-
Framework based on complex networks to model and mine patient pathways
Authors:
Caroline de Oliveira Costa Souza Rosa,
Márcia Ito,
Alex Borges Vieira,
Klaus Wehmuth,
Antônio Tadeu Azevedo Gomes
Abstract:
The automatic discovery of a model to represent the history of encounters of a group of patients with the healthcare system -- the so-called "pathway of patients" -- is a new field of research that supports clinical and organisational decisions to improve the quality and efficiency of the treatment provided. The pathways of patients with chronic conditions tend to vary significantly from one perso…
▽ More
The automatic discovery of a model to represent the history of encounters of a group of patients with the healthcare system -- the so-called "pathway of patients" -- is a new field of research that supports clinical and organisational decisions to improve the quality and efficiency of the treatment provided. The pathways of patients with chronic conditions tend to vary significantly from one person to another, have repetitive tasks, and demand the analysis of multiple perspectives (interventions, diagnoses, medical specialities, among others) influencing the results. Therefore, modelling and mining those pathways is still a challenging task. In this work, we propose a framework comprising: (i) a pathway model based on a multi-aspect graph, (ii) a novel dissimilarity measurement to compare pathways taking the elapsed time into account, and (iii) a mining method based on traditional centrality measures to discover the most relevant steps of the pathways. We evaluated the framework using the study cases of pregnancy and diabetes, which revealed its usefulness in finding clusters of similar pathways, representing them in an easy-to-interpret way, and highlighting the most significant patterns according to multiple perspectives.
△ Less
Submitted 27 October, 2023; v1 submitted 25 September, 2023;
originally announced September 2023.
-
Physics-Aware Neural Networks for Boundary Layer Linear Problems
Authors:
Antonio Tadeu Azevedo Gomes,
Larissa Miguez da Silva,
Frederic Valentin
Abstract:
Physics-Informed Neural Networks (PINNs) are machine learning tools that approximate the solution of general partial differential equations (PDEs) by adding them in some form as terms of the loss/cost function of a Neural Network. Most pieces of work in the area of PINNs tackle non-linear PDEs. Nevertheless, many interesting problems involving linear PDEs may benefit from PINNs; these include para…
▽ More
Physics-Informed Neural Networks (PINNs) are machine learning tools that approximate the solution of general partial differential equations (PDEs) by adding them in some form as terms of the loss/cost function of a Neural Network. Most pieces of work in the area of PINNs tackle non-linear PDEs. Nevertheless, many interesting problems involving linear PDEs may benefit from PINNs; these include parametric studies, multi-query problems, and parabolic (transient) PDEs. The purpose of this paper is to explore PINNs for linear PDEs whose solutions may present one or more boundary layers. More specifically, we analyze the steady-state reaction-advection-diffusion equation in regimes in which the diffusive coefficient is small in comparison with the reactive or advective coefficients. We show that adding information about these coefficients as predictor variables in a PINN results in better prediction models than in a PINN that only uses spatial information as predictor variables. This finding may be instrumental in multiscale problems where the coefficients of the PDEs present high variability in small spatiotemporal regions of the domain, and therefore PINNs may be employed together with domain decomposition techniques to efficiently approximate the PDEs locally at each partition of the spatiotemporal domain, without resorting to different learned PINN models at each of these partitions.
△ Less
Submitted 15 July, 2022;
originally announced August 2022.
-
Modelling and Mining of Patient Pathways: A Sco** Review
Authors:
Caroline de Oliveira Costa Souza Rosa,
Marcia Ito,
Alex Borges Vieira,
Antonio Tadeu Azevedo Gomes
Abstract:
The sequence of visits and procedures performed by the patient in the health system, also known as the patient's pathway or trajectory, can reveal important information about the clinical treatment adopted and the health service provided. The rise of electronic health data availability made it possible to assess the pathways of a large number of patients. Nevertheless, some challenges also arose c…
▽ More
The sequence of visits and procedures performed by the patient in the health system, also known as the patient's pathway or trajectory, can reveal important information about the clinical treatment adopted and the health service provided. The rise of electronic health data availability made it possible to assess the pathways of a large number of patients. Nevertheless, some challenges also arose concerning how to synthesize these pathways and how to mine them from the data, fostering a new field of research. The objective of this review is to survey this new field of research, highlighting representation models, mining techniques, methods of analysis, and examples of case studies.
△ Less
Submitted 4 June, 2022;
originally announced June 2022.
-
Panoptic Segmentation Meets Remote Sensing
Authors:
Osmar Luiz Ferreira de Carvalho,
Osmar Abílio de Carvalho Júnior,
Cristiano Rosa e Silva,
Anesmar Olino de Albuquerque,
Nickolas Castro Santana,
Dibio Leandro Borges,
Roberto Arnaldo Trancoso Gomes,
Renato Fontes Guimarães
Abstract:
Panoptic segmentation combines instance and semantic predictions, allowing the detection of "things" and "stuff" simultaneously. Effectively approaching panoptic segmentation in remotely sensed data can be auspicious in many challenging problems since it allows continuous map** and specific target counting. Several difficulties have prevented the growth of this task in remote sensing: (a) most a…
▽ More
Panoptic segmentation combines instance and semantic predictions, allowing the detection of "things" and "stuff" simultaneously. Effectively approaching panoptic segmentation in remotely sensed data can be auspicious in many challenging problems since it allows continuous map** and specific target counting. Several difficulties have prevented the growth of this task in remote sensing: (a) most algorithms are designed for traditional images, (b) image labelling must encompass "things" and "stuff" classes, and (c) the annotation format is complex. Thus, aiming to solve and increase the operability of panoptic segmentation in remote sensing, this study has five objectives: (1) create a novel data preparation pipeline for panoptic segmentation, (2) propose an annotation conversion software to generate panoptic annotations; (3) propose a novel dataset on urban areas, (4) modify the Detectron2 for the task, and (5) evaluate difficulties of this task in the urban setting. We used an aerial image with a 0,24-meter spatial resolution considering 14 classes. Our pipeline considers three image inputs, and the proposed software uses point shapefiles for creating samples in the COCO format. Our study generated 3,400 samples with 512x512 pixel dimensions. We used the Panoptic-FPN with two backbones (ResNet-50 and ResNet-101), and the model evaluation considered semantic instance and panoptic metrics. We obtained 93.9, 47.7, and 64.9 for the mean IoU, box AP, and PQ. Our study presents the first effective pipeline for panoptic segmentation and an extensive database for other researchers to use and deal with other data or related problems requiring a thorough scene understanding.
△ Less
Submitted 30 November, 2021; v1 submitted 23 November, 2021;
originally announced November 2021.
-
Bounding Box-Free Instance Segmentation Using Semi-Supervised Learning for Generating a City-Scale Vehicle Dataset
Authors:
Osmar Luiz Ferreira de Carvalho,
Osmar Abílio de Carvalho Júnior,
Anesmar Olino de Albuquerque,
Nickolas Castro Santana,
Dibio Leandro Borges,
Roberto Arnaldo Trancoso Gomes,
Renato Fontes Guimarães
Abstract:
Vehicle classification is a hot computer vision topic, with studies ranging from ground-view up to top-view imagery. In remote sensing, the usage of top-view images allows for understanding city patterns, vehicle concentration, traffic management, and others. However, there are some difficulties when aiming for pixel-wise classification: (a) most vehicle classification studies use object detection…
▽ More
Vehicle classification is a hot computer vision topic, with studies ranging from ground-view up to top-view imagery. In remote sensing, the usage of top-view images allows for understanding city patterns, vehicle concentration, traffic management, and others. However, there are some difficulties when aiming for pixel-wise classification: (a) most vehicle classification studies use object detection methods, and most publicly available datasets are designed for this task, (b) creating instance segmentation datasets is laborious, and (c) traditional instance segmentation methods underperform on this task since the objects are small. Thus, the present research objectives are: (1) propose a novel semi-supervised iterative learning approach using GIS software, (2) propose a box-free instance segmentation approach, and (3) provide a city-scale vehicle dataset. The iterative learning procedure considered: (1) label a small number of vehicles, (2) train on those samples, (3) use the model to classify the entire image, (4) convert the image prediction into a polygon shapefile, (5) correct some areas with errors and include them in the training data, and (6) repeat until results are satisfactory. To separate instances, we considered vehicle interior and vehicle borders, and the DL model was the U-net with the Efficient-net-B7 backbone. When removing the borders, the vehicle interior becomes isolated, allowing for unique object identification. To recover the deleted 1-pixel borders, we proposed a simple method to expand each prediction. The results show better pixel-wise metrics when compared to the Mask-RCNN (82% against 67% in IoU). On per-object analysis, the overall accuracy, precision, and recall were greater than 90%. This pipeline applies to any remote sensing target, being very efficient for segmentation and generating datasets.
△ Less
Submitted 23 November, 2021;
originally announced November 2021.
-
Analyzing Flight Delay Prediction Under Concept Drift
Authors:
Lucas Giusti,
Leonardo Carvalho,
Antonio Tadeu Gomes,
Rafaelli Coutinho,
Jorge Soares,
Eduardo Ogasawara
Abstract:
Flight delays impose challenges that impact any flight transportation system. Predicting when they are going to occur is an important way to mitigate this issue. However, the behavior of the flight delay system varies through time. This phenomenon is known in predictive analytics as concept drift. This paper investigates the prediction performance of different drift handling strategies in aviation…
▽ More
Flight delays impose challenges that impact any flight transportation system. Predicting when they are going to occur is an important way to mitigate this issue. However, the behavior of the flight delay system varies through time. This phenomenon is known in predictive analytics as concept drift. This paper investigates the prediction performance of different drift handling strategies in aviation under different scales (models trained from flights related to a single airport or the entire flight system). Specifically, two research questions were proposed and answered: (i) How do drift handling strategies influence the prediction performance of delays? (ii) Do different scales change the results of drift handling strategies? In our analysis, drift handling strategies are relevant, and their impacts vary according to scale and machine learning models used.
△ Less
Submitted 4 April, 2021;
originally announced April 2021.
-
On the Implementation of a Scalable Simulator for Multiscale Hybrid-Mixed Methods
Authors:
Antonio Tadeu A. Gomes,
Weslley S. Pereira,
Frederic Valentin,
Diego Paredes
Abstract:
The family of Multiscale Hybrid-Mixed (MHM) finite element methods has received considerable attention from the mathematics and engineering community in the last few years. The MHM methods allow solving highly heterogeneous problems on coarse meshes while providing solutions with high-order precision. It embeds independent local problems which are responsible for upscaling unresolved scales into t…
▽ More
The family of Multiscale Hybrid-Mixed (MHM) finite element methods has received considerable attention from the mathematics and engineering community in the last few years. The MHM methods allow solving highly heterogeneous problems on coarse meshes while providing solutions with high-order precision. It embeds independent local problems which are responsible for upscaling unresolved scales into the numerical solution. These local contributions are brought together through a global problem defined on the skeleton of the coarse partition. Since the local problems are completely independent, they can be easily computed in parallel. In this paper, we present two simulator prototypes specifically crafted for the MHM methods, which adopt two different implementation strategies: (i) a multi-programming language approach, each language tackling different simulation issues; and (ii) a classical, single-programming language approach. Specifically, we use C++ for numerical computation of the global and local problems in a modular way; for process distribution in the simulator, we adopt the Erlang concurrent language in the first approach, and the MPI standard in the second approach. The aim of exploring these different approaches is twofold: (i) allow for the deployment of the simulator both in high-performance computing (with MPI) and in cloud computing environments (with Erlang); and (ii) pave the way for further exploration of quality attributes related to software productivity and fault-tolerance, which are key to Exascale systems. We present a performance evaluation of the two simulator prototypes taking into account their efficiency.
△ Less
Submitted 30 March, 2017;
originally announced March 2017.
-
Expressando Atributos Não-Funcionais em Workflows Científicos
Authors:
Vivian Medeiros,
Antonio Tadeu Azevedo Gomes
Abstract:
In this paper we present OSC, a scientific workflow specification language based on software architecture principles. In contrast with other approaches, OSC employs connectors as first-class constructs. In this way, we leverage reusability and compositionality in the workflow modeling process, specially in the configuration of mechanisms that manage non-functional attributes.
In this paper we present OSC, a scientific workflow specification language based on software architecture principles. In contrast with other approaches, OSC employs connectors as first-class constructs. In this way, we leverage reusability and compositionality in the workflow modeling process, specially in the configuration of mechanisms that manage non-functional attributes.
△ Less
Submitted 18 April, 2013;
originally announced April 2013.
-
DANCE: A Framework for the Distributed Assessment of Network Centralities
Authors:
Klaus Wehmuth,
Antonio Tadeu A. Gomes,
Artur Ziviani
Abstract:
The analysis of large-scale complex networks is a major challenge in the Big Data domain. Given the large-scale of the complex networks researchers commonly deal with nowadays, the use of localized information (i.e. restricted to a limited neighborhood around each node of the network) for centrality-based analysis is gaining momentum in the recent literature. In this context, we propose a framewor…
▽ More
The analysis of large-scale complex networks is a major challenge in the Big Data domain. Given the large-scale of the complex networks researchers commonly deal with nowadays, the use of localized information (i.e. restricted to a limited neighborhood around each node of the network) for centrality-based analysis is gaining momentum in the recent literature. In this context, we propose a framework for the Distributed Assessment of Network Centralities (DANCE) in complex networks. DANCE offers a single environment that allows the use of different localized centrality proposals, which can be tailored to specific applications. This environment can be thus useful given the vast potential applicability of centrality-based analysis on large-scale complex networks found in different areas, such as Biology, Physics, Sociology, or Computer Science. Since the localized centrality proposals DANCE implements employ only localized information, DANCE can easily benefit from parallel processing environments and run on different computing architectures. To illustrate this, we present a parallel implementation of DANCE and show how it can be applied to the analysis of large-scale complex networks using different kinds of network centralities. This implementation is made available to complex network researchers and practitioners interested in using it through a scientific web portal.
△ Less
Submitted 17 April, 2014; v1 submitted 4 August, 2011;
originally announced August 2011.
