Predicting Alzheimer's disease CSF core biomarkers: a multimodal Machine Learning approach.

Introduction: Alzheimer's disease (AD) is a progressive neurodegenerative disorder. Current core cerebrospinal fluid (CSF) AD biomarkers, widely employed for diagnosis, require a lumbar puncture to be performed, making them impractical as screening tools. Considering the role of sleep disturbances in AD, recent research suggests quantitative sleep electroencephalography features as potential non-invasive biomarkers of AD pathology. However, quantitative analysis of comprehensive polysomnography (PSG) signals remains relatively understudied. PSG is a non-invasive test enabling qualitative and quantitative analysis of a wide range of parameters, offering additional insights alongside other biomarkers. Machine Learning (ML) gained interest for its ability to discern intricate patterns within complex datasets, offering promise in AD neuropathology detection. Therefore, this study aims to evaluate the effectiveness of a multimodal ML approach in predicting core AD CSF biomarkers. Methods: Mild-moderate AD patients were prospectively recruited for PSG, followed by testing of CSF and blood samples for biomarkers. PSG signals underwent preprocessing to extract non-linear, time domain and frequency domain statistics quantitative features. Multiple ML algorithms were trained using four subsets of input features: clinical variables (CLINVAR), conventional PSG parameters (SLEEPVAR), quantitative PSG signal features (PSGVAR) and a combination of all subsets (ALL). Cross-validation techniques were employed to evaluate model performance and ensure generalizability. Regression models were developed to determine the most effective variable combinations for explaining variance in the biomarkers. Results: On 49 subjects, Gradient Boosting Regressors achieved the best results in estimating biomarkers levels, using different loss functions for each biomarker: least absolute deviation (LAD) for the Aß42, least squares (LS) for p-tau and Huber for t-tau. The ALL subset demonstrated the lowest training errors for all three biomarkers, albeit with varying test performance. Specifically, the SLEEPVAR subset yielded the best test performance in predicting Aß42, while the ALL subset most accurately predicted p-tau and t-tau due to the lowest test errors. Conclusions: Multimodal ML can help predict the outcome of CSF biomarkers in early AD by utilizing non-invasive and economically feasible variables. The integration of computational models into medical practice offers a promising tool for the screening of patients at risk of AD, potentially guiding clinical decisions.

Applications and advancements of polysaccharide-based nanostructures for enhanced drug delivery.

Growing demand for highly effective, site-specific delivery of pharmaceuticals and nutraceuticals using nano-sized carriers has prompted increased scrutiny of carrier biocompatibility and biodegradability. To address these concerns, biodegradable natural polymers have emerged as a transformative domain, offering non-toxic, precisely targetable carriers capable of finely modulating cargo pharmacokinetics while generating innocuous decomposition by-products. This comprehensive review illuminates the emergence of polysaccharide-based nanoparticulate drug delivery systems. These systems establish an interactive interface between drug and targeted organs, guided by strategic modifications to polysaccharide backbones, which facilitate the creation of morphologically, constitutionally, and characteristically vibrant nanostructures through various fabrication routes, underpinning their pivotal role in biomedical applications. Advancements crucial to enhancing polysaccharide-based drug delivery, such as surface modifications and bioinspired modifications for enhanced targeting, and stimuli-responsive release, strategies to overcome biological barriers, enhance tumor penetration, and optimize therapeutic outcomes are highlighted. This review also examines some potent challenges, and the contemporary way out of them, and discusses future perspectives in the field.

Parametric analysis of craniocerebral injury mechanism in pedestrian traffic accidents based on finite element methods.

PURPOSE: The toughest challenge in pedestrian traffic accident identification lies in ascertaining injury manners. This study aimed to systematically simulate and parameterize 3 types of craniocerebral injury including impact injury, fall injury, and run-over injury, to compare the injury response outcomes of different injury manners. METHODS: Based on the total human model for safety (THUMS) and its enhanced human model THUMS-hollow structures, a total of 84 simulations with 3 injury manners, different loading directions, and loading velocities were conducted. Von Mises stress, intracranial pressure, maximum principal strain, cumulative strain damage measure, shear stress, and cranial strain were employed to analyze the injury response of all areas of the brain. To examine the association between injury conditions and injury consequences, correlation analysis, principal component analysis, linear regression, and stepwise linear regression were utilized. RESULTS: There is a significant correlation observed between each criterion of skull and brain injury (p < 0.01 in all Pearson correlation analysis results). A 2-phase increase of cranio-cerebral stress and strain as impact speed increases. In high-speed impact (> 40 km/h), the Von Mises stress on the skull was with a high possibility exceed the threshold for skull fracture (100 MPa). When falling and making temporal and occipital contact with the ground, the opposite side of the impacted area experiences higher frequency stress concentration than contact at other conditions. Run-over injuries tend to have a more comprehensive craniocerebral injury, with greater overall deformation due to more adequate kinetic energy conduction. The mean value of maximum principal strain of brain and Von Mises stress of cranium at run-over condition are 1.39 and 403.8 MPa, while they were 1.31, 94.11 MPa and 0.64, 120.5 MPa for the impact and fall conditions, respectively. The impact velocity also plays a significant role in craniocerebral injury in impact and fall loading conditions (the p of all F-test < 0.05). A regression equation of the craniocerebral injury manners in pedestrian accidents was established. CONCLUSION: The study distinguished the craniocerebral injuries caused in different manners, elucidated the biomechanical mechanisms of craniocerebral injury, and provided a biomechanical foundation for the identification of craniocerebral injury in legal contexts.

Suitability of Banana and Plantain Fruits in Modulating Neurodegenerative Diseases: Implicating the In Vitro and In Vivo Evidence from Neuroactive Narratives of Constituent Biomolecules.

Active principles in plant-based foods, especially staple fruits, such as bananas and plantains, possess inter-related anti-inflammatory, anti-apoptotic, antioxidative, and neuromodulatory activities. Neurodegenerative diseases affect the functionality of the central and peripheral nervous system, with attendant cognitive deficits being hallmarks of these conditions. The dietary constitution of a wide range of bioactive compounds identified in this review further iterates the significance of the banana and plantain in compromising, halting, or preventing the pathological mechanisms of neurological disorders. The neuroprotective mechanisms of these biomolecules have been identified by using protein expression regulation and specific gene/pathway targeting, such as the nuclear and tumor necrosis factors, extracellular signal-regulated and mitogen-activated protein kinases, activator protein-1, and the glial fibrillary acidic protein. This review establishes the potential double-edged neuro-pharmacological fingerprints of banana and plantain fruits in their traditionally consumed pulp and less utilized peel component for human nutrition.

Current state and progress of research on forensic biomechanics in China.

Forensic biomechanics gradually has become a significant component of forensic science. Forensic biomechanics is evidence-based science that applies biomechanical principles and methods to forensic practice, which has constituted one of the most potential research areas. In this review, we introduce how finite element techniques can be used to simulate forensic cases, how injury criteria and injury scales can be used to describe injury severity, and how tests of postmortem human subjects and dummy can be used to provide essential validation data. This review also describes research progress and new applications of forensic biomechanics in China.Key pointsThe review shows the main research progress and new applications of forensic biomechanics in China.The review introduces eight cases about the application of forensic biomechanics, including the multiple rigid body reconstruction, the finite element applications, study of mechanical properties, traffic crash reconstruction based on multiple techniques and analysis of morphomechanical mechanism about blood dispersal.Though forensic biomechanics has a great advantage for the evaluation of injury mechanisms, it still has some uncertainties owing to the uniqueness of the human anatomy, the complexity of biological materials, and the uncertainty of injury-causing circumstances.

Potentials of Musa Species Fruits against Oxidative Stress-Induced and Diet-Linked Chronic Diseases: In Vitro and In Vivo Implications of Micronutritional Factors and Dietary Secondary Metabolite Compounds.

Nutritional quality and the well-being of the body system are directly linked aspects of human survival. From the unborn foetus to adulthood, the need for sustainable access to micronutrient-rich foods is pertinent and the global consumption of banana and plantain fruits, in effect, contributes to the alleviation of the scourge of malnutrition. This review is particularly aimed at evaluating the pharmacological dimensions through the biological mechanisms of Musa fruits in the body, which represent correlations with their constituent micronutrient factors and dietary polyphenolic constituents such as minerals, vitamin members, anthocyanins, lutein, α-,ß- carotenes, neoxanthins and cryptoxanthins, epi- and gallo catechins, catecholamines, 3-carboxycoumarin, ß-sitosterol, monoterpenoids, with series of analytical approaches for the various identified compounds being highlighted therein. Derivative value-products from the compartments (flesh and peel) of Musa fruits are equally highlighted, bringing forth the biomedicinal and nutritional relevance, including the potentials of Musa species in dietary diversification approaches.

[Biomechanics and imaging evaluation on stable structure of pelvis].

The treatment of difficulty and prognosis of pelvic fracture are directly related to the pelvic girdle stability. Diagnosis of pelvic fracture is mainly imaging manifestations based on biomechanics of pelvic anatomy. With the progress of biomechanics experiment technology, previousopinion has changed, such as separation of symphysis pubis 2.5 cm could not be seen as distinguishing feature of type I and II for anterior-posterior compression;displacement of sacroilliac joints less than 1 cm could cause loss of vertical stability;lateral extrusion could also cause vertical instability;part description of Young-Burgess classification is not suitable for experiment results;ligament plays an important role in restricting displacement and having proprioceptors;SPECT-CT could improve sensitivity of diagnosis, but could not evaluate stability of pelvic fractures precisely.