ObjectiveEndoscopic mucosal resection (EMR) is the preferred treatment for non-invasive large (>= 20 mm) non-pedunculated colorectal polyps (LNPCPs) but is associated with an early recurrence... Show moreObjectiveEndoscopic mucosal resection (EMR) is the preferred treatment for non-invasive large (>= 20 mm) non-pedunculated colorectal polyps (LNPCPs) but is associated with an early recurrence rate of up to 30%. We evaluated whether standardised EMR training could reduce recurrence rates in Dutch community hospitals.DesignIn this multicentre cluster randomised trial, 59 endoscopists from 30 hospitals were randomly assigned to the intervention group (e-learning and 2-day training including hands-on session) or control group. From April 2019 to August 2021, all consecutive EMR-treated LNPCPs were included. Primary endpoint was recurrence rate after 6 months.ResultsA total of 1412 LNPCPs were included; 699 in the intervention group and 713 in the control group (median size 30 mm vs 30 mm, 45% vs 52% size, morphology, site and access (SMSA) score IV, 64% vs 64% proximal location). Recurrence rates were lower in the intervention group compared with controls (13% vs 25%, OR 0.43; 95% CI 0.23 to 0.78; p=0.005) with similar complication rates (8% vs 9%, OR 0.93; 95% CI 0.64 to 1.36; p=0.720). Recurrences were more often unifocal in the intervention group (92% vs 76%; p=0.006). In sensitivity analysis, the benefit of the intervention on recurrence rate was only observed in the 20-40 mm LNPCPs (5% vs 20% in 20-29 mm, p=0.001; 10% vs 21% in 30-39 mm, p=0.013) but less evident in >= 40 mm LNPCPs (24% vs 31%; p=0.151). In a post hoc analysis, the training effect was maintained in the study group, while in the control group the recurrence rate remained high.ConclusionA compact standardised EMR training for LNPCPs significantly reduced recurrences in community hospitals. This strongly argues for a national dedicated training programme for endoscopists performing EMR of >= 20 mm LNPCPs. Interestingly, in sensitivity analysis, this benefit was limited for LNPCPs >= 40 mm.Trial registration numberNTR7477. Show less
Transport inspectorates are looking for novel methods to identify dangerous behavior, ultimately to reduce risks associated to the movements of people and goods. We explore a data-driven approach... Show moreTransport inspectorates are looking for novel methods to identify dangerous behavior, ultimately to reduce risks associated to the movements of people and goods. We explore a data-driven approach to arrive at smart inspections of vehicles. Inspections are smart when they are performed (1) accurate, (2) automated, (3) fair, and (4) in an interpretable manner. We leverage tools from the network science and machine learning domain to encode the behavioral aspect of vehicle’s behavior. Tools used in this thesis include community detection, link prediction, and assortativity. We explore their applicability and provide technical methods. In the final chapter, we also discuss the matter of fairness in machine learning. Show less
Cargo ships navigating global waters are required to be sufficiently safe and compliant with international treaties. Governmental inspectorates currently assess in a rule-based manner whether a... Show moreCargo ships navigating global waters are required to be sufficiently safe and compliant with international treaties. Governmental inspectorates currently assess in a rule-based manner whether a ship is potentially noncompliant and thus needs inspection. One of the dominant ship characteristics in this assessment is the ‘colour’ of the flag a ship is flying, where countries with a positive reputation have a so-called ‘white flag’. The colour of a flag may disproportionately influence the inspector, causing more frequent and stricter inspections of ships flying a non-white flag, resulting in confirmation bias in historical inspection data.In this paper, we propose an automated approach for the assessment of ship noncompliance, realising two important contributions. First, we reduce confirmation bias by using fair classifiers that decorrelate the flag from the risk classification returned by the model. Second, we extract mobility patterns from a cargo ship network, allowing us to derive meaningful features for ship classification. Crucially, these features model the behaviour of a ship, rather than its static properties. Our approach shows both a higher overall prediction performance and improved fairness with respect to the flag. Ultimately, this work enables inspectorates to better target noncompliant ships, thereby improving overall maritime safety and environmental protection. Show less
Link prediction is a well-studied technique for inferring the missing edges between two nodes in some static representation of a network. In modern day social networks, the timestamps associated... Show moreLink prediction is a well-studied technique for inferring the missing edges between two nodes in some static representation of a network. In modern day social networks, the timestamps associated with each link can be used to predict future links between so-far unconnected nodes. In these so-called temporal networks, we speak of temporal link prediction. This paper presents a systematic investigation of supervised temporal link prediction on 26 temporal, structurally diverse, real-world networks ranging from thousands to a million nodes and links. We analyse the relation between global structural properties of each network and the obtained temporal link prediction performance, employing a set of well-established topological features commonly used in the link prediction literature. We report on four contributions. First, using temporal information, an improvement of prediction performance is observed. Second, our experiments show that degree disassortative networks perform better in temporal link prediction than assortative networks. Third, we present a new approach to investigate the distinction between networks modelling discrete events and networks modelling persistent relations. Unlike earlier work, our approach utilises information on all past events in a systematic way, resulting in substantially higher link prediction performance. Fourth, we report on the influence of the temporal activity of the node or the edge on the link prediction performance, and show that the performance differs depending on the considered network type. In the studied information networks, temporal information on the node appears most important. The findings in this paper demonstrate how link prediction can effectively be improved in temporal networks, explicitly taking into account the type of connectivity modelled by the temporal edge. More generally, the findings contribute to a better understanding of the mechanisms behind the evolution of networks. Show less
In link prediction, the goal is to predict which links will appear in the future of an evolving network. To estimate the performance of these models in a supervised machine learning model, disjoint... Show moreIn link prediction, the goal is to predict which links will appear in the future of an evolving network. To estimate the performance of these models in a supervised machine learning model, disjoint and independent train and test sets are needed. However, objects in a real-world network are inherently related to each other. Therefore, it is far from trivial to separate candidate links into these disjoint sets.Here we characterize and empirically investigate the two dominant approaches from the literature for creating separate train and test sets in link prediction, referred to as random and temporal splits. Comparing the performance of these two approaches on several large temporal network datasets, we find evidence that random splits may result in too optimistic results, whereas a temporal split may give a more fair and realistic indication of performance. Results appear robust to the selection of temporal intervals. These findings will be of interest to researchers that employ link prediction or other machine learning tasks in networks. Show less
The goal of this paper is to learn the dynamics of truck co-driving behaviour. Understanding this behaviour is important because co-driving has a potential positive impact on the environment. In... Show moreThe goal of this paper is to learn the dynamics of truck co-driving behaviour. Understanding this behaviour is important because co-driving has a potential positive impact on the environment. In the so-called co-driving network, trucks are nodes while links indicate that two trucks frequently drive together. To understand the network’s dynamics, we use a link prediction approach employing a machine learning classifier. The features of the classifier can be categorized into spatio-temporal features, neighbourhood features, path features, and node features. The very different types of features allow us to understand the social processes underlying the co-driving behaviour. Our work is based on a spatio-temporal data not studied before. Data is collected from 18 million truck movements in the Netherlands. We find that co-driving behaviour is best described by using neighbourhood features, and to lesser extent by path and spatio-temporal features. Node features are deemed unimportant. Findings suggest that the dynamics of a truck co-driving network has clear social network effects. Show less
Pereira Barata, A.P.; Bruin, G.J. de; Takes, F.W.; Veenman, C.J.; Herik, H.J. van den 2018
