A Model for Improving the Learning Curves of Artificial Neural Networks

Beyond multilayer perceptrons: Investigating complex topologies in neural networks

This study delves into the crucial aspect of network topology in artificial neural networks (NNs) and its impact on model performance. Addressing the need to comprehend how network structures influence learning capabilities, the research contrasts traditional multilayer perceptrons (MLPs) with models built on various complex topologies using novel network generation techniques. Drawing insights from synthetic datasets, the study reveals the remarkable accuracy of complex NNs, particularly in high-difficulty scenarios, outperforming MLPs. Our exploration extends to real-world datasets, highlighting the task-specific nature of optimal network topologies and unveiling trade-offs, including increased computational demands and reduced robustness to graph damage in complex NNs compared to MLPs. This research underscores the pivotal role of complex topologies in addressing challenging learning tasks. However, it also signals the necessity for deeper insights into the complex interplay among topological attributes influencing NN performance. By shedding light on the advantages and limitations of complex topologies, this study provides valuable guidance for practitioners and paves the way for future endeavors to design more efficient and adaptable neural architectures across various applications.

Analysis of intermunicipal journeys for cardiac surgery in Brazilian Unified Health System (SUS): an approach based on network theory

INTRODUCTION

The transformation of data into information is important to support decision making and, thus, to induce improvements in healthcare services. The regionalized organization of healthcare systems is necessary to ensure the integrity of citizen care. From this perspective, the creation of mechanisms to guide and assess the behavior of a healthcare services network becomes necessary. However, these mechanisms must consider the interaction between different municipalities. The objective of this study is to apply network analysis as a supporting tool in the Brazilian Unified Health System (Sistema Único de Saúde-SUS) management.

METHODS

The stages of the proposed method are described and applied in a real situation, analyzing the intermunicipal interaction network for cardiovascular surgery in the municipality of Vitória da Conquista, Bahia, Brazil, from 2008 to 2020. The metrics analyzed were journeys indices, flow of patients and distance of the journeys, considering the journeys from and to the municipality in focus.

RESULT

There was an increase of the incoming flow and in-degree indices combined with a decrease in outgoing flow, showing the growing importance of this municipality as a provider of these services.

CONCLUSION

The method used in the study has potential to be adopted as a management tool to assess the behavior of the interactions network of the selected service, aiding the regionalized organization of the healthcare system.