The enormous volume of data produced in the context of healthcare processes has determined the need to develop advanced analytical techniques useful for management purposes. In particular, clinical research and some medical practices are experiencing a radical change related to the introduction of learning algorithms that can facilitate the analysis of a considerable amount of data relating to healthcare. Artificial intelligence allows the integration of guidelines in decision support systems and clinical workflow promising prospects for standardization and improvement of the quality of healthcare. Through artificial intelligence techniques, it is possible to proceed with the construction of algorithms capable of learning from past events and predicting unknown events. The implementation of predictive modelling tools for clinical decision-making support represents one of the most advanced technological solutions to be developed in the software field. The application of similar systems in the clinical field and - more precisely - in the medico-legal activities of necropsy and litigation management presupposes a rigorous methodological framework that allows to overcome the skepticism shown by healthcare professionals and encourage the replacement of systems currently used. The objective of the present project is to develop an infrastructure for the development of a medico-legal decision support system field in the form of a set of objective and standardized tools. The system must be able to provide diagnostic considerations auxiliary to the autopsy activity, as well as to carry out risk assessments in healthcare liability litigation, and to find a link between exposure and possible diseases in the health surveillance of workers. The purpose corresponds to the introduction and application of scientific evidence to the typical functions of the medico-legal and occupational medicine activities carried out in the context of the National Health System.
The possibility of using reliable and accessible data for medical-legal evaluation in the hospital setting is currently characterized by a lack of concreteness as it is based on a significant empirical component and lacking in rigorous scientific models. In accordance with what is univocally expressed by the recent scientific literature, the implementation of systems for decision support in the clinical setting represents, due to the technological advancement and the amount of data available, the most promising frontier for the optimization of rigorous predictive models. For these reasons, the application potential of the analytical approach proposed in this project is high. In fact, through the development of the models, it will be possible to subject the data to complex and more accurate analyses than those carried out by the single operator with the advantage of producing results capable of assisting the professional in the different diagnostic and predictive activities related to autopsy and management of claims arising from healthcare liability. The structuring of Decision Support Systems will facilitate the integrated analysis of large amounts of data by quickly extracting information useful for decision-making processes and providing valuable aid in managing problems that are difficult to solve immediately. In addition, the objectivity of the analytical techniques used and the results obtained will eliminate the limitations inherent in the individual evaluation, notoriously influenced by the sensitivity and experience of the healthcare professional. The application of statistical and probabilistic methods to the uncertainty of diagnostic paths and to the random nature of the risk will allow the overcoming of today's epistemological limits through a methodological standardization that contemplates the systematic determination of the measurable parameters. The outlined perspective will allow a fervent scientific growth of the peculiar activities of hospital legal medicine, hesitating in the formation of evidence and in the consolidation of evaluation tools - possibly convergent and interconnected - which will confer objectivity and predictive value to medico-legal evaluations. In addition, the analysis conducted starting from the data concerning the autoptic diagnostic and litigation management activities, passing through the managerial and strategic manifestations of the same, will highlight the importance of the standardization of the operating procedures and the increasingly decisive role of legal medicine in the policies of the health systems. Ultimately, the proposed research constitutes a pioneering step towards the application of artificial intelligence to medico-legal evaluation activities, as well as an attempt to systematize decision-making processes currently based on empiricism. Finally, the planned approach will allow to precisely define the decision-making strategies developed by professionals as well as obtain judgments made through clear, reliable and evidence-based reasoning.
Finally, considering the occupational medicine field, the main innovation will be the use of the big data technologies for health and safety risks analytics in the healthcare workers domain, that consider large data sets of health and safety risks, which are usually sparse and noisy (Ajayi et al, 2019).