Multimodal automatic assessment of acute pain through facial videos and heart rate signals utilizing transformer-based architectures.

Accurate and objective pain evaluation is crucial in developing effective pain management protocols, aiming to alleviate distress and prevent patients from experiencing decreased functionality. A multimodal automatic assessment framework for acute pain utilizing video and heart rate signals is introduced in this study. The proposed framework comprises four pivotal modules: the Spatial Module, responsible for extracting embeddings from videos; the Heart Rate Encoder, tasked with mapping heart rate signals into a higher dimensional space; the AugmNet, designed to create learning-based augmentations in the latent space; and the Temporal Module, which utilizes the extracted video and heart rate embeddings for the final assessment. The Spatial-Module undergoes pre-training on a two-stage strategy: first, with a face recognition objective learning universal facial features, and second, with an emotion recognition objective in a multitask learning approach, enabling the extraction of high-quality embeddings for the automatic pain assessment. Experiments with the facial videos and heart rate extracted from electrocardiograms of the BioVid database, along with a direct comparison to 29 studies, demonstrate state-of-the-art performances in unimodal and multimodal settings, maintaining high efficiency. Within the multimodal context, 82.74% and 39.77% accuracy were achieved for the binary and multi-level pain classification task, respectively, utilizing 9.62 million parameters for the entire framework.

FHBF: Federated hybrid boosted forests with dropout rates for supervised learning tasks across highly imbalanced clinical datasets.

Although several studies have deployed gradient boosting trees (GBT) as a robust classifier for federated learning tasks (federated GBT [FGBT]), even with dropout rates (federated gradient boosting trees with dropout rate [FDART]), none of them have investigated the overfitting effects of FGBT across heterogeneous and highly imbalanced datasets within federated environments nor the effect of dropouts in the loss function. In this work, we present the federated hybrid boosted forests (FHBF) algorithm, which incorporates a hybrid weight update approach to overcome ill-posed problems that arise from overfitting effects during the training across highly imbalanced datasets in the cloud. Eight case studies were conducted to stress the performance of FHBF against existing algorithms toward the development of robust AI models for lymphoma development across 18 European federated databases. Our results highlight the robustness of FHBF, yielding an average loss of 0.527 compared with FGBT (0.611) and FDART (0.584) with increased classification performance (0.938 sensitivity, 0.732 specificity).

Comparison of pulmonary and small airways function between idiopathic inflammatory myopathies patients with and without interstitial lung disease.

OBJECTIVES: To evaluate pulmonary and small airway function in patients with idiopathic inflammatory myopathies (IIM) and make comparisons between patients with and without interstitial lung disease (ILD). METHODS: Newly diagnosed IIM patients with and without ILD determined by high resolution computed tomography were included in the study. Pulmonary and small airway function was assessed by spirometry, diffusing capacity for carbon monoxide (DLCO), body plethysmography, single and multiple breath nitrogen washout, impulse oscillometry and measurement of respiratory resistance by the interrupter technique (Rint) using the Q-box system. We used discrepancies between lung volumes measured by multiple breath nitrogen washout and body plethysmography to evaluate for small airway dysfunction. RESULTS: Study cohort comprised of 26 IIM patients, 13 with and 13 without ILD. IIM-ILD patients presented more frequently with dyspnoea, fever, arthralgias and positive anti-synthetase antibodies, compared to IIM patients without ILD. Classic spirometric parameters and most lung physiology parameters assessing small airway function did not differ between the two groups. Predicted total lung capacity and residual volume (TLCN2WO, RVN2WO) measured by multiple breath nitrogen washout and the TLCN2WO/TLCpleth ratio were significantly lower in IIM-ILD patients compared to those without ILD (mean: 111.1% vs. 153.4%, p=0.034, median: 171% vs. 210%, p=0.039 and median: 1.28 vs. 1.45, p=0.039, respectively). Rint tended to be higher among IIM-ILD patients (mean:100.5% vs. 76.6%, p=0.053). CONCLUSIONS: Discrepancies between lung volumes measured by multiple breath nitrogen washout and body plethysmography in IIM-ILD patients indicate an early small airways dysfunction in these patients.

Boosting the performance of MALT lymphoma classification in patients with primary Sjögren's Syndrome through data augmentation: a case study.

Sjögren's Syndrome (SS) patients with mucosa associated lymphoid tissue lymphomas (MALTLs) and diffuse large B-cell lymphomas (DLBCLs) have 10-year survival rates of 80% and 40%, respectively. This highlights the unique biologic burden of the two histologic forms, as well as, the need for early detection and thorough monitoring of these patients. The lack of MALTL patients and the fact that most studies are single cohort and combine patients with different lymphoma subtypes narrow the understanding of MALTL progression. Here, we propose a data augmentation pipeline that utilizes an advanced synthetic data generator which is trained on a Pan European data hub with primary SS (pSS) patients to yield a high-quality synthetic data pool. The latter is used for the development of an enhanced MALTL classification model. Four scenarios were defined to assess the reliability of augmentation. Our results revealed an overall improvement in the accuracy, sensitivity, specificity, and AUC by 7%, 6.3%, 9%, and 6.3%, respectively. This is the first case study that utilizes data augmentation to reflect the progression of MALTL in pSS.

Visual Analogue Scale for the Evaluation of Olfactory and Gustatory Dysfunction of COVID-19 Patients in Northwestern Greece.

Background The visual analogue scale (VAS) has been used as a diagnostic tool for the evaluation of the severity of olfactory and gustatory dysfunction (OGD) caused by SARS-CoV2 infection. The main objective of the present study was the evaluation of OGD with VAS in COVID-19-positive patients in Northwestern Greece and its possible association with the patients' self-reported symptoms of olfactory and gustatory dysfunction. Methods The presence of olfactory and gustatory symptoms and their severity were assessed by questionnaire along with the use of specific odorants and tastant ingredients, in three time periods: prior to COVID-19, during COVID-19 (initial diagnosis) and post-COVID-19 disease (at four weeks from disease onset). Three hundred COVID-19-positive patients (home-quarantined and hospitalized) tested with RT-PCR test in the University Hospital of Ioannina Greece were included in this study. Statistical analysis was performed on SPSS Statistics 26.0 (IBM Corp., Armonk, NY) Results Out of a total of 300 patients, 146 and 190 patients had mild hyposmia and hypogeusia respectively, followed by patients with severe hyposmia or hypogeusia (118 and 88 respectively), at the time of COVID-19 onset (initial diagnosis). An increase in the number of patients with recovery of symptoms was observed during the follow-up period, during which only eight patients had non-resolving severe symptoms (six patients with hyposmia and two with hypogeusia). On further analysis, a statistically significant association was found between the severity of symptoms (assessed by VAS score) and the self-reported symptoms of sensory dysfunction by the patients. There was a significant association between the groups of patients with mild hyposmia and patients that reported no loss of smell; between the patients with moderate hyposmia and the patients who reported "loss of smell"; and between the patients with severe hyposmia and the group of patients who reported a loss of smell, at the COVID-19 onset period. Similarly, patients with mild hyposmia were associated with those that reported a loss of smell at the same time. The severity of hyposmia was also associated with the reported symptom of "loss of taste" at the time of COVID-19 diagnosis. Similar findings were observed regarding the severity of hypogeusia and the reported symptom of "loss of taste" among the groups of patients. Finally, the severity of hypogeusia was associated with smell loss at the time of initial diagnosis of the infection. Conclusion Similar to the literature data, our findings indicate that hyposmia and hypogeusia are common symptoms of COVID-19 disease with varying severity. In our study, most of the patients exerted a complete recovery of these OGD symptoms. In addition, we found an association between olfactory dysfunction and self-reported sensory of taste as well as gustatory dysfunction and sensory of smell. Finally, we found that the VAS score was a reliable diagnostic tool in the estimation of OGD in this cohort of patients. However, our results need to be confirmed by larger-scale trials.

An explainable machine learning-driven proposal of pulmonary fibrosis biomarkers.

Pulmonary fibrosing diseases are in the very epicenter of biomedical research both due to their increasing prevalence and their association with SARS-CoV-2 infections. Research of idiopathic pulmonary fibrosis, the most lethal among the interstitial lung diseases, is in need for new biomarkers and potential disease targets, a goal that could be accelerated using machine learning techniques. In this study, we have used Shapley values to explain the decisions made by an ensemble learning model trained to classify samples to an either pulmonary fibrosis or steady state based on the expression values of deregulated genes. This process resulted in a full and a laconic set of features capable of separating phenotypes to an at least equal degree as previously published marker sets. Indicatively, a maximum increase of 6% in specificity and 5% in Mathew's correlation coefficient was achieved. Evaluation with an additional independent dataset showed our feature set having a greater generalization potential than the rest. Ultimately, the proposed gene lists are expected not only to serve as new sets of diagnostic marker elements, but also as a target pool for future research initiatives.

Region-adaptive magnetic resonance image enhancement for improving CNN-based segmentation of the prostate and prostatic zones.

Automatic segmentation of the prostate of and the prostatic zones on MRI remains one of the most compelling research areas. While different image enhancement techniques are emerging as powerful tools for improving the performance of segmentation algorithms, their application still lacks consensus due to contrasting evidence regarding performance improvement and cross-model stability, further hampered by the inability to explain models' predictions. Particularly, for prostate segmentation, the effectiveness of image enhancement on different Convolutional Neural Networks (CNN) remains largely unexplored. The present work introduces a novel image enhancement method, named RACLAHE, to enhance the performance of CNN models for segmenting the prostate's gland and the prostatic zones. The improvement in performance and consistency across five CNN models (U-Net, U-Net++, U-Net3+, ResU-net and USE-NET) is compared against four popular image enhancement methods. Additionally, a methodology is proposed to explain, both quantitatively and qualitatively, the relation between saliency maps and ground truth probability maps. Overall, RACLAHE was the most consistent image enhancement algorithm in terms of performance improvement across CNN models with the mean increase in Dice Score ranging from 3 to 9% for the different prostatic regions, while achieving minimal inter-model variability. The integration of a feature driven methodology to explain the predictions after applying image enhancement methods, enables the development of a concrete, trustworthy automated pipeline for prostate segmentation on MR images.

AI-Enhanced Predictive Modeling for Identifying Depression and Delirium in Cardiovascular Patients Scheduled for Cardiac Surgery.

Several studies have demonstrated a critical association between cardiovascular disease (CVD) and mental health, revealing that approximately one-third of individuals with CVD also experience depression. This comorbidity significantly increases the risk of cardiac complications and mortality, a risk that persists regardless of traditional factors. Addressing this issue, our study pioneers a straightforward, explainable, and data-driven pipeline for predicting depression in CVD patients. METHODS: Our study was conducted at a cardiac surgical intensive care unit. A total of 224 participants who were scheduled for elective coronary artery bypass graft surgery (CABG) were enrolled in the study. Prior to surgery, each patient underwent psychiatric evaluation to identify major depressive disorder (MDD) based on the DSM-5 criteria. An advanced data curation workflow was applied to eliminate outliers and inconsistencies and improve data quality. An explainable AI-empowered pipeline was developed, where sophisticated machine learning techniques, including the AdaBoost, random forest, and XGBoost algorithms, were trained and tested on the curated data based on a stratified cross-validation approach. RESULTS: Our findings identified a significant correlation between the biomarker "sRAGE" and depression (r = 0.32, p = 0.038). Among the applied models, the random forest classifier demonstrated superior accuracy in predicting depression, with notable scores in accuracy (0.62), sensitivity (0.71), specificity (0.53), and area under the curve (0.67). CONCLUSIONS: This study provides compelling evidence that depression in CVD patients, particularly those with elevated "sRAGE" levels, can be predicted with a 62% accuracy rate. Our AI-driven approach offers a promising way for early identification and intervention, potentially revolutionizing care strategies in this vulnerable population.

Small airways dysfunction in patients with systemic sclerosis and interstitial lung disease.

Background: A number of studies report small airways involvement in patients with systemic sclerosis (SSc). Furthermore, small airways dysfunction is increasingly recognized in patients with interstitial lung disease (ILD) of idiopathic or autoimmune etiology. The objectives of this study were to evaluate small airways function in SSc patients with ILD and explore the effect of treatment on small airways function by using conventional and contemporary pulmonary function tests (PFTs). Methods: This single-center, prospective, observational study included a total of 35 SSc patients, with and without ILD based on HRCT scan, evaluated by a special radiologist blindly. Clinical data were collected from all patients who were also assessed for HRCT findings of small airways disease. Small airways function was assessed by classic spirometry, measurement of diffusing capacity for carbon monoxide, body plethysmography, single breath nitrogen washout (N2SBW) and impulse oscillometry (IOS). The prevalence of small airways dysfunction according to R5-R20, phase III slopeN2SBW and CV/VC methodologies was calculated in the total SSc population. Pulmonary function tests were compared between: (a) SSc-ILD and non-ILD patients and (b) two time points (baseline and follow up visit) in a subset of SSc-ILD patients who received treatment for ILD and were re-evaluated at a follow up visit after 12 months. Results: Phase III slopeN2SBW and R5-R20 showed the highest diagnostic performance for detecting small airways dysfunction among SSc patients (61 and 37.5%, respectively). Twenty three SSc patients were found with ILD and 14 of them had a 12-month follow up visit. SSc-ILD patients compared to those without ILD exhibited increased phase III slopeN2SBW ≥120% (p = 0.04), R5-R20 ≥0.07 kPa/L/s (p = 0.025), airway resistance (Raw) (p = 0.011), and special airway resistance (sRaw) (p = 0.02), and decreased specific airway conductance (sGaw) (p = 0.022), suggesting impaired small airways function in the SSc-ILD group. Radiographic features of SAD on HRCT were observed in 22% of SSc-ILD patients and in none of SSc-non-ILD patients. Comparison of PFTs between baseline and follow-up visit after 12 months in the 14 SSc-ILD treated patients, showed improvement of phase III slopeN2SBW (p = 0.034), R5-R20 (p = 0.035) and Raw (p = 0.044) but not sRaw and sGaw parameters. Conclusion: Phase III slopeN2SBW and R5-R20 may reveal small airways dysfunction in SSc associated ILD before structural damage and may be partially improved in a subset of patients receiving treatment for ILD.

Machine Learning Models for Cardiovascular Disease Events Prediction.

Cardiovascular diseases (CVDs) are among the most serious disorders leading to high mortality rates worldwide. CVDs can be diagnosed and prevented early by identifying risk biomarkers using statistical and machine learning (ML) models, In this work, we utilize clinical CVD risk factors and biochemical data using machine learning models such as Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), Extreme Grading Boosting (XGB) and Adaptive Boosting (AdaBoost) to predict death caused by CVD within ten years of follow-up. We used the cohort of the Ludwigshafen Risk and Cardiovascular Health (LURIC) study and 2943 patients were included in the analysis (484 annotated as dead due to CVD). We calculated the Accuracy (ACC), Precision, Recall, F1-Score, Specificity (SPE) and area under the receiver operating characteristic curve (AUC) of each model. The findings of the comparative analysis show that Logistic Regression has been proven to be the most reliable algorithm having accuracy 72.20 %. These results will be used in the TIMELY study to estimate the risk score and mortality of CVD in patients with 10-year risk.

Predicting the need for mechanical ventilation and mortality in hospitalized COVID-19 patients who received heparin.

Although several studies have utilized AI (artificial intelligence)-based solutions to enhance the decision making for mechanical ventilation, as well as, for mortality in COVID-19, the extraction of explainable predictors regarding heparin's effect in intensive care and mortality has been left unresolved. In the present study, we developed an explainable AI (XAI) workflow to shed light into predictors for admission in the intensive care unit (ICU), as well as, for mortality across those hospitalized COVID-19 patients who received heparin. AI empowered classifiers, such as, the hybrid Extreme gradient boosting (HXGBoost) with customized loss functions were trained on time-series curated clinical data to develop robust AI models. Shapley additive explanation analysis (SHAP) was conducted to determine the positive or negative impact of the predictors in the model's output. The HXGBoost predicted the risk for intensive care and mortality with 0.84 and 0.85 accuracy, respectively. SHAP analysis indicated that the low percentage of lymphocytes at day 7 along with increased FiO2 at days 1 and 5, low SatO2 at days 3 and 7 increase the probability for mortality and highlight the positive effect of heparin administration at the early days of hospitalization for reducing mortality.

A "smart" Imputation Approach for Effective Quality Control Across Complex Clinical Data Structures.

The overwhelming need to improve the quality of complex data structures in healthcare is more important than ever. Although data quality has been the point of interest in many studies, none of them has focused on the development of quantitative and explainable methods for data imputation. In this work, we propose a "smart" imputation workflow to address missing data across complex data structures in the context of in silico clinical trials. AI algorithms were utilized to produce high-quality virtual patient profiles. A search algorithm was then developed to extract the best virtual patient profiles through the definition of a profile matching score (PMS). A case study was conducted, where the real dataset was randomly contaminated with multiple missing values (e.g., 10 to 50%). In total, 10000 virtual patient profiles with less than 0.02 Kullback-Leibler (KL) divergence were produced to estimate the PMS distribution. The best generator achieved the lowest average squared absolute difference (0.4) and average correlation difference (0.02) with the real dataset highlighting its increased effectiveness for data imputation across complex clinical data structures.

Clinical Symptoms in Hospitalized and Self-Quarantined Patients with SARS-CoV-2 Infection in Northwestern Greece - Association with Olfactory and Gustatory Dysfunction.

Backround: We aimed to assess the relation of chemosensory dysfunction with the reported symptoms in two subgroups of patients in Northwestern Greece: the first one included patients with moderate to severe symptomatology who needed hospitalization and the second one, patients with mild symptoms who recovered at home. Methods:We used a questionnaire to select information about patient demographics, medical history and reported symptoms during infection. Three hundred COVID-19 positive patients who were identified via RT-PCR test in the University Hospital of Ioannina, Greece, were included in the present study, of which 150 recovered at home and the remaining 150 needed hospitalization. Statistical analysis was based on IBM-SPSS Statistics 26.0. Results:The majority of patients had fever during infection, while o minor percentage of those who needed hospitalization (12.67%) suffered from sore throat. There was a statistically significant difference between the loss of smell and clinical symptoms including fatigue, nose congestion, body aches and headache, and loss of taste and reported symptoms including fatigue, body aches, runny nose, headache and sore throat. Conclusion: Fever was the symptom with the highest percentage rate, while sore throat was the symptom with the lowest percentage rate. There are reported clinical symptoms related with olfactory and gustatory dysfunction during COVID-19 infection.

Dizziness in the Emergency Department: Insights and Epidemiological Data - a Population Based Study.

Background:Dizziness is a commonly referred symptom in emergency departments (EDs). The aim of this study is to describe the epidemiology of dizziness included acute vestibular syndrome (AVS) in the ED of the University Hospital of Ioannina, Grecce, during a six-month period. Methods:A total of 60 patients presenting with dizziness to the ED of our hospital during a short period of six months in 2021 were identified. Data were obtained through retrospective and prospective review of medical records. Statistical analysis was based on ÉBM-SPSS Statistics 26.0. Results:Among the 60 patients, 16.67% received the diagnosis of cerebellar stroke, 3.33% Meniere disease, 16.67% vestibular neuritis, 20% benign paroxysmal positional vertigo, 3.33% cardiovascular disease, and 1.67% had a neurological disease. Finally, 35% of patients left the ED undiagnosed. Conclusion:Benign paroxysmal positional vertigo was found to be the most common cause of dizziness in the ED, followed by cerebellar stroke and vestibular neuritis. A detailed neurological examination is recommended for the diagnosis of dizziness in the ED. Our data confirm the findings of previous studies in the GreeK population of patients presenting with dizziness to the ED of our hospital.

The Impact of Age and Gender and Their Association with Chemosensory Dysfunction, in Hospitalized and Self-Quarantine Patients with Covid-19 Infection, in Epirus, Greece.

Objective:Olfactory and gustatory dysfunction that relates with the infection from severe acute respiratory syndrome-related coronavirus (SARS-CoV-2) has already improved. The relation between chemosensory dysfunction and age and gender in covid-19 positive patients is the main objective of the present study. Methods:We used a questionnaire to select information about medical history, patient demographics and reported symptoms during infection. Three hundred covid-19 positive patients, who underwent a RT-PCR test in the University Hospital of Ioannina, Grecce, were included in this study; 150 of them recovered at home and the remaining 150 were admitted to hospital. Statistical analysis based on ÉBM-SPSS Statistics 26.0 was done. Results:The total sample included 300 patients, of which 106 females and 194 males. There was a statistically significant difference between the subgroup of patients aged 21-25, 61-65 and 71-75 with loss of smell, that of hospitalized patients aged 41-45 with loss of smell and the subgroup of those aged 31-35 and 71-75 with loss of taste. Conclusion:There is a significant association between chemosensory dysfunction and younger age groups. Olfactory and gustatory dysfunction appears more frequently in women than men. Male gender relates with disease severity.

Disease Progression of Hypertrophic Cardiomyopathy: Modeling Using Machine Learning.

BACKGROUND: Cardiovascular disorders in general are responsible for 30% of deaths worldwide. Among them, hypertrophic cardiomyopathy (HCM) is a genetic cardiac disease that is present in about 1 of 500 young adults and can cause sudden cardiac death (SCD). OBJECTIVE: Although the current state-of-the-art methods model the risk of SCD for patients, to the best of our knowledge, no methods are available for modeling the patient's clinical status up to 10 years ahead. In this paper, we propose a novel machine learning (ML)-based tool for predicting disease progression for patients diagnosed with HCM in terms of adverse remodeling of the heart during a 10-year period. METHODS: The method consisted of 6 predictive regression models that independently predict future values of 6 clinical characteristics: left atrial size, left atrial volume, left ventricular ejection fraction, New York Heart Association functional classification, left ventricular internal diastolic diameter, and left ventricular internal systolic diameter. We supplemented each prediction with the explanation that is generated using the Shapley additive explanation method. RESULTS: The final experiments showed that predictive error is lower on 5 of the 6 constructed models in comparison to experts (on average, by 0.34) or a consortium of experts (on average, by 0.22). The experiments revealed that semisupervised learning and the artificial data from virtual patients help improve predictive accuracies. The best-performing random forest model improved R2 from 0.3 to 0.6. CONCLUSIONS: By engaging medical experts to provide interpretation and validation of the results, we determined the models' favorable performance compared to the performance of experts for 5 of 6 targets.

ICU admission and mortality classifiers for COVID-19 patients based on subgroups of dynamically associated profiles across multiple timepoints.

The coronavirus disease 2019 (COVID-19) which is caused by severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) is consistently causing profound wounds in the global healthcare system due to its increased transmissibility. Currently, there is an urgent unmet need to identify the underlying dynamic associations among COVID-19 patients and distinguish patient subgroups with common clinical profiles towards the development of robust classifiers for ICU admission and mortality. To address this need, we propose a four step pipeline which: (i) enhances the quality of multiple timeseries clinical data through an automated data curation workflow, (ii) deploys Dynamic Bayesian Networks (DBNs) for the detection of features with increased connectivity based on dynamic association analysis across multiple points, (iii) utilizes Self Organizing Maps (SOMs) and trajectory analysis for the early identification of COVID-19 patients with common clinical profiles, and (iv) trains robust multiple additive regression trees (MART) for ICU admission and mortality classification based on the extracted homogeneous clusters, to identify risk factors and biomarkers for disease progression. The contribution of the extracted clusters and the dynamically associated clinical data improved the classification performance for ICU admission to sensitivity 0.83 and specificity 0.83, and for mortality to sensitivity 0.74 and specificity 0.76. Additional information was included to enhance the performance of the classifiers yielding an increase by 4% in sensitivity and specificity for mortality. According to the risk factor analysis, the number of lymphocytes, SatO2, PO2/FiO2, and O2 supply type were highlighted as risk factors for ICU admission and the percentage of neutrophils and lymphocytes, PO2/FiO2, LDH, and ALP for mortality, among others. To our knowledge, this is the first study that combines dynamic modeling with clustering analysis to identify homogeneous groups of COVID-19 patients towards the development of robust classifiers for ICU admission and mortality.

Addressing the clinical unmet needs in primary Sjögren's Syndrome through the sharing, harmonization and federated analysis of 21 European cohorts.

For many decades, the clinical unmet needs of primary Sjögren's Syndrome (pSS) have been left unresolved due to the rareness of the disease and the complexity of the underlying pathogenic mechanisms, including the pSS-associated lymphomagenesis process. Here, we present the HarmonicSS cloud-computing exemplar which offers beyond the state-of-the-art data analytics services to address the pSS clinical unmet needs, including the development of lymphoma classification models and the identification of biomarkers for lymphomagenesis. The users of the platform have been able to successfully interlink, curate, and harmonize 21 regional, national, and international European cohorts of 7,551 pSS patients with respect to the ethical and legal issues for data sharing. Federated AI algorithms were trained across the harmonized databases, with reduced execution time complexity, yielding robust lymphoma classification models with 85% accuracy, 81.25% sensitivity, 85.4% specificity along with 5 biomarkers for lymphoma development. To our knowledge, this is the first GDPR compliant platform that provides federated AI services to address the pSS clinical unmet needs.

Bayesian Inference-Based Gaussian Mixture Models With Optimal Components Estimation Towards Large-Scale Synthetic Data Generation for In Silico Clinical Trials.

Goal: To develop a computationally efficient and unbiased synthetic data generator for large-scale in silico clinical trials (CTs). Methods: We propose the BGMM-OCE, an extension of the conventional BGMM (Bayesian Gaussian Mixture Models) algorithm to provide unbiased estimations regarding the optimal number of Gaussian components and yield high-quality, large-scale synthetic data at reduced computational complexity. Spectral clustering with efficient eigenvalue decomposition is applied to estimate the hyperparameters of the generator. A case study is conducted to compare the performance of BGMM-OCE against four straightforward synthetic data generators for in silico CTs in hypertrophic cardiomyopathy (HCM). Results: The BGMM-OCE generated 30000 virtual patient profiles having the lowest coefficient-of-variation (0.046), inter- and intra-correlation differences (0.017, and 0.016, respectively) with the real ones in reduced execution time. Conclusions: BGMM-OCE overcomes the lack of population size in HCM which obscures the development of targeted therapies and robust risk stratification models.

A federated AI strategy for the classification of patients with Mucosa Associated Lymphoma Tissue (MALT) lymphoma across multiple harmonized cohorts.

Mucosa Associated Lymphoma Tissue (MALT) type is an extremely rare type of lymphoma which occurs in less than 3% of patients with primary Sjögren's Syndrome (pSS). No reported studies so far have been able to investigate risk factors for MALT development across multiple cohort databases with sufficient statistical power. Here, we present a generalized, federated AI (artificial intelligence) strategy which enables the training of AI algorithms across multiple harmonized databases. A case study is conducted towards the development of MALT classification models across 17 databases on pSS. Advanced AI algorithms were developed, including federated Multinomial Naïve Bayes (FMNB), federated gradient boosting trees (FGBT), FGBT with dropouts (FDART), and the federated Multilayer Perceptron (FMLP). The FDART with dropout rate 0.3 achieved the best performance with sensitivity 0.812, and specificity 0.829, yielding 8 biomarkers as prominent for MALT development.