Medical-informed machine learning: integrating prior knowledge into medical decision systems.

BACKGROUND: Clinical medicine offers a promising arena for applying Machine Learning (ML) models. However, despite numerous studies employing ML in medical data analysis, only a fraction have impacted clinical care. This article underscores the importance of utilising ML in medical data analysis, recognising that ML alone may not adequately capture the full complexity of clinical data, thereby advocating for the integration of medical domain knowledge in ML. METHODS: The study conducts a comprehensive review of prior efforts in integrating medical knowledge into ML and maps these integration strategies onto the phases of the ML pipeline, encompassing data pre-processing, feature engineering, model training, and output evaluation. The study further explores the significance and impact of such integration through a case study on diabetes prediction. Here, clinical knowledge, encompassing rules, causal networks, intervals, and formulas, is integrated at each stage of the ML pipeline, resulting in a spectrum of integrated models. RESULTS: The findings highlight the benefits of integration in terms of accuracy, interpretability, data efficiency, and adherence to clinical guidelines. In several cases, integrated models outperformed purely data-driven approaches, underscoring the potential for domain knowledge to enhance ML models through improved generalisation. In other cases, the integration was instrumental in enhancing model interpretability and ensuring conformity with established clinical guidelines. Notably, knowledge integration also proved effective in maintaining performance under limited data scenarios. CONCLUSIONS: By illustrating various integration strategies through a clinical case study, this work provides guidance to inspire and facilitate future integration efforts. Furthermore, the study identifies the need to refine domain knowledge representation and fine-tune its contribution to the ML model as the two main challenges to integration and aims to stimulate further research in this direction.

Exploring machine learning for untargeted metabolomics using molecular fingerprints.

BACKGROUND: Metabolomics, the study of substrates and products of cellular metabolism, offers valuable insights into an organism's state under specific conditions and has the potential to revolutionise preventive healthcare and pharmaceutical research. However, analysing large metabolomics datasets remains challenging, with available methods relying on limited and incompletely annotated metabolic pathways. METHODS: This study, inspired by well-established methods in drug discovery, employs machine learning on metabolite fingerprints to explore the relationship of their structure with responses in experimental conditions beyond known pathways, shedding light on metabolic processes. It evaluates fingerprinting effectiveness in representing metabolites, addressing challenges like class imbalance, data sparsity, high dimensionality, duplicate structural encoding, and interpretable features. Feature importance analysis is then applied to reveal key chemical configurations affecting classification, identifying related metabolite groups. RESULTS: The approach is tested on two datasets: one on Ataxia Telangiectasia and another on endothelial cells under low oxygen. Machine learning on molecular fingerprints predicts metabolite responses effectively, and feature importance analysis aligns with known metabolic pathways, unveiling new affected metabolite groups for further study. CONCLUSION: In conclusion, the presented approach leverages the strengths of drug discovery to address critical issues in metabolomics research and aims to bridge the gap between these two disciplines. This work lays the foundation for future research in this direction, possibly exploring alternative structural encodings and machine learning models.

Agent-based systems in healthcare.

This Special Issue is dedicated to discussing which are the advantages, challenges and open issues in the application of the agent-based approach as a part of the digital transformation in the healthcare sector. Agent-based technology in healthcare optimises resource allocation and coordination and supports clinical decision-making. Challenges, such as model reliability and interdisciplinary collaboration, must be addressed for widespread adoption. Embracing this technology promises improved healthcare delivery and better patient outcomes. Six papers, out of the many submitted, have been accepted for publication, each one discussing an aspect of this broad field.

A survey on agents applications in healthcare: Opportunities, challenges and trends.

BACKGROUND AND OBJECTIVE: The agent abstraction is a powerful one, developed decades ago to represent crucial aspects of artificial intelligence research. The meaning has transformed over the years and now there are different nuances across research communities. At its core, an agent is an autonomous computational entity capable of sensing, acting, and capturing interactions with other agents and its environment. This review examines how agent-based techniques have been implemented and evaluated in a specific and very important domain, i.e. healthcare research. METHODS: We survey key areas of agent-based research in healthcare, e.g. individual and collective behaviours, communicable and non-communicable diseases, and social epidemiology. We propose a systematic search and critical review of relevant recent works, introduced by an exploratory network analysis. RESULTS: Network analysis enables to devise out 5 main research clusters, the most active authors, and 4 main research topics. CONCLUSIONS: Our findings support discussion of some future directions for increasing the value of agent-based approaches in healthcare.

Machine Learning in Hypertension Detection: A Study on World Hypertension Day Data.

Many modifiable and non-modifiable risk factors have been associated with hypertension. However, current screening programs are still failing in identifying individuals at higher risk of hypertension. Given the major impact of high blood pressure on cardiovascular events and mortality, there is an urgent need to find new strategies to improve hypertension detection. We aimed to explore whether a machine learning (ML) algorithm can help identifying individuals predictors of hypertension. We analysed the data set generated by the questionnaires administered during the World Hypertension Day from 2015 to 2019. A total of 20206 individuals have been included for analysis. We tested five ML algorithms, exploiting different balancing techniques. Moreover, we computed the performance of the medical protocol currently adopted in the screening programs. Results show that a gain of sensitivity reflects in a loss of specificity, bringing to a scenario where there is not an algorithm and a configuration which properly outperforms against the others. However, Random Forest provides interesting performances (0.818 sensitivity - 0.629 specificity) compared with medical protocols (0.906 sensitivity - 0.230 specificity). Detection of hypertension at a population level still remains challenging and a machine learning approach could help in making screening programs more precise and cost effective, when based on accurate data collection. More studies are needed to identify new features to be acquired and to further improve the performances of ML models.

Augmenting BDI Agency with a Cognitive Service: Architecture and Validation in Healthcare Domain.

Autonomous intelligent systems are starting to influence clinical practice, as ways to both readily exploit experts' knowledge when contextual conditions demand so, and harness the overwhelming amount of patient related data currently at clinicians' disposal. However, these two approaches are rarely synergistically exploited, and tend to be used without integration. In this paper, we follow recent efforts reported in the literature regarding integration of BDI agency with machine learning based Cognitive Services, by proposing an integration architecture, and by validating such architecture in the complex domain of trauma management. In particular, we show that augmentation of a BDI agent, endowed with predefined plans encoding experts' knowledge, with a Cognitive Service, trained on past observed data, can enhance trauma management by reducing over triage episodes.

The Impact of Self-Loops on Boolean Networks Attractor Landscape and Implications for Cell Differentiation Modelling.

Boolean networks are a notable model of gene regulatory networks and, particularly, prominent theories discuss how they can capture cellular differentiation processes. One frequent motif in gene regulatory networks, especially in those circuits involved in cell differentiation, is autoregulation. In spite of this, the impact of autoregulation on Boolean network attractor landscape has not yet been extensively discussed in literature. In this paper we propose to model autoregulation as self-loops, and analyse how the number of attractors and their robustness may change once they are introduced in a well-known and widely used Boolean networks model, namely random Boolean networks. Results show that self-loops provide an evolutionary advantage in dynamic mechanisms of cells, by increasing both number and maximal robustness of attractors. These results provide evidence to the hypothesis that autoregulation is a straightforward functional component to consolidate cell dynamics, mainly in differentiation processes.

Xenobiotic-Free Medium Guarantees Expansion of Adipose Tissue-Derived Canine Mesenchymal Stem Cells Both in 3D Fibrin-Based Matrices and in 2D Plastic Surface Cultures.

Mesenchymal stem cells (MSCs) have been recently introduced in veterinary medicine as a potential therapeutic tool for several pathologies. The large-scale in vitro expansion needed to ensure the preparation of a suitable number of MSCs for clinical application usually requires the use of xenogeneic supplements like the fetal bovine serum (FBS). The substitution of FBS with species-specific supplements would improve the safety of implanted cells, reducing the risk of undesired immune responses following cell therapy. We have evaluated the effectiveness of canine adipose tissue-derived stromal vascular fraction (SVF) and MSCs (ADMSCs) expansion in the presence of canine blood-derived supplements. Cells were cultured on traditional plastic surface and inside a 3D environment derived from the jellification of different blood-derived products, i.e., platelet-poor plasma (PPP), platelet-rich plasma (PRP), or platelet lysate (PL). PPP, PRP, and PL can contribute to canine ADMSCs in vitro expansion. Both allogeneic and autologous PPP and PL can replace FBS for ADMSCs culture on a plastic surface, exhibiting either a similar (PPP) or a more effective (PL) stimulus to cell replication. Furthermore, the 3D environment based on homospecific blood-derived products polymerization provides a strong stimulus to ADMSCs replication, producing a higher number of cells in comparison to the plastic surface environment. Allogeneic or autologous blood products behave similarly. The work suggests that canine ADMSCs can be expanded in the absence of xenogeneic supplements, thus increasing the safety of cellular preparations. Furthermore, the 3D fibrin-based matrices could represent a simple, readily available environments for effective in vitro expansion of ADMSCs using allogeneic or autologous blood-products.

Complementing Agents with Cognitive Services: A Case Study in Healthcare.

Personal Agents (PAs) have longly been explored as assistants to support users in their daily activities. Surprisingly, few works refer to the adoption of PAs in the healthcare domain, where they can assist physicians' activities reducing medical errors. Although literature proposes different approaches for modelling and engineering PAs, none of them discusses how they can be integrated with cognitive services in order to empower their reasoning capabilities. In this paper we present an integration model, specifically devised for healthcare applications, that enhances Belief-Desire-Intention agents reasoning with advanced cognitive capabilities. As a case study, we adopt this integrated model in the critical care path of trauma resuscitation, stepping forward to the vision of Smart Hospitals.

On the Integration of Agents and Digital Twins in Healthcare.

A digital twin is a digital representation of a physical asset reproducing its data model, its behaviour and its communication with other physical assets. Digital twins act as a digital replica for the physical object or process they represent, providing nearly real-time monitoring and evaluation without being in close proximity. Although most of their concrete applications can be found mainly in the industrial context, healthcare represents another relevant area where digital twins can have a disruptive impact. The main research question tackled by this paper is about the integration of digital twins with agents and Multi-Agent Systems (MAS) technologies in healthcare. After providing an overview of the application of digital twins in healthcare, in this paper, we discuss our vision about agent-based digital twins, and we present a first case study, about the application of agent-based digital twins to the management of severe traumas.

Real-time tracking and documentation in trauma management.

In trauma resuscitation, an accurate documentation is crucial to improve the quality of trauma care. Hospital emergency departments typically adopt handwritten paper records and flow sheets for acquiring data, which are often inaccurate. In this article, we describe TraumaTracker, a computer-based system for trauma tracking and documentation. Results demonstrate that completeness and accuracy of trauma documentation significantly improved using TraumaTracker, since it enables to add data and information that were not recorded in paper documentation - especially precise times and locations of events.

BDI personal medical assistant agents: The case of trauma tracking and alerting.

Personal assistant agents can have an important role in healthcare as a smart technology to support physicians in their daily work, helping to tackle the increasing complexity of their task environment. In this paper we present and discuss a personal medical assistant agent technology for trauma documentation and management, based on the Belief-Desire-Intention (BDI) architecture. The purpose of the personal assistant agent is twofold: to assist the Trauma Team in doing precision tracking during a trauma resuscitation, so as to (automatically) produce an accurate documentation of the trauma, and to generate alerts at real-time, to be eventually displayed either on smart-glasses or room-display.

Modeling Intercellular Communication as a Survival Strategy of Cancer Cells: An In Silico Approach on a Flexible Bioinformatics Framework.

Intercellular communication is very important for cell development and allows a group of cells to survive as a population. Cancer cells have a similar behavior, presenting the same mechanisms and characteristics of tissue formation. In this article, we model and simulate the formation of different communication channels that allow an interaction between two cells. This is a first step in order to simulate in the future processes that occur in healthy tissue when normal cells surround a cancer cell and to interrupt the communication, thus preventing the spread of malignancy into these cells. The purpose of this study is to propose key molecules, which can be targeted to allow us to break the communication between cancer cells and surrounding normal cells. The simulation is carried out using a flexible bioinformatics platform that we developed, which is itself based on the metaphor chemistry-based model.