Inflammation and weight change related to neurocognitive and functional impairment in diabetes and psychiatric disorders.

INTRODUCTION: Obesity is a global pandemic associated with various cardio-metabolic and psychiatric disorders. Neurocognitive and functional deficits have been associated with several somatic and psychiatric disorders. Adiposity-related inflammation has recently emerged as a key risk factor for neurocognitive and functional impairments. This prospective transdiagnostic study aimed to investigate the role of adiposity-related inflammatory markers in neurocognitive and functional outcomes associated with weight changes. METHODS: Peripheral blood inflammatory and oxidative stress biomarkers and neurocognitive and functional performance were assessed twice over 1 year in 165 individuals, including 30 with schizophrenia, 42 with bipolar disorder, 35 with major depressive disorder, 30 with type 2 diabetes mellitus (T2DM), and 28 healthy controls. Participants were stratified by body mass index into categories of type 2 obesity (T2OB; n=30), type 1 obesity (T1OB; n=42), overweight (OW; n=53), and average weight (NW; n=40). Mixed one-way analysis of covariance and linear and binary logistic regression analyses were performed. RESULTS: Compared with NW, T2OB and T1OB were significantly associated with impaired neurocognitive and functional performance (p<0.01; η2p=0.06-0.12) and higher levels of C-reactive protein and platelets (PLT) (p<0.01; η2p=0.08-0.16), with small-to-moderate effect sizes. IL-6, IL-10, and PLT were key factors for detecting significant weight changes in T1OB and T2OB over time. Regression models revealed that inflammatory and oxidative stress biomarkers and cellular adhesion molecules were significantly associated with neurocognitive and functional performance (p<0.05). DISCUSSION: Obesity is characterized by neurocognitive and functional impairments alongside low-grade systemic inflammation. Adiposity-related inflammatory biomarkers may contribute to neurocognitive and functional decline in individuals with T2DM and psychiatric disorders. Our data suggest that these biomarkers facilitate the identification of specific subgroups of individuals at higher risk of developing obesity.

Inflammation and lipid metabolism as potential biomarkers of memory impairment across type 2 diabetes mellitus and severe mental disorders.

INTRODUCTION: Neurocognitive impairment is a transdiagnostic feature across several psychiatric and cardiometabolic conditions. The relationship between inflammatory and lipid metabolism biomarkers and memory performance is not fully understood. This study aimed to identify peripheral biomarkers suitable to signal memory decline from a transdiagnostic and longitudinal perspective. METHODS: Peripheral blood biomarkers of inflammation, oxidative stress and lipid metabolism were assessed twice over a 1-year period in 165 individuals, including 30 with schizophrenia (SZ), 42 with bipolar disorder (BD), 35 with major depressive disorder (MDD), 30 with type 2 diabetes mellitus (T2DM), and 28 healthy controls (HCs). Participants were stratified by memory performance quartiles, taking as a reference their global memory score (GMS) at baseline, into categories of high memory (H; n = 40), medium to high memory (MH; n = 43), medium to low memory (ML; n = 38) and low memory (L; n = 44). Exploratory and confirmatory factorial analysis, mixed one-way analysis of covariance and discriminatory analyses were performed. RESULTS: L group was significantly associated with higher levels of tumor necrosis factor-alpha (TNF-α) and lower levels of apolipoprotein A1 (Apo-A1) compared to those from the MH and H groups (p < 0.05; η2p = 0.06-0.09), with small to moderate effect sizes. Moreover, the combination of interleukin-6 (IL-6), TNF-α, c-reactive protein (CRP), Apo-A1 and Apo-B compounded the transdiagnostic model that best discriminated between groups with different degrees of memory impairment (χ2 = 11.9-49.3, p < 0.05-0.0001). CONCLUSIONS: Inflammation and lipid metabolism seem to be associated with memory across T2DM and severe mental illnesses (SMI). A panel of biomarkers may be a useful approach to identify individuals at greater risk of neurocognitive impairment. These findings may have a potential translational utility for early intervention and advance precision medicine in these disorders.

Immune-Inflammatory Biomarkers Predict Cognition and Social Functioning in Patients With Type 2 Diabetes Mellitus, Major Depressive Disorder, Bipolar Disorder, and Schizophrenia: A 1-Year Follow-Up Study.

Background: Systemic, low-grade immune-inflammatory activity, together with social and neurocognitive performance deficits are a transdiagnostic trait of people suffering from type 2 diabetes mellitus (T2DM) and severe mental illnesses (SMIs), such as schizophrenia (SZ), major depressive disorder (MDD), and bipolar disorder (BD). We aimed to determine if immune-inflammatory mediators were significantly altered in people with SMIs or T2DM compared with healthy controls (HC) and whether these biomarkers could help predict their cognition and social functioning 1 year after assessment. Methods: We performed a prospective, 1-year follow-up cohort study with 165 participants at baseline (TB), including 30 with SZ, 42 with BD, 35 with MDD, 30 with T2DM, and 28 HC; and 125 at 1-year follow-up (TY), and determined executive domain (ED), global social functioning score (GSFS), and peripheral blood immune-inflammatory and oxidative stress biomarkers. Results: Participants with SMIs and T2DM showed increased peripheral levels of inflammatory markers, such as interleukin-10 (p < 0.01; η2 p = 0.07) and tumor necrosis factor-α (p < 0.05; η2 p = 0.08); and oxidative stress biomarkers, such as reactive oxygen species (ROS) (p < 0.05; η2 p = 0.07) and mitochondrial ROS (p < 0.01; η2 p = 0.08). The different combinations of the exposed biomarkers anticipated 46-57.3% of the total ED and 23.8-35.7% of GSFS for the participants with SMIs. Limitations: Participants' treatment, as usual, was continued without no specific interventions; thus, it was difficult to anticipate substantial changes related to the psychopharmacological pattern. Conclusion: People with SMIs show significantly increased levels of peripheral immune-inflammatory biomarkers, which may contribute to the neurocognitive and social deficits observed in SMIs, T2DM, and other diseases with systemic immune-inflammatory activation of chronic development. These parameters could help identify the subset of patients who could benefit from immune-inflammatory modulator strategies to ameliorate their functional outcomes.

Prediction of the risk of cancer and the grade of dysplasia in leukoplakia lesions using deep learning.

OBJECTIVES: To estimate the probability of malignancy of an oral leukoplakia lesion using Deep Learning, in terms of evolution to cancer and high-risk dysplasia. MATERIALS AND METHODS: A total of 261 oral leukoplakia lesions with a mean of 5.5 years follow-up were analysed from standard digital photographs. A deep learning pipeline composed by a U-Net based segmentation of the lesion followed by a multi-task CNN classifier was used to predict the malignant transformation and the risk of dysplasia of the lesion. An explainability heatmap is constructed using LIME in order to interpret the decision of the model for each output. RESULTS: A Dice coefficient of 0.561 was achieved on the segmentation task. For the prediction of a malignant transformation, the model provided a sensitivity of 1 with a specificity of 0.692. For the prediction of high-risk dysplasia, the model achieved a specificity of 0.740 and a sensitivity of 0.928. CONCLUSION: The proposed model using deep learning can be a helpful tool for predicting the possible malignant evolution of oral leukoplakias. The generated heatmap provides a high confidence on the output of the model and enables its interpretability.

Specific metabolic syndrome components predict cognition and social functioning in people with type 2 diabetes mellitus and severe mental disorders.

OBJECTIVE: Obesity and metabolic diseases such as metabolic syndrome (MetS) are more prevalent in people with type 2 diabetes mellitus (T2DM), major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia (SZ). MetS components might be associated with neurocognitive and functional impairments in these individuals. The predictive and discriminatory validity of MetS and its components regarding those outcomes were assessed from prospective and transdiagnostic perspectives. METHODS: Metabolic syndrome components and neurocognitive and social functioning were assessed in 165 subjects, including 30 with SZ, 42 with BD, 35 with MDD, 30 with T2DM, and 28 healthy controls (HCs). A posteriori, individuals were classified into two groups. The MetS group consisted of those who met at least three of the following criteria: abdominal obesity (AO), elevated triglycerides (TG), reduced high-density lipoprotein cholesterol (HDL), elevated blood pressure (BP), and elevated fasting glucose (FPG); the remaining participants comprised the No-MetS group. Mixed one-way analysis of covariance and linear and binary logistic regression analyses were performed. RESULTS: Cognitive impairment was significantly greater in the MetS group (n = 82) than in the No-MetS group (n = 83), with small effect sizes (p < 0.05; η²p = 0.02 - 0.03). In both groups, the most robust associations between MetS components and neurocognitive and social functioning were observed with TG and FPG (p < 0.05). There was also evidence for a significant relationship between cognition and BP in the MetS group (p < 0.05). The combination of TG, FPG, elevated systolic BP and HDL best classified individuals with greater cognitive impairment (p < 0.001), and TG was the most accurate (p < 0.0001). CONCLUSIONS: Specific MetS components are significantly associated with cognitive impairment across somatic and psychiatric disorders. Our findings provide further evidence on the summative effect of MetS components to predict cognition and social functioning and allow the identification of individuals with worse outcomes. Transdiagnostic, lifestyle-based therapeutic interventions targeted at that group hold the potential to improve health outcomes.

Transdiagnostic neurocognitive deficits in patients with type 2 diabetes mellitus, major depressive disorder, bipolar disorder, and schizophrenia: A 1-year follow-up study.

BACKGROUND: Neurocognition impairments are critical factors in patients with major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia (SZ), and also in those with somatic diseases such as type 2 diabetes mellitus (T2DM). Intriguingly, these severe mental illnesses are associated with an increased co-occurrence of diabetes (direct comorbidity). This study sought to investigate the neurocognition and social functioning across T2DM, MDD, BD, and SZ using a transdiagnostic and longitudinal approach. METHODS: A total of 165 participants, including 30 with SZ, 42 with BD, 35 with MDD, 30 with T2DM, and 28 healthy controls (HC), were assessed twice at a 1-year interval using a comprehensive, integrated test battery on neuropsychological and social functioning. RESULTS: Common neurocognitive impairments in somatic and psychiatric disorders were identified, including deficits in short-term memory and cognitive reserve (p < 0.01, η²p=0.08-0.31). Social functioning impairments were observed in almost all the disorders (p < 0.0001; η²p=0.29-0.49). Transdiagnostic deficits remained stable across the 1-year follow-up (p < 0.001; η²p=0.13-0.43) and could accurately differentiate individuals with somatic and psychiatric disorders (χ² = 48.0, p < 0.0001). LIMITATIONS: The initial sample size was small, and high experimental mortality was observed after follow-up for one year. CONCLUSIONS: This longitudinal study provides evidence of some possible overlap in neurocognition deficits across somatic and psychiatric diagnostic categories, such as T2DM, MDD, BD, and SZ, which have high comorbidity. This overlap may be a result of shared genetic and environmental etiological factors. The findings open promising avenues for research on transdiagnostic phenotypes of neurocognition in these disorders, in addition to their biological bases.

Prediction of the hemoglobin level in hemodialysis patients using machine learning techniques.

Patients who suffer from chronic renal failure (CRF) tend to suffer from an associated anemia as well. Therefore, it is essential to know the hemoglobin (Hb) levels in these patients. The aim of this paper is to predict the hemoglobin (Hb) value using a database of European hemodialysis patients provided by Fresenius Medical Care (FMC) for improving the treatment of this kind of patients. For the prediction of Hb, both analytical measurements and medication dosage of patients suffering from chronic renal failure (CRF) are used. Two kinds of models were trained, global and local models. In the case of local models, clustering techniques based on hierarchical approaches and the adaptive resonance theory (ART) were used as a first step, and then, a different predictor was used for each obtained cluster. Different global models have been applied to the dataset such as Linear Models, Artificial Neural Networks (ANNs), Support Vector Machines (SVM) and Regression Trees among others. Also a relevance analysis has been carried out for each predictor model, thus finding those features that are most relevant for the given prediction.

Foetal ECG recovery using dynamic neural networks.

Non-invasive electrocardiography has proven to be a very interesting method for obtaining information about the foetus state and thus to assure its well-being during pregnancy. One of the main applications in this field is foetal electrocardiogram (ECG) recovery by means of automatic methods. Evident problems found in the literature are the limited number of available registers, the lack of performance indicators, and the limited use of non-linear adaptive methods. In order to circumvent these problems, we first introduce the generation of synthetic registers and discuss the influence of different kinds of noise to the modelling. Second, a method which is based on numerical (correlation coefficient) and statistical (analysis of variance, ANOVA) measures allows us to select the best recovery model. Finally, finite impulse response (FIR) and gamma neural networks are included in the adaptive noise cancellation (ANC) scheme in order to provide highly non-linear, dynamic capabilities to the recovery model. Neural networks are benchmarked with classical adaptive methods such as the least mean squares (LMS) and the normalized LMS (NLMS) algorithms in simulated and real registers and some conclusions are drawn. For synthetic registers, the most determinant factor in the identification of the models is the foetal-maternal signal-to-noise ratio (SNR). In addition, as the electromyogram contribution becomes more relevant, neural networks clearly outperform the LMS-based algorithm. From the ANOVA test, we found statistical differences between LMS-based models and neural models when complex situations (high foetal-maternal and foetal-noise SNRs) were present. These conclusions were confirmed after doing robustness tests on synthetic registers, visual inspection of the recovered signals and calculation of the recognition rates of foetal R-peaks for real situations. Finally, the best compromise between model complexity and outcomes was provided by the FIR neural network. Both the methodology for selecting a model and the introduction of advanced neural models are the main contributions of this paper.