The coronavirus disease 2019 (COVID-19) pandemic is a scientific, medical, and social challenge. The complexity of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is centered on the ...unpredictable clinical course of the disease that can rapidly develop, causing severe and deadly complications. The identification of effective laboratory biomarkers able to classify patients based on their risk is imperative in being able to guarantee prompt treatment. The analysis of recently published studies highlights the role of systemic vasculitis and cytokine mediated coagulation disorders as the principal actors of multi organ failure in patients with severe COVID-19 complications. The following biomarkers have been identified: hematological (lymphocyte count, neutrophil count, neutrophil-lymphocyte ratio (NLR)), inflammatory (C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), procalcitonin (PCT)), immunological (interleukin (IL)-6 and biochemical (D-dimer, troponin, creatine kinase (CK), aspartate aminotransferase (AST)), especially those related to coagulation cascades in disseminated intravascular coagulation (DIC) and acute respiratory distress syndrome (ARDS). New laboratory biomarkers could be identified through the accurate analysis of multicentric case series; in particular, homocysteine and angiotensin II could play a significant role.
Celotno besedilo
Dostopno za:
DOBA, IJS, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK, VSZLJ
The US Food and Drug Administration (FDA) and European Medicines Agency (EMA) are currently willing to consider a 30% to 40% glomerular filtration rate (GFR) decline as a surrogate end point for ...kidney failure for clinical trials of kidney disease progression under appropriate conditions. However, these end points may not be practical for early stages of kidney disease. In March 2018, the National Kidney Foundation sponsored a scientific workshop in collaboration with the FDA and EMA to evaluate changes in albuminuria or GFR as candidate surrogate end points. Three parallel efforts were presented: meta-analyses of observational studies (cohorts), meta-analyses of clinical trials, and simulations of trial design. In cohorts, after accounting for measurement error, relationships between change in urinary albumin-creatinine ratio (UACR) or estimated GFR (eGFR) slope and the clinical outcome of kidney disease progression were strong and consistent. In trials, the posterior median R2 of treatment effects on the candidate surrogates with the clinical outcome was 0.47 (95% Bayesian credible interval BCI, 0.02-0.96) for early change in UACR and 0.72 (95% BCI, 0.05-0.99) when restricted to baseline UACR>30mg/g, and 0.97 (95% BCI, 0.78-1.00) for total eGFR slope at 3 years and 0.96 (95% BCI, 0.63-1.00) for chronic eGFR slope (ie, the slope excluding the first 3 months from baseline, when there might be acute changes in eGFR). The magnitude of the relationships of changes in the candidate surrogates with risk for clinical outcome was consistent across cohorts and trials: a UACR reduction of 30% or eGFR slope reduction by 0.5 to 1.0mL/min/1.73m2 per year were associated with an HR of ∼0.7 for the clinical outcome in cohorts and trials. In simulations, using GFR slope as an end point substantially reduced the required sample size and duration of follow-up compared with the clinical end point when baseline eGFR was high, treatment effects were uniform, and there was no acute effect of the treatment. We conclude that both early change in albuminuria and GFR slope fulfill criteria for surrogacy for use as end points in clinical trials for chronic kidney disease progression under certain conditions, with stronger support for change in GFR than albuminuria. Implementation requires understanding conditions under which each surrogate is likely to perform well and restricting its use to those settings.
The National Kidney Foundation–Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) guideline for evaluation, classification, and stratification of chronic kidney disease (CKD) was published in ...2002. The KDOQI guideline was well accepted by the medical and public health communities, but concerns and criticisms arose as new evidence became available since the publication of the original guidelines. KDIGO (Kidney Disease: Improving Global Outcomes) recently published an updated guideline to clarify the definition and classification of CKD and to update recommendations for the evaluation and management of individuals with CKD based on new evidence published since 2002. The primary recommendations were to retain the current definition of CKD based on decreased glomerular filtration rate or markers of kidney damage for 3 months or more and to include the cause of kidney disease and level of albuminuria, as well as level of glomerular filtration rate, for CKD classification. NKF-KDOQI convened a work group to write a commentary on the KDIGO guideline in order to assist US practitioners in interpreting the KDIGO guideline and determining its applicability within their own practices. Overall, the commentary work group agreed with most of the recommendations contained in the KDIGO guidelines, particularly the recommendations regarding the definition and classification of CKD. However, there were some concerns about incorporating the cause of disease into CKD classification, in addition to certain recommendations for evaluation and management.
Biopsy-confirmed liver fibrosis is a prognostic factor for patients with nonalcoholic fatty liver disease (NAFLD). We performed a systematic review to quantify the prognostic value of fibrosis stage ...in patients with NAFLD and the subgroup of patients with nonalcoholic steatohepatitis (NASH) and to assess the evidence that change in fibrosis stage is a surrogate endpoint.
We searched the MEDLINE, Embase, Cochrane Library, and trial registry databases through August 2018 for prospective or retrospective cohort studies of liver-related clinical events and outcomes in adults with NAFLD or NASH. We collected data on mortality (all cause and liver related) and morbidity (cirrhosis, liver cancer, and all liver-related events) by stage of fibrosis, determined by biopsy, for patients with NAFLD or NASH. Using fibrosis stage 0 as a reference population, we calculated fibrosis stage-specific relative risk (RR) and 95% confidence interval (CI) values for mortality and morbidities. We performed fixed-effect and random-effect model meta-analyses. Metaregression was used to examine associations among study design (prospective vs retrospective cohort), overall risk of bias (medium or high), and mean duration of follow-up (in years).
Our meta-analysis included 13 studies, comprising 4428 patients with NAFLD; 2875 of these were reported to have NASH. Compared with no fibrosis (stage 0), unadjusted risk increased with increasing stage of fibrosis (stage 0 vs 4): all-cause mortality RR, 3.42 (95% CI, 2.63–4.46); liver-related mortality RR, 11.13 (95% CI, 4.15–29.84); liver transplant RR, 5.42 (95% CI, 1.05–27.89); and liver-related events RR, 12.78 (95% CI, 6.85–23.85). The magnitude of RR did not differ significantly after adjustment for confounders, including age or sex in the subgroup of NAFLD patients with NASH. Three studies examined the effects of increasing fibrosis on quality of life had inconsistent findings.
In a systematic review and meta-analysis, we found biopsy-confirmed fibrosis to be associated with risk of mortality and liver-related morbidity in patients with NAFLD, with and without adjustment for confounding factors and in patients with reported NASH. Further studies are needed to assess the association between fibrosis stage and patient quality of life and establish that change in liver fibrosis stage is a valid endpoint for use in clinical trials.
•We explored the fusion of heterogeneous multimodal time series data for Parkinson’s prediction with different problem formulation strategies.•Features selection optimization plays an essential role ...in classification and progression detection.•Different data balancing strategies have been explored to prepare the data set accurately.•The Bayesian optimization for model hyper-parameters shows a remarkable performance.•Extensive explainability techniques have been investigated to interpret the model decisions.
Parkinson’s Disease (PD) is a devastating chronic neurological condition. Machine learning (ML) techniques have been used in the early prediction of PD progression. Fusion of heterogeneous data modalities proved its capability to improve the performance of ML models. Time series data fusion supports the tracking of the disease over time. In addition, the trustworthiness of the resulting models is improved by adding model explainability features. The literature on PD has not sufficiently explored these three points.
In this work, we proposed an ML pipeline for predicting the progression of PD that is both accurate and explainable. We explore the fusion of different combinations of five time series modalities from the Parkinson’s Progression Markers Initiative (PPMI) real-world dataset, including patient characteristics, biosamples, medication history, motor, and non-motor function data. Each patient has six visits. The problem has been formulated in two ways: ❶ a three-class based progression prediction with 953 patients in each time series modality, and ❷ a four-class based progression prediction with 1,060 patients in each time series modality. The statistical features of these six visits were calculated from each modality and diverse feature selection methods were applied to select the most informative feature sets. The extracted features were used to train a set of well-known ML models including Support vector machines (SVM), random forests (RF), extra tree classifier (ETC), light gradient boosting machines (LGBM), and stochastic gradient descent (SGD). We examined a number of data-balancing strategies in the pipeline with different combinations of modalities. ML models have been optimized using the Bayesian optimizer. A comprehensive evaluation of various ML methods has been conducted, and the best models have been extended to provide different explainability features.
We compare the performance of ML models before and after optimization and using and without using feature selection. In the three-class experiment and with various modality fusions, the LGBM model produced the most accurate results with a 10-fold cross-validation (10-CV) accuracy of 90.73% using non-motor function modality. RF produced the best results in the four-class experiment with various modality fusions with a 10-CV accuracy of 94.57% using non-motor modality. With the fused dataset of non-motor and motor function modalities, the LGBM model outperformed the other ML models in both the 3-class and 4-class experiments (i.e., 10-CV accuracy of 94.89% and 93.73%, respectively). Using the Shapely Additive Explanations (SHAP) framework, we employed global and instance-based explanations to explain the behavior of each ML classifier. Moreover, we extended the explainability by implementing the LIME and SHAPASH local explainers. The consistency of these explainers has been explored. The resultant classifiers were accurate, explainable, and thus medically more relevant and applicable.
The select modalities and feature sets were confirmed by the literature and medical experts. The various explainers suggest that the bradykinesia (NP3BRADY) feature was the most dominant and consistent. By providing thorough insights into the influence of multiple modalities on the disease risk, the suggested approach is expected to help improve the clinical knowledge of PD progression processes.
Long recognized as an evolutionarily ancient cell type involved in tissue homeostasis and immune defense against pathogens, macrophages are being re-discovered as regulators of several diseases, ...including cancer. Tumor-associated macrophages (TAMs) represent the most abundant innate immune population in the tumor microenvironment (TME). Macrophages are professional phagocytic cells of the hematopoietic system specializing in the detection, phagocytosis and destruction of bacteria and other harmful micro-organisms, apoptotic cells and metabolic byproducts. In contrast to these healthy macrophage functions, TAMs support cancer cell growth and metastasis and mediate immunosuppressive effects on the adaptive immune cells of the TME. Cancer is one of the most potent insults on macrophage physiology, inducing changes that are intimately linked with disease progression. In this Review, we outline hallmarks of TAMs and discuss the emerging mechanisms that contribute to their pathophysiological adaptations and the vulnerabilities that provide attractive targets for therapeutic exploitation in cancer.
Display omitted
•An end-to-end integrated framework for modeling AD progression via predicting the future clinical scores and 3D MRI concurrently.•The integrated framework unifies regression and GAN ...models with three strategies.•A new ROI loss for guiding the framework to pay attention on the brain regions that reflect disease progression significantly.
Alzheimer’s disease is a common neurodegenerative brain disease that affects the elderly population worldwide. Its early automatic detection is vital for early intervention and treatment. A common solution is to perform future cognitive score prediction based on the baseline brain structural magnetic resonance image (MRI), which can directly infer the potential severity of disease. Recently, several studies have modelled disease progression by predicting the future brain MRI that can provide visual information of brain changes over time. Nevertheless, no studies explore the intra correlation of these two solutions, and it is unknown whether the predicted MRI can assist the prediction of cognitive score. Here, instead of independent prediction, we aim to predict disease progression in multi-view, i.e., predicting subject-specific changes of cognitive score and MRI volume concurrently. To achieve this, we propose an end-to-end integrated framework, where a regression model and a generative adversarial network are integrated together and then jointly optimized. Three integration strategies are exploited to unify these two models. Moreover, considering that some brain regions, such as hippocampus and middle temporal gyrus, could change significantly during the disease progression, a region-of-interest (ROI) mask and a ROI loss are introduced into the integrated framework to leverage this anatomical prior knowledge. Experimental results on the longitudinal Alzheimer’s Disease Neuroimaging Initiative dataset demonstrated that the integrated framework outperformed the independent regression model for cognitive score prediction. And its performance can be further improved with the ROI loss for both cognitive score and MRI prediction.