Distributed learning: a reliable privacy-preserving strategy to change multicenter collaborations using AI.

PURPOSE: The present scoping review aims to assess the non-inferiority of distributed learning over centrally and locally trained machine learning (ML) models in medical applications. METHODS: We performed a literature search using the term "distributed learning" OR "federated learning" in the PubMed/MEDLINE and EMBASE databases. No start date limit was used, and the search was extended until July 21, 2020. We excluded articles outside the field of interest; guidelines or expert opinion, review articles and meta-analyses, editorials, letters or commentaries, and conference abstracts; articles not in the English language; and studies not using medical data. Selected studies were classified and analysed according to their aim(s). RESULTS: We included 26 papers aimed at predicting one or more outcomes: namely risk, diagnosis, prognosis, and treatment side effect/adverse drug reaction. Distributed learning was compared to centralized or localized training in 21/26 and 14/26 selected papers, respectively. Regardless of the aim, the type of input, the method, and the classifier, distributed learning performed close to centralized training, but two experiments focused on diagnosis. In all but 2 cases, distributed learning outperformed locally trained models. CONCLUSION: Distributed learning resulted in a reliable strategy for model development; indeed, it performed equally to models trained on centralized datasets. Sensitive data can get preserved since they are not shared for model development. Distributed learning constitutes a promising solution for ML-based research and practice since large, diverse datasets are crucial for success.

Insights into how universal, tax-funded, single payer health systems manage their waiting lists: A review of the literature.

Background: A conspicuous consequence of gatekeeping arrangements in universal, tax-funded, single-payer health care systems is the long waiting times. Besides limiting equal access to care, long waiting times can have a negative impact on health outcomes. Long waiting times can create obstacles in a patient's care pathway. Organization for Economic Co-operation and Development (OECD) countries have implemented various strategies to tackle this issue, but there is little evidence for which approach is the most effective. This literature review examined waiting times for ambulatory care. Objective: The aim was to identify the main policies or combinations of policies universal, tax-funded, and single-payer healthcare systems have implemented to improve the governance of outpatient waiting times. Methods: Starting from 1040 potentially eligible articles, a total of 41 studies were identified via a 2-step selection process. Findings: Despite the relevance of the issue, the literature is limited. A set of 15 policies for the governance of ambulatory waiting time was identified and categorized by the type of intervention: generation of supply capacity, control of demand, and mixed interventions. Even if a primary intervention was always identifiable, rarely a policy was implemented solo. The most frequent primary strategies were: guidelines implementation and/or clinical pathways, including triage, guidelines for referral and maxim waiting times (14 studies), task shifting (9 studies), and telemedicine (6 studies). Most studies were observational, with no data on costs of intervention and impact on clinical outcomes.

Adversarial Data Augmentation for HMM-Based Anomaly Detection.

In this work, we concentrate on the detection of anomalous behaviors in systems operating in the physical world and for which it is usually not possible to have a complete set of all possible anomalies in advance. We present a data augmentation and retraining approach based on adversarial learning for improving anomaly detection. In particular, we first define a method for generating adversarial examples for anomaly detectors based on Hidden Markov Models (HMMs). Then, we present a data augmentation and retraining technique that uses these adversarial examples to improve anomaly detection performance. Finally, we evaluate our adversarial data augmentation and retraining approach on four datasets showing that it achieves a statistically significant performance improvement and enhances the robustness to adversarial attacks. Key differences from the state-of-the-art on adversarial data augmentation are the focus on multivariate time series (as opposed to images), the context of one-class classification (in contrast to standard multi-class classification), and the use of HMMs (in contrast to neural networks).

Combining rate-adaptive cardiac pacing algorithms via multiagent negotiation.

Simulating and controlling physiological phenomena are notoriously complex tasks to tackle and require accurate models of the phenomena of interest. Currently, most physiological processes are described by a set of partial models capturing specific aspects of the phenomena, and usually their composition does not produce effective comprehensive models. A current open issue is thus the development of techniques able to effectively describe a phenomenon starting from partial models. This is particularly relevant for heart rate regulation modeling where a large number of heterogeneous partial models exists. In this paper we make the original proposal of adopting a multiagent paradigm, called anthropic agency, to provide a powerful and flexible tool for combining partial models of heart rate regulation for adaptive cardiac pacing applications. The partial models are embedded in autonomous computational entities, called agents, that cooperatively negotiate in order to smooth their conflicts on the values of the variables forming the global model the multiagent system provides. We experimentally evaluate our approach and we analyze its properties.

Anthropic agency: a multiagent system for physiological processes.

Multiagent systems are powerful and flexible tools for modelling and regulating complex phenomena. In fact, a way to manage the complexity of a phenomenon is to decompose it in such a way that each agent embeds the control model for a portion of the phenomenon. In this perspective, the cooperative interaction among the agents results in the controller for the whole phenomenon. Since the portions in which the phenomenon is decomposed may overlap, the actions the single agents undertake to regulate these portions may conflict; hence a balanced negotiation is required. A class of complex phenomena that present several difficulties in their satisfactory modelling and controlling is the class of physiological processes. The purpose of this paper is to introduce a general multiagent architecture, called anthropic agency, for the modelling and the regulation of complex physiological phenomena.