A unified method for assessing the observability of dynamic complex systems.

PROBLEM: Systems theory applied to biology and medicine assumes that the complexity of a system can be described by quasi-generic models to predict the behavior of many other similar systems. To this end, the aim of various research works in systems theory is to develop inductive modeling (based on data-intensive analysis) or deductive modeling (based on the deduction of mechanistic principles) to discover patterns and identify plausible correlations between past and present events, or to connect different causal relationships of interacting elements at different scales and compute mathematical predictions. Mathematical principles assume that there are constant and observable universal causal principles that apply to all biological systems. Nowadays, there are no suitable tools to assess the soundness of these universal causal principles, especially considering that organisms not only respond to environmental stimuli (and inherent processes) across multiple scales but also integrate information about and within these scales. This implies an uncontrollable degree of uncertainty. METHODOLOGY: A method has been developed to detect the stability of causal processes by evaluating the information contained in the trajectories identified in a phase space. Time series patterns are analyzed using concepts from geometric information theory and persistent homology. In essence, recognizing these patterns in different time periods and evaluating their geometrically integrated information leads to the assessment of causal relationships. With this method, and together with the evaluation of persistent entropy in trajectories in relation to different individual systems, we have developed a method called Φ-S diagram as a complexity measure to recognize when organisms follow causal pathways leading to mechanistic responses. RESULTS: We calculated the Φ-S diagram of a deterministic dataset available in the ICU repository to test the method's interpretability. We also calculated the Φ-S diagram of time series from health data available in the same repository. This includes patients' physiological response to sport measured with wearables outside laboratory conditions. We confirmed the mechanistic nature of both datasets in both calculations. In addition, there is evidence that some individuals show a high degree of autonomous response and variability. Therefore, persistent individual variability may limit the ability to observe the cardiac response. In this study, we present the first demonstration of the concept of developing a more robust framework for representing complex biological systems.

Bayesian logical neural networks for human-centered applications in medicine.

Background: Medicine is characterized by its inherent uncertainty, i.e., the difficulty of identifying and obtaining exact outcomes from available data. Electronic Health Records aim to improve the exactitude of health management, for instance using automatic data recording techniques or the integration of structured as well as unstructured data. However, this data is far from perfect and is usually noisy, implying that epistemic uncertainty is almost always present in all biomedical research fields. This impairs the correct use and interpretation of the data not only by health professionals but also in modeling techniques and AI models incorporated in professional recommender systems. Method: In this work, we report a novel modeling methodology combining structural explainable models, defined on Logic Neural Networks which replace conventional deep-learning methods with logical gates embedded in neural networks, and Bayesian Networks to model data uncertainties. This means, we do not account for the variability of the input data, but we train single models according to the data and deliver different Logic-Operator neural network models that could adapt to the input data, for instance, medical procedures (Therapy Keys depending on the inherent uncertainty of the observed data. Result: Thus, our model does not only aim to assist physicians in their decisions by providing accurate recommendations; it is above all a user-centered solution that informs the physician when a given recommendation, in this case, a therapy, is uncertain and must be carefully evaluated. As a result, the physician must be a professional who does not solely rely on automatic recommendations. This novel methodology was tested on a database for patients with heart insufficiency and can be the basis for future applications of recommender systems in medicine.

Graph neural network modelling as a potentially effective method for predicting and analyzing procedures based on patients' diagnoses.

BACKGROUND: Currently, the healthcare sector strives to improve the quality of patient care management and to enhance/increase its economic performance/efficiency (e.g., cost-effectiveness) by healthcare providers. The data stored in electronic health records (EHRs) offer the potential to uncover relevant patterns relating to diseases and therapies, which in turn could help identify empirical medical guidelines to reflect best practices in a healthcare system. Based on this pattern of identification model, it is thus possible to implement recommender systems with the notion that a higher volume of procedures is often associated with better high-quality models. METHODS: Although there are several different applications that uses machine learning methods to identify such patterns, such identification is still a challenge, due in part because these methods often ignore the basic structure of the population, or even considering the similarity of diagnoses and patient typology. To this end, we have developed a method based on graph-data representation aimed to cluster 'similar' patients. Using such a model, patients will be linked when there is a same and/or similar patterns are being observed amongst them, a concept that will enable the construction of a network-like structure which is called a patient graph.1 This structure can be then analyzed by Graph Neural Networks (GNN) to identify relevant labels, and in this case the appropriate medical procedures that will be recommended. RESULTS: We were able to construct a patient graph structure based on the patient's basic information like age and gender as well as the diagnosis and the trained GNNs models to identify the corresponding patient's therapies using a synthetic patient database. We have even compared our GNN models against different baseline models (using the SCIKIT-learn library of python) and also against the performance of these different model-methods. We have found that the GNNs models are superior, with an average improvement of the f1 score of 6.48 % in respect to the baseline models. In addition, the GNNs models are useful in performing additional clustering analysis which allow a distinctive identification of specific therapeutic/treatment clusters relating to a particular combination of diagnoses. CONCLUSIONS: We found that the GNNs models offer a promising lead to model the distribution of diagnoses in patient population, and is thus a better model in identifying patients with similar phenotype based on the combination of morbidities and/or comorbidities. Nevertheless, network/graph building is still challenging and prone to biases as it is highly dependent on how the ICD distribution affects the patient network embedding space. This graph setup not only requires a high quality of the underlying diagnostic ecosystem, but it also requires a good understanding on how patients at hand are identified by disease respectively. For this reason, additional work is still needed to better improve patient embedding in graph structures for future investigations and the applications of this service-based technology. Therefore, there has not been any interventional study yet.

Elastic Multi-scale Mechanisms: Computation and Biological Evolution.

Explanations based on low-level interacting elements are valuable and powerful since they contribute to identify the key mechanisms of biological functions. However, many dynamic systems based on low-level interacting elements with unambiguous, finite, and complete information of initial states generate future states that cannot be predicted, implying an increase of complexity and open-ended evolution. Such systems are like Turing machines, that overlap with dynamical systems that cannot halt. We argue that organisms find halting conditions by distorting these mechanisms, creating conditions for a constant creativity that drives evolution. We introduce a modulus of elasticity to measure the changes in these mechanisms in response to changes in the computed environment. We test this concept in a population of predators and predated cells with chemotactic mechanisms and demonstrate how the selection of a given mechanism depends on the entire population. We finally explore this concept in different frameworks and postulate that the identification of predictive mechanisms is only successful with small elasticity modulus.

Quantitative Evaluation and Prediction of Drug Effects and Toxicological Risk Using Mechanistic Multiscale Models.

Integrating in vitro and in silico approaches has great potential for reducing experimental effort and delivering know-how and intellectual property in drug development. Here, we focus on a possible framework for multiscale modeling in pharmaceutical drug development. Looking at the modeling frameworks at different scales, it is obvious that choosing the proper level of complexity and abstraction is not a trivial task. At cellular level, we consider that the application of validated kinetic models of cellular toxicity mechanisms of drugs is particularly important for deriving valid predictions. These kinetic models can be applied for integrating inter-individual differences, e.g. obtained from data measured in surgical liver samples, into predictions of drug effects. Challenges identified include (i) the development of sufficiently detailed, structured organ models, (ii) definition of multiscale models that can be efficiently handled by available super-computing facilities, and (iii) availability of validated cell-type and organ-specific kinetic metabolic models. Multiscale models can streamline drug development by facilitating the design of experiments and trials, by providing and testing hypotheses, and by reducing time and costs due to less experiments and improved decision-making. In this review, we discuss the required pieces, possibilities, and challenges in multiscale modeling for the prediction of drug effects.

A multi-scale modeling framework for individualized, spatiotemporal prediction of drug effects and toxicological risk.

In this study, we focus on a novel multi-scale modeling approach for spatiotemporal prediction of the distribution of substances and resulting hepatotoxicity by combining cellular models, a 2D liver model, and whole body model. As a case study, we focused on predicting human hepatotoxicity upon treatment with acetaminophen based on in vitro toxicity data and potential inter-individual variability in gene expression and enzyme activities. By aggregating mechanistic, genome-based in silico cells to a novel 2D liver model and eventually to a whole body model, we predicted pharmacokinetic properties, metabolism, and the onset of hepatotoxicity in an in silico patient. Depending on the concentration of acetaminophen in the liver and the accumulation of toxic metabolites, cell integrity in the liver as a function of space and time as well as changes in the elimination rate of substances were estimated. We show that the variations in elimination rates also influence the distribution of acetaminophen and its metabolites in the whole body. Our results are in agreement with experimental results. What is more, the integrated model also predicted variations in drug toxicity depending on alterations of metabolic enzyme activities. Variations in enzyme activity, in turn, reflect genetic characteristics or diseases of individuals. In conclusion, this framework presents an important basis for efficiently integrating inter-individual variability data into models, paving the way for personalized or stratified predictions of drug toxicity and efficacy.