Black persons bear a disproportionate burden of peripheral artery disease (PAD) and experience higher rates of endovascular revascularization failure (ERF) when compared with non-Hispanic White ...persons. We aimed to identify predictors of ERF in Black persons using predictive modeling.
This retrospective study included all persons identifying as Black who underwent an initial endovascular revascularization procedure for PAD between 2011 and 2018 at a midwestern tertiary care center. Three predictive models were developed using (1) logistic regression, (2) penalized logistic regression (least absolute shrinkage and selection operator LASSO), and (3) random forest (RF). Predictive performance was evaluated under repeated cross-validation.
Of the 163 individuals included in the study, 113 (63.1%) experienced ERF at 1 y. Those with ERF had significant differences in symptom status (P < 0.001), lesion location (P < 0.001), diabetes status (P = 0.037), and annual procedural volume of the attending surgeon (P < 0.001). Logistic regression and LASSO models identified tissue loss, smoking, femoro-popliteal lesion location, and diabetes control as risk factors for ERF. The RF model identified annual procedural volume, age, PAD symptoms, number of comorbidities, and lesion location as most predictive variables. LASSO and RF models were more sensitive than logistic regression but less specific, although all three methods had an overall accuracy of ≥75%.
Black persons undergoing endovascular revascularization for PAD are at high risk of ERF, necessitating need for targeted intervention. Predictive models may be clinically useful for identifying high-risk patients, although individual predictors of ERF varied by model. Further exploration into these models may improve limb salvage for this population.
Internet gaming disorder (IGD), a worldwide mental health issue, has been widely studied using neuroimaging techniques during the last decade. Although dysfunctions in resting‐state functional ...connectivity have been reported in IGD, mapping relationships from abnormal connectivity patterns to behavioral measures have not been fully investigated. Connectome‐based predictive modeling (CPM)—a recently developed machine‐learning approach—has been used to examine potential neural mechanisms in addictions and other psychiatric disorders. To identify the resting‐state connections associated with IGD, we modified the CPM approach by replacing its core learning algorithm with a support vector machine. Resting‐state functional magnetic resonance imaging (fMRI) data were acquired in 72 individuals with IGD and 41 healthy comparison participants. The modified CPM was conducted with respect to classification and regression. A comparison of whole‐brain and network‐based analyses showed that the default‐mode network (DMN) is the most informative network in predicting IGD both in classification (individual identification accuracy = 78.76%) and regression (correspondence between predicted and actual psychometric scale score: r = 0.44, P < 0.001). To facilitate the characterization of the aberrant resting‐state activity in the DMN, the identified networks have been mapped into a three‐subsystem division of the DMN. Results suggest that individual differences in DMN function at rest could advance our understanding of IGD and variability in disorder etiology and intervention outcomes.
We integrated the connectome‐based predictive modeling approach with the support vector machine to establish the link between brain connectivity profiles and behavior in internet gaming disorder. We found that the default‐mode network is the most informative network in predicting internet gaming disorder.
This paper presents a novel approach called physics-informed probabilistic slow feature analysis. The probabilistic slow feature analysis method has been employed to extract slowly varying latent ...patterns from high-dimensional measured data. The extracted slow features have proven effective in industrial applications such as soft sensing and process monitoring. However, industrial processes come with various physical constraints that must be taken into account, such as energy requirements, equipment limitations, and safety considerations. The conventional black-box nature of the slow feature model often leads to physically inconsistent or unacceptable results. To address this issue, we propose integrating physics principles into the probabilistic slow feature model, ensuring that the extracted features adhere to physics laws. Our formulation incorporates two types of physical constraints: linear algebraic equality and inequality constraints. Through an industrial case study, we demonstrate the effectiveness of our methodology, showcasing the advantages of incorporating physics in feature extraction. These advantages include improved interpretability, reduced data dimensionality, and enhanced generalization performance.
Background
Anhedonia is a key symptom of major depressive disorder (MDD) and other psychiatric diseases. The neural basis of anhedonia has been widely examined, yet the interindividual variability in ...neuroimaging biomarkers underlying individual‐specific symptom severity is not well understood.
Methods
To establish an individualized prediction model of anhedonia, we applied connectome‐based predictive modeling (CPM) to whole‐brain resting‐state functional connectivity profiles of MDD patients.
Results
The CPM can successfully and reliably predict individual consummatory but not anticipatory anhedonia. The predictive model mainly included salience network (SN), frontoparietal network (FPN), default mode network (DMN), and motor network. Importantly, subsequent computational lesion prediction and consummatory‐specific model prediction revealed that connectivity of the SN with DMN and FPN is essential and specific for the prediction of consummatory anhedonia.
Conclusions
This study shows that brain functional connectivity, especially the connectivity of SN‐FPN and SN‐DMN, can specifically predict individualized consummatory anhedonia in MDD. These findings suggest the potential of functional connectomes for the diagnosis and prognosis of anhedonia in MDD and other disorders.
Buildings consume about 40 % of globally-produced energy. A notable amount of this energy is used to provide sufficient comfort levels to the building occupants. Moreover, given recent increases in ...global temperatures as a result of climate change and the associated decrease in comfort levels, providing adequate comfort levels in indoor spaces has become increasingly important. However, striking a balance between reducing building energy use and providing adequate comfort levels is a significant challenge. Conventional control methods for indoor spaces, such as on/off, proportional-integral (PI), and proportional-integral-derivative (PID) controllers, display significant instabilities and frequently overshoot thermostats, resulting in unnecessary energy use. Additionally, conventional building control methods rarely include comfort regulatory schemes. Consequently, recent research efforts have focused on the use of advanced artificial intelligence (AI) methods to optimize building energy usage while maintaining occupant thermal comfort. We present a review of the current AI-based methodologies being used to enhance thermal comfort in indoor spaces. we focus on thermal comfort predictive models using diverse machine learning (ML) algorithms and their deployment in building control systems for energy saving purposes. We then discuss gaps in the existing literature and highlight potential future research directions.
Land-surface parameters derived from digital land surface models (DLSMs) (for example, slope, surface curvature, topographic position, topographic roughness, aspect, heat load index, and topographic ...moisture index) can serve as key predictor variables in a wide variety of mapping and modeling tasks relating to geomorphic processes, landform delineation, ecological and habitat characterization, and geohazard, soil, wetland, and general thematic mapping and modeling. However, selecting features from the large number of potential derivatives that may be predictive for a specific feature or process can be complicated, and existing literature may offer contradictory or incomplete guidance. The availability of multiple data sources and the need to define moving window shapes, sizes, and cell weightings further complicate selecting and optimizing the feature space. This review focuses on the calculation and use of DLSM parameters for empirical spatial predictive modeling applications, which rely on training data and explanatory variables to make predictions of landscape features and processes over a defined geographic extent. The target audience for this review is researchers and analysts undertaking predictive modeling tasks that make use of the most widely used terrain variables.
To outline best practices and highlight future research needs, we review a range of land-surface parameters relating to steepness, local relief, rugosity, slope orientation, solar insolation, and moisture and characterize their relationship to geomorphic processes. We then discuss important considerations when selecting such parameters for predictive mapping and modeling tasks to assist analysts in answering two critical questions: What landscape conditions or processes does a given measure characterize? How might a particular metric relate to the phenomenon or features being mapped, modeled, or studied? We recommend the use of landscape- and problem-specific pilot studies to answer, to the extent possible, these questions for potential features of interest in a mapping or modeling task. We describe existing techniques to reduce the size of the feature space using feature selection and feature reduction methods, assess the importance or contribution of specific metrics, and parameterize moving windows or characterize the landscape at varying scales using alternative methods while highlighting strengths, drawbacks, and knowledge gaps for specific techniques. Recent developments, such as explainable machine learning and convolutional neural network (CNN)-based deep learning, may guide and/or minimize the need for feature space engineering and ease the use of DLSMs in predictive modeling tasks.
Are intelligence and creativity distinct abilities, or do they rely on the same cognitive and neural systems? We sought to quantify the extent to which intelligence and creative cognition overlap in ...brain and behavior by combining machine learning of fMRI data and latent variable modeling of cognitive ability data in a sample of young adults (N = 186) who completed a battery of intelligence and creative thinking tasks. The study had 3 analytic goals: (a) to assess contributions of specific facets of intelligence (e.g., fluid and crystallized intelligence) and general intelligence to creative ability (i.e., divergent thinking originality), (b) to model whole-brain functional connectivity networks that predict intelligence facets and creative ability, and (c) to quantify the degree to which these predictive networks overlap in the brain. Using structural equation modeling, we found moderate to large correlations between intelligence facets and creative ability, as well as a large correlation between general intelligence and creative ability (r = .63). Using connectome-based predictive modeling, we found that functional brain networks that predict intelligence facets overlap to varying degrees with a network that predicts creative ability, particularly within the prefrontal cortex of the executive control network. Notably, a network that predicted general intelligence shared 46% of its functional connections with a network that predicted creative ability-including connections linking executive control and salience/ventral attention networks-suggesting that intelligence and creative thinking rely on similar neural and cognitive systems.
Sleep is critical to a variety of cognitive functions and insufficient sleep can have negative consequences for mood and behavior across the lifespan. An important open question is how sleep duration ...is related to functional brain organization which may in turn impact cognition. To characterize the functional brain networks related to sleep across youth and young adulthood, we analyzed data from the publicly available Human Connectome Project (HCP) dataset, which includes n‐back task‐based and resting‐state fMRI data from adults aged 22–35 years (task n = 896; rest n = 898). We applied connectome‐based predictive modeling (CPM) to predict participants' mean sleep duration from their functional connectivity patterns. Models trained and tested using 10‐fold cross‐validation predicted self‐reported average sleep duration for the past month from n‐back task and resting‐state connectivity patterns. We replicated this finding in data from the 2‐year follow‐up study session of the Adolescent Brain Cognitive Development (ABCD) Study, which also includes n‐back task and resting‐state fMRI for adolescents aged 11–12 years (task n = 786; rest n = 1274) as well as Fitbit data reflecting average sleep duration per night over an average duration of 23.97 days. CPMs trained and tested with 10‐fold cross‐validation again predicted sleep duration from n‐back task and resting‐state functional connectivity patterns. Furthermore, demonstrating that predictive models are robust across independent datasets, CPMs trained on rest data from the HCP sample successfully generalized to predict sleep duration in the ABCD Study sample and vice versa. Thus, common resting‐state functional brain connectivity patterns reflect sleep duration in youth and young adults.
Resting‐state functional connections that predict more sleep (left) and less sleep (right) in the Human Connectome Project and Adolescent Brain Cognitive Development Study datasets.
One of the challenges with predictive modeling is how to quantify the reliability of the models' predictions on new objects. In this work we give an introduction to conformal prediction, a framework ...that sits on top of traditional machine learning algorithms and which outputs valid confidence estimates to predictions from QSAR models in the form of prediction intervals that are specific to each predicted object. For regression, a prediction interval consists of an upper and a lower bound. For classification, a prediction interval is a set that contains none, one, or many of the potential classes. The size of the prediction interval is affected by a user-specified confidence/significance level, and by the nonconformity of the predicted object; i.e., the strangeness as defined by a nonconformity function. Conformal prediction provides a rigorous and mathematically proven framework for in silico modeling with guarantees on error rates as well as a consistent handling of the models’ applicability domain intrinsically linked to the underlying machine learning model. Apart from introducing the concepts and types of conformal prediction, we also provide an example application for modeling ABC transporters using conformal prediction, as well as a discussion on general implications for drug discovery.
ABSTRACT
Background
Although coronavirus disease 2019 (COVID-19) patients who develop in-hospital acute kidney injury (AKI) have worse short-term outcomes, their long-term outcomes have not been ...fully characterized. We investigated 90-day and 1-year outcomes after hospital AKI grouped by time to recovery from AKI.
Methods
This study consisted of 3296 COVID-19 patients with hospital AKI stratified by early recovery (<48 hours), delayed recovery (2–7 days) and prolonged recovery (>7–90 days). Demographics, comorbidities and laboratory values were obtained at admission and up to the 1-year follow-up. The incidence of major adverse cardiovascular events (MACE) and major adverse kidney events (MAKE), rehospitalization, recurrent AKI and new-onset chronic kidney disease (CKD) were obtained 90-days after COVID-19 discharge.
Results
The incidence of hospital AKI was 28.6%. Of the COVID-19 patients with AKI, 58.0% experienced early recovery, 14.8% delayed recovery and 27.1% prolonged recovery. Patients with a longer AKI recovery time had a higher prevalence of CKD (P < .05) and were more likely to need invasive mechanical ventilation (P < .001) and to die (P < .001). Many COVID-19 patients developed MAKE, recurrent AKI and new-onset CKD within 90 days, and these incidences were higher in the prolonged recovery group (P < .05). The incidence of MACE peaked 20–40 days postdischarge, whereas MAKE peaked 80–90 days postdischarge. Logistic regression models predicted 90-day MACE and MAKE with 82.4 ± 1.6% and 79.6 ± 2.3% accuracy, respectively.
Conclusion
COVID-19 survivors who developed hospital AKI are at high risk for adverse cardiovascular and kidney outcomes, especially those with longer AKI recovery times and those with a history of CKD. These patients may require long-term follow-up for cardiac and kidney complications.
Graphical Abstract
Graphical Abstract