Comparative assessment of synthetic time series generation approaches in healthcare: leveraging patient metadata for accurate data synthesis.

BACKGROUND: Synthetic data is an emerging approach for addressing legal and regulatory concerns in biomedical research that deals with personal and clinical data, whether as a single tool or through its combination with other privacy enhancing technologies. Generating uncompromised synthetic data could significantly benefit external researchers performing secondary analyses by providing unlimited access to information while fulfilling pertinent regulations. However, the original data to be synthesized (e.g., data acquired in Living Labs) may consist of subjects' metadata (static) and a longitudinal component (set of time-dependent measurements), making it challenging to produce coherent synthetic counterparts. METHODS: Three synthetic time series generation approaches were defined and compared in this work: only generating the metadata and coupling it with the real time series from the original data (A1), generating both metadata and time series separately to join them afterwards (A2), and jointly generating both metadata and time series (A3). The comparative assessment of the three approaches was carried out using two different synthetic data generation models: the Wasserstein GAN with Gradient Penalty (WGAN-GP) and the DöppelGANger (DGAN). The experiments were performed with three different healthcare-related longitudinal datasets: Treadmill Maximal Effort Test (TMET) measurements from the University of Malaga (1), a hypotension subset derived from the MIMIC-III v1.4 database (2), and a lifelogging dataset named PMData (3). RESULTS: Three pivotal dimensions were assessed on the generated synthetic data: resemblance to the original data (1), utility (2), and privacy level (3). The optimal approach fluctuates based on the assessed dimension and metric. CONCLUSION: The initial characteristics of the datasets to be synthesized play a crucial role in determining the best approach. Coupling synthetic metadata with real time series (A1), as well as jointly generating synthetic time series and metadata (A3), are both competitive methods, while separately generating time series and metadata (A2) appears to perform more poorly overall.

Synthetic Tabular Data Evaluation in the Health Domain Covering Resemblance, Utility, and Privacy Dimensions.

BACKGROUND: Synthetic tabular data generation is a potentially valuable technology with great promise for data augmentation and privacy preservation. However, prior to adoption, an empirical assessment of generated synthetic tabular data is required across dimensions relevant to the target application to determine its efficacy. A lack of standardized and objective evaluation and benchmarking strategy for synthetic tabular data in the health domain has been found in the literature. OBJECTIVE: The aim of this paper is to identify key dimensions, per dimension metrics, and methods for evaluating synthetic tabular data generated with different techniques and configurations for health domain application development and to provide a strategy to orchestrate them. METHODS: Based on the literature, the resemblance, utility, and privacy dimensions have been prioritized, and a collection of metrics and methods for their evaluation are orchestrated into a complete evaluation pipeline. This way, a guided and comparative assessment of generated synthetic tabular data can be done, categorizing its quality into three categories ("Excellent," "Good," and "Poor"). Six health care-related datasets and four synthetic tabular data generation approaches have been chosen to conduct an analysis and evaluation to verify the utility of the proposed evaluation pipeline. RESULTS: The synthetic tabular data generated with the four selected approaches has maintained resemblance, utility, and privacy for most datasets and synthetic tabular data generation approach combination. In several datasets, some approaches have outperformed others, while in other datasets, more than one approach has yielded the same performance. CONCLUSION: The results have shown that the proposed pipeline can effectively be used to evaluate and benchmark the synthetic tabular data generated by various synthetic tabular data generation approaches. Therefore, this pipeline can support the scientific community in selecting the most suitable synthetic tabular data generation approaches for their data and application of interest.

Spatial characterization of the effect of age and sex on macular layer thicknesses and foveal pit morphology.

Characterizing the effect of age and sex on macular retinal layer thicknesses and foveal pit morphology is crucial to differentiating between natural and disease-related changes. We applied advanced image analysis techniques to optical coherence tomography (OCT) to: 1) enhance the spatial description of age and sex effects, and 2) create a detailed open database of normative retinal layer thickness maps and foveal pit shapes. The maculae of 444 healthy subjects (age range 21-88) were imaged with OCT. Using computational spatial data analysis, thickness maps were obtained for retinal layers and averaged into 400 (20 x 20) sectors. Additionally, the geometry of the foveal pit was radially analyzed by computing the central foveal thickness, rim height, rim radius, and mean slope. The effect of age and sex on these parameters was analyzed with multiple regression mixed-effects models. We observed that the overall age-related decrease of the total retinal thickness (TRT) (-1.1% per 10 years) was mainly driven by the ganglion cell-inner plexiform layer (GCIPL) (-2.4% per 10 years). Both TRT and GCIPL thinning patterns were homogeneous across the macula when using percentual measurements. Although the male retina was 4.1 µm thicker on average, the greatest differences were mainly present for the inner retinal layers in the inner macular ring (up to 4% higher TRT than in the central macula). There was an age-related decrease in the rim height (1.0% per 10 years) and males had a higher rim height, shorter rim radius, and steeper mean slope. Importantly, the radial analysis revealed that these changes are present and relatively uniform across angular directions. These findings demonstrate the capacity of advanced analysis of OCT images to enhance the description of the macula. This, together with the created dataset, could aid the development of more accurate diagnosis models for macular pathologies.

Estimation of the epidemiology of dementia and associated neuropsychiatric symptoms by applying machine learning to real-world data.

INTRODUCTION: Incidence rates of dementia-related neuropsychiatric symptoms (NPS) are not known and this hampers the assessment of their population burden. The objective of this study was to obtain an approximate estimate of the population incidence and prevalence of both dementia and NPS. METHODS: Given the dynamic nature of the population with dementia, a retrospective study was conducted within the database of the Basque Health Service (real-world data) at the beginning and end of 2019. Validated random forest models were used to identify separately depressive and psychotic clusters according to their presence in the electronic health records of all patients diagnosed with dementia. RESULTS: Among the 631,949 individuals over 60 years registered, 28,563 were diagnosed with dementia, of whom 15,828 (55.4%) showed psychotic symptoms and 19,461 (68.1%) depressive symptoms. The incidence of dementia in 2019 was 6.8/1000 person-years. Most incident cases of depressive (72.3%) and psychotic (51.9%) NPS occurred in cases of incident dementia. The risk of depressive-type NPS grows with years since dementia diagnosis, living in a nursing home, and female sex, but falls with older age. In the psychotic cluster model, the effects of male sex, and older age are inverted, both increasing the probability of this type of symptoms. CONCLUSIONS: The stigmatization factor conditions the social and attitudinal environment, delaying the diagnosis of dementia, preventing patients from receiving adequate care and exacerbating families' suffering. This study evidences the synergy between big data and real-world data for psychiatric epidemiological research.

Estimation of the epidemiology of dementia and associated neuropsychiatric symptoms by applying machine learning to real-world data.

INTRODUCTION: Incidence rates of dementia-related neuropsychiatric symptoms (NPS) are not known and this hampers the assessment of their population burden. The objective of this study was to obtain an approximate estimate of the population incidence and prevalence of both dementia and NPS. METHODS: Given the dynamic nature of the population with dementia, a retrospective study was conducted within the database of the Basque Health Service (real-world data) at the beginning and end of 2019. Validated random forest models were used to identify separately depressive and psychotic clusters according to their presence in the electronic health records of all patients diagnosed with dementia. RESULTS: Among the 631,949 individuals over 60 years registered, 28,563 were diagnosed with dementia, of whom 15,828 (55.4%) showed psychotic symptoms and 19,461 (68.1%) depressive symptoms. The incidence of dementia in 2019 was 6.8/1000 person-years. Most incident cases of depressive (72.3%) and psychotic (51.9%) NPS occurred in cases of incident dementia. The risk of depressive-type NPS grows with years since dementia diagnosis, living in a nursing home, and female sex, but falls with older age. In the psychotic cluster model, the effects of male sex, and older age are inverted, both increasing the probability of this type of symptoms. CONCLUSIONS: The stigmatization factor conditions the social and attitudinal environment, delaying the diagnosis of dementia, preventing patients from receiving adequate care and exacerbating families' suffering. This study evidences the synergy between big data and real-world data for psychiatric epidemiological research.

Validation of Random Forest Machine Learning Models to Predict Dementia-Related Neuropsychiatric Symptoms in Real-World Data.

BACKGROUND: Neuropsychiatric symptoms (NPS) are the leading cause of the social burden of dementia but their role is underestimated. OBJECTIVE: The objective of the study was to validate predictive models to separately identify psychotic and depressive symptoms in patients diagnosed with dementia using clinical databases representing the whole population to inform decision-makers. METHODS: First, we searched the electronic health records of 4,003 patients with dementia to identify NPS. Second, machine learning (random forest) algorithms were applied to build separate predictive models for psychotic and depressive symptom clusters in the training set (N = 3,003). Third, calibration and discrimination were assessed in the test set (N = 1,000) to assess the performance of the models. RESULTS: Neuropsychiatric symptoms were noted in the electronic health record of 58% of patients. The area under the receiver operating curve reached 0.80 for the psychotic cluster model and 0.74 for the depressive cluster model. The Kappa index and accuracy also showed better discrimination in the psychotic model. Calibration plots indicated that both types of model had less predictive accuracy when the probability of neuropsychiatric symptoms was <25%. The most important variables in the psychotic cluster model were use of risperidone, level of sedation, use of quetiapine and haloperidol and the number of antipsychotics prescribed. In the depressive cluster model, the most important variables were number of antidepressants prescribed, escitalopram use, level of sedation, and age. CONCLUSION: Given their relatively good performance, the predictive models can be used to estimate prevalence of NPS in population databases.

Aging Modulates the Resting Brain after a Memory Task: A Validation Study from Multivariate Models.

Recent work has demonstrated that aging modulates the resting brain. However, the study of these modulations after cognitive practice, resulting from a memory task, has been scarce. This work aims at examining age-related changes in the functional reorganization of the resting brain after cognitive training, namely, neuroplasticity, by means of the most innovative tools for data analysis. To this end, electroencephalographic activity was recorded in 34 young and 38 older participants. Different methods for data analyses, including frequency, time-frequency and machine learning-based prediction models were conducted. Results showed reductions in Alpha power in old compared to young adults in electrodes placed over posterior and anterior areas of the brain. Moreover, young participants showed Alpha power increases after task performance, while their older counterparts exhibited a more invariant pattern of results. These results were significant in the 140-160 s time window in electrodes placed over anterior regions of the brain. Machine learning analyses were able to accurately classify participants by age, but failed to predict whether resting state scans took place before or after the memory task. These findings greatly contribute to the development of multivariate tools for electroencephalogram (EEG) data analysis and improve our understanding of age-related changes in the functional reorganization of the resting brain.

Smart Home-Based Prediction of Multidomain Symptoms Related to Alzheimer's Disease.

As members of an increasingly aging society, one of our major priorities is to develop tools to detect the earliest stage of age-related disorders such as Alzheimer's Disease (AD). The goal of this paper is to evaluate the possibility of using unobtrusively collected activity-aware smart home behavior data to detect the multimodal symptoms that are often found to be impaired in AD. After gathering longitudinal smart home data for 29 older adults over an average duration of 2 years, we automatically labeled the data with corresponding activity classes and extracted time-series statistics containing ten behavioral features. Mobility, cognition, and mood were evaluated every six months. Using these data, we created regression models to predict symptoms as measured by the tests and a feature selection analysis was performed. Classification models were built to detect reliable absolute changes in the scores predicting symptoms and SmoteBOOST and wRACOG algorithms were used to overcome class imbalance where needed. Results show that all mobility, cognition, and depression symptoms can be predicted from activity-aware smart home data. Similarly, these data can be effectively used to predict reliable changes in mobility and memory skills. Results also suggest that not all behavioral features contribute equally to the prediction of every symptom. Future work therefore can improve model sensitivity by including additional longitudinal data and by further improving strategies to extract relevant features and address class imbalance. The results presented herein contribute toward the development of an early change detection system based on smart home technology.

On the early diagnosis of Alzheimer's Disease from multimodal signals: A survey.

INTRODUCTION: The number of Alzheimer's Disease (AD) patients is increasing with increased life expectancy and 115.4 million people are expected to be affected in 2050. Unfortunately, AD is commonly diagnosed too late, when irreversible damages have been caused in the patient. OBJECTIVE: An automatic, continuous and unobtrusive early AD detection method would be required to improve patients' life quality and avoid big healthcare costs. Thus, the objective of this survey is to review the multimodal signals that could be used in the development of such a system, emphasizing on the accuracy that they have shown up to date for AD detection. Some useful tools and specific issues towards this goal will also have to be reviewed. METHODS: An extensive literature review was performed following a specific search strategy, inclusion criteria, data extraction and quality assessment in the Inspec, Compendex and PubMed databases. RESULTS: This work reviews the extensive list of psychological, physiological, behavioural and cognitive measurements that could be used for AD detection. The most promising measurements seem to be magnetic resonance imaging (MRI) for AD vs control (CTL) discrimination with an 98.95% accuracy, while electroencephalogram (EEG) shows the best results for mild cognitive impairment (MCI) vs CTL (97.88%) and MCI vs AD distinction (94.05%). Available physiological and behavioural AD datasets are listed, as well as medical imaging analysis steps and neuroimaging processing toolboxes. Some issues such as "label noise" and multi-site data are discussed. CONCLUSIONS: The development of an unobtrusive and transparent AD detection system should be based on a multimodal system in order to take full advantage of all kinds of symptoms, detect even the smallest changes and combine them, so as to detect AD as early as possible. Such a multimodal system might probably be based on physiological monitoring of MRI or EEG, as well as behavioural measurements like the ones proposed along the article. The mentioned AD datasets and image processing toolboxes are available for their use towards this goal. Issues like "label noise" and multi-site neuroimaging incompatibilities may also have to be overcome, but methods for this purpose are already available.

Towards an automatic early stress recognition system for office environments based on multimodal measurements: A review.

Stress is a major problem of our society, as it is the cause of many health problems and huge economic losses in companies. Continuous high mental workloads and non-stop technological development, which leads to constant change and need for adaptation, makes the problem increasingly serious for office workers. To prevent stress from becoming chronic and provoking irreversible damages, it is necessary to detect it in its early stages. Unfortunately, an automatic, continuous and unobtrusive early stress detection method does not exist yet. The multimodal nature of stress and the research conducted in this area suggest that the developed method will depend on several modalities. Thus, this work reviews and brings together the recent works carried out in the automatic stress detection looking over the measurements executed along the three main modalities, namely, psychological, physiological and behavioural modalities, along with contextual measurements, in order to give hints about the most appropriate techniques to be used and thereby, to facilitate the development of such a holistic system.