare data and integrating it with additional open and social data in a secure way can support the monumental challenge policy-makers face in safely accessing all relevant data to assist in managing the health and wellbeing of all. The goal of this study was to develop a novel health data platform within the MIDAS (Meaningful Integration of Data Analytics and Services) project, that harnesses the potential of latent healthcare data in combination with open and social data to support evidence-based health policy decision-making in a privacy-preserving manner. Methods: The MIDAS platform was developed in an iterative and collaborative way with close involvement of academia, industry, healthcare staff and policy-makers, to solve tasks including data storage, data harmonization, data analytics and visualizations, and open and social data analytics. The platform has been piloted and tested by health departments in four European countries, each focusing on different region-specific health challenges and related data sources. Results: A novel health data platform solving the needs of Public Health decision-makers was successfully implemented within the four pilot regions connecting heterogeneous healthcare datasets and open datasets and turning large amounts of previously isolated data into actionable information allowing for evidence-based health policy-making and risk stratification through the application and visualization of advanced analytics. Conclusions: The MIDAS platform delivers a secure, effective and integrated solution to deal with health data, providing support for health policy decision-making, planning of public health activities and the implementation of the Health in All Policies approach. The platform has proven transferable, sustainable and scalable across policies, data and regions.

Pharmacogenetics of Bleeding and Thromboembolic Events in Direct Oral Anticoagulant Users.

This study aimed to analyze associations between genetic variants and the occurrence of clinical outcomes in dabigatran, apixaban, and rivaroxaban users. This was a retrospective real-world study linking genotype data of three Finnish biobanks with national register data on drug dispensations and healthcare encounters. We investigated several single-nucleotide variants (SNVs) in the ABCG2, ABCB1, CES1, and CYP3A5 genes potentially associated with bleeding or thromboembolic events in direct oral anticoagulant (DOAC) users based on earlier research. We used Cox regression models to compare the incidence of clinical outcomes between carriers and noncarriers of the SNVs or haplotypes. In total, 1,806 patients on apixaban, dabigatran, or rivaroxaban were studied. The ABCB1 c.3435C>T (p.Ile1145=, rs1045642) SNV (hazard ratio (HR) 0.42, 95% confidence interval (CI), 0.18-0.98, P = 0.044) and 1236T-2677T-3435T (rs1128503-rs2032582-rs1045642) haplotype (HR 0.44, 95% CI, 0.20-0.95, P = 0.036) were associated with a reduced risk for thromboembolic outcomes, and the 1236C-2677G-3435C (HR 2.55, 95% CI, 1.03-6.36, P = 0.044) and 1236T-2677G-3435C (HR 5.88, 95% CI, 2.35-14.72, P < 0.001) haplotypes with an increased risk for thromboembolic outcomes in rivaroxaban users. The ABCB1 c.2482-2236G>A (rs4148738) SNV associated with a lower risk for bleeding events (HR 0.37, 95% CI, 0.16-0.89, P = 0.025) in apixaban users. ABCB1 variants are potential factors affecting thromboembolic events in rivaroxaban users and bleeding events in apixaban users. Studies with larger numbers of patients are warranted for comprehensive assessment of the pharmacogenetic associations of DOACs and their relevance for clinical practice.

Pharmacogenetics of Anticoagulation and Clinical Events in Warfarin-Treated Patients: A Register-Based Cohort Study with Biobank Data and National Health Registries in Finland.

PURPOSE: To assess the association between VKORC1 and CYP2C9 variants and the incidence of adverse drug reactions in warfarin-treated patients in a real-world setting. MATERIALS AND METHODS: This was a register-based cohort study (PreMed) linking data from Finnish biobanks, national health registries and patient records between January 1st 2007 and June 30th 2018. The inclusion criteria were: 1) ≥18 years of age, 2) CYP2C9 and VKORC1 genotype information available, 3) a diagnosis of a cardiovascular disease, 4) at least one warfarin purchase, 5) regular INR tests. Eligible individuals were divided into two warfarin sensitivity groups; normal responders, and sensitive and highly sensitive responders based on their VKORC1 and CYP2C9 genotypes. The incidences of clinical events were compared between the groups using Cox regression models. RESULTS: The cohort consisted of 2508 participants (45% women, mean age of 69 years), of whom 65% were categorized as normal responders and 35% sensitive or highly sensitive responders. Compared to normal responders, sensitive and highly sensitive responders had fewer INR tests below 2 (median: 33.3% vs 43.8%, 95% CI: -13.3%, -10.0%) and more above 3 (median: 18.2% vs 6.7%, 95% Cl: 8.3%, 10.8%). The incidence (per 100 patient-years) of bleeding outcomes was 5.4 for normal responders and 5.6 for the sensitive and highly sensitive responder group (HR=1.03, 95% CI: 0.74, 1.44). The incidence of thromboembolic outcomes was 4.9 and 7.8, respectively (HR=1.48, 95% CI: 1.08, 2.03). CONCLUSION: In a real-world setting, genetically sensitive and highly sensitive responders to warfarin had more high INR tests and required a lower daily dose of warfarin than normal responders. However, the risk for bleeding events was not increased in sensitive and highly sensitive responders. Interestingly, the risk of thromboembolic outcomes was lower in normal responders compared to the sensitive and highly sensitive responders. TRIAL REGISTRATION: NCT04001166.

Platform for systems medicine research and diagnostic applications in psychotic disorders-The METSY project.

Psychotic disorders are associated with metabolic abnormalities including alterations in glucose and lipid metabolism. A major challenge in the treatment of psychosis is to identify patients with vulnerable metabolic profiles who may be at risk of developing cardiometabolic co-morbidities. It is established that both central and peripheral metabolic organs use lipids to control energy balance and regulate peripheral insulin sensitivity. The endocannabinoid system, implicated in the regulation of glucose and lipid metabolism, has been shown to be dysregulated in psychosis. It is currently unclear how these endocannabinoid abnormalities relate to metabolic changes in psychosis. Here we review recent research in the field of metabolic co-morbidities in psychotic disorders as well as the methods to study them and potential links to the endocannabinoid system. We also describe the bioinformatics platforms developed in the EU project METSY for the investigations of the biological etiology in patients at risk of psychosis and in first episode psychosis patients. The METSY project was established with the aim to identify and evaluate multi-modal peripheral and neuroimaging markers that may be able to predict the onset and prognosis of psychiatric and metabolic symptoms in patients at risk of developing psychosis and first episode psychosis patients. Given the intrinsic complexity and widespread role of lipid metabolism, a systems biology approach which combines molecular, structural and functional neuroimaging methods with detailed metabolic characterisation and multi-variate network analysis is essential in order to identify how lipid dysregulation may contribute to psychotic disorders. A decision support system, integrating clinical, neuropsychological and neuroimaging data, was also developed in order to aid clinical decision making in psychosis. Knowledge of common and specific mechanisms may aid the etiopathogenic understanding of psychotic and metabolic disorders, facilitate early disease detection, aid treatment selection and elucidate new targets for pharmacological treatments.

How Many Is Enough? Effect of Sample Size in Inter-Subject Correlation Analysis of fMRI.

Inter-subject correlation (ISC) is a widely used method for analyzing functional magnetic resonance imaging (fMRI) data acquired during naturalistic stimuli. A challenge in ISC analysis is to define the required sample size in the way that the results are reliable. We studied the effect of the sample size on the reliability of ISC analysis and additionally addressed the following question: How many subjects are needed for the ISC statistics to converge to the ISC statistics obtained using a large sample? The study was realized using a large block design data set of 130 subjects. We performed a split-half resampling based analysis repeatedly sampling two nonoverlapping subsets of 10-65 subjects and comparing the ISC maps between the independent subject sets. Our findings suggested that with 20 subjects, on average, the ISC statistics had converged close to a large sample ISC statistic with 130 subjects. However, the split-half reliability of unthresholded and thresholded ISC maps improved notably when the number of subjects was increased from 20 to 30 or more.

Brain hemodynamic activity during viewing and re-viewing of comedy movies explained by experienced humor.

Humor is crucial in human social interactions. To study the underlying neural processes, three comedy clips were shown twice to 20 volunteers during functional magnetic resonance imaging (fMRI). Inter-subject similarities in humor ratings, obtained immediately after fMRI, explained inter-subject correlation of hemodynamic activity in right frontal pole and in a number of other brain regions. General linear model analysis also indicated activity in right frontal pole, as well as in additional cortical areas and subcortically in striatum, explained by humorousness. The association of the right frontal pole with experienced humorousness is a novel finding, which might be related to humor unfolding over longer time scales in the movie clips. Specifically, frontal pole has been shown to exhibit longer temporal receptive windows than, e.g., sensory areas, which might have enabled processing of humor in the clips based on holding information and reinterpreting that in light of new information several (even tens of) seconds later. As another novel finding, medial and lateral prefrontal areas, frontal pole, posterior-inferior temporal areas, posterior parietal areas, posterior cingulate, striatal structures and amygdala showed reduced activity upon re-viewing of the clips, suggesting involvement in processing of humor related to novelty of the comedic events.

Effects of spatial smoothing on inter-subject correlation based analysis of FMRI.

This study evaluates the effects of spatial smoothing on inter-subject correlation (ISC) analysis for FMRI data using the traditional model based analysis as a reference. So far within ISC analysis the effects of smoothing have not been studied systematically and linear Gaussian filters with varying kernel widths have been used without better knowledge about the effects of filtering. Instead, with the traditional general linear model (GLM) based analysis, the effects of smoothing have been studied extensively. In this study, ISC and GLM analyses were computed with two experimental and one simulated block-design datasets. The test statistics and the detected activation areas were compared numerically with correlation and Dice similarity measures, respectively. The study verified that (1) the choice of the filter substantially affected the activations detected by ISC analysis, (2) the detected activations according to ISC and GLM methods were highly similar regardless of the smoothing kernel and (3) the effect of spatial smoothing was mildly smaller on ISC than GLM analysis. Our results indicated that a good selection of the full width at half maximum of the Gaussian smoothing kernel for ISC was slightly larger than double the original voxel size.

A versatile software package for inter-subject correlation based analyses of fMRI.

In the inter-subject correlation (ISC) based analysis of the functional magnetic resonance imaging (fMRI) data, the extent of shared processing across subjects during the experiment is determined by calculating correlation coefficients between the fMRI time series of the subjects in the corresponding brain locations. This implies that ISC can be used to analyze fMRI data without explicitly modeling the stimulus and thus ISC is a potential method to analyze fMRI data acquired under complex naturalistic stimuli. Despite of the suitability of ISC based approach to analyze complex fMRI data, no generic software tools have been made available for this purpose, limiting a widespread use of ISC based analysis techniques among neuroimaging community. In this paper, we present a graphical user interface (GUI) based software package, ISC Toolbox, implemented in Matlab for computing various ISC based analyses. Many advanced computations such as comparison of ISCs between different stimuli, time window ISC, and inter-subject phase synchronization are supported by the toolbox. The analyses are coupled with re-sampling based statistical inference. The ISC based analyses are data and computation intensive and the ISC toolbox is equipped with mechanisms to execute the parallel computations in a cluster environment automatically and with an automatic detection of the cluster environment in use. Currently, SGE-based (Oracle Grid Engine, Son of a Grid Engine, or Open Grid Scheduler) and Slurm environments are supported. In this paper, we present a detailed account on the methods behind the ISC Toolbox, the implementation of the toolbox and demonstrate the possible use of the toolbox by summarizing selected example applications. We also report the computation time experiments both using a single desktop computer and two grid environments demonstrating that parallelization effectively reduces the computing time. The ISC Toolbox is available in https://code.google.com/p/isc-toolbox/

Inter-subject correlation in fMRI: method validation against stimulus-model based analysis.

Within functional magnetic resonance imaging (fMRI), the use of the traditional general linear model (GLM) based analysis methods is often restricted to strictly controlled research setups requiring a parametric activation model. Instead, Inter-Subject Correlation (ISC) method is based on voxel-wise correlation between the time series of the subjects, which makes it completely non-parametric and thus suitable for naturalistic stimulus paradigms such as movie watching. In this study, we compared an ISC based analysis results with those of a GLM based in five distinct controlled research setups. We used International Consortium for Brain Mapping functional reference battery (FRB) fMRI data available from the Laboratory of Neuro Imaging image data archive. The selected data included measurements from 37 right-handed subjects, who all had performed the same five tasks from FRB. The GLM was expected to locate activations accurately in FRB data and thus provide good grounds for investigating relationship between ISC and stimulus induced fMRI activation. The statistical maps of ISC and GLM were compared with two measures. The first measure was the Pearson's correlation between the non-thresholded ISC test-statistics and absolute values of the GLM Z-statistics. The average correlation value over five tasks was 0.74. The second was the Dice index between the activation regions of the methods. The average Dice value over the tasks and three threshold levels was 0.73. The results of this study indicated how the data driven ISC analysis found the same foci as the model-based GLM analysis. The agreement of the results is highly interesting, because ISC is applicable in situations where GLM is not suitable, for example, when analyzing data from a naturalistic stimuli experiment.