Finding new ways to cost-effectively facilitate population screening and improve cancer diagnoses at an early stage supported by data-driven AI models provides unprecedented opportunities to reduce cancer related mortality. This work presents the INCISIVE project initiative towards enhancing AI solutions for health imaging by unifying, harmonizing, and securely sharing scattered cancer-related data to ensure large datasets which are critically needed to develop and evaluate trustworthy AI models. The adopted solutions of the INCISIVE project have been outlined in terms of data collection, harmonization, data sharing, and federated data storage in compliance with legal, ethical, and FAIR principles. Experiences and examples feature breast cancer data integration and mammography collection, indicating the current progress, challenges, and future directions.
Maritime traffic emissions are a major concern to governments as they heavily impact the Air Quality in coastal cities. Ships use the Automatic Identification System (AIS) to continuously report position and speed among other features, and therefore this data is suitable to be used to estimate emissions, if it is combined with engine data. However, important ship features are often inaccurate or missing. State-of-the-art complex systems, like CALIOPE at the Barcelona Supercomputing Center, are used to model Air Quality. These systems can benefit from AIS based emission models as they are very precise in positioning the pollution. Unfortunately, these models are sensitive to missing or corrupted data, and therefore they need data curation techniques to significantly improve the estimation accuracy. In this work, we propose a methodology for treating ship data using Conditional Restricted Boltzmann Machines (CRBMs) plus machine learning methods to improve the quality of data passed to emission models that can also be applied to other GPS and time-series problems. Results show that we can improve the default methods proposed to cover missing data. In our results, we observed that using our method the models boosted their accuracy to detect otherwise undetectable emissions. In particular, we used a real data-set of AIS data, provided by the Spanish Port Authority, to estimate that thanks to our method, the model was able to detect 45% of additional emissions, representing 152 tonnes of pollutants per week in Barcelona and propose new features that may enhance emission modeling.
(English) In recent years IoT and Smart Cities have become a popular paradigm of computing that is based on network-enabled devices connected providing different functionalities, from sensor measures to domotic actions. With this paradigm, it is possible to provide to the stakeholders near-realtime information of the field, e.g. the current pollution of the city. Along with the mentioned paradigms, Fog Computing enables computation near the sensors where the data is produced, i.e. Edge nodes. This paradigm provides low latency and fault tolerance given the possible independence of the sensor devices. Moreover, pushing this computation enables derived results in a near-realtime fashion. This ability to push the computation to where the data is produced can be beneficial in many situations, however it also requires to include in the Edge the data preparation processes that ensure the fitness for use of the data as the incoming data can be erroneous. Given this situation, Machine Learning can be useful to correct data and also to produce predictions of the future values. Even though there have been studies regarding on the uses of data at the Edge, to our knowledge there is no evaluation of the different modeling situations and the viability of the approach. Therefore, this thesis aims to evaluate the possibility of building a distributed system that ensures the fitness for use of the incoming data through Machine Learning enabled Data Preparation, estimates the emissions and predicts the future status of the city in a near-realtime fashion. We evaluate the viability through three contributions. The first contribution focuses on forecasting in a distributed scenario with road traffic dataset for evaluation. It provides a robust solution to build a central model. This approach is based on Federated Learning, which allows training models at the Edge nodes and then merging them centrally. This way the models in the Edge can be independent but also can be synchronized. The results show the trade-off between accuracy versions training time and a comparison between low-powered devices versus server-class machines. These analyses show that it is viable to use Machine Learning with this paradigm. The second contribution focuses on a particular use case of ship emission estimation. To estimate exhaust emissions data must be correct, which is not always the case. This contribution explores the different techniques available to correct ship registry data and proposes the usage of simple Machine Learning techniques to do imputation of missing or erroneous values. This contribution analyzes the different variables and their relationship to provide the practitioners with guidelines for correction and data treatment. The results show that with classical Machine Learning it is possible to improve the state-of-the-art results. Moreover, as these algorithms are simple enough, they can be used in an Edge device if required. The third contribution focuses on generating new variables from the ones available with a ship trace dataset obtained from the Automatic Identification System (AIS). We use a pipeline of two different methods, a Neural Networks and a clustering algorithm, to group movements into movement patterns or \emph{behaviors}. We test the predicting power of these behaviors to predict ship type, main engine power, and navigational status. The prediction of the main engine power is compared against the standard technique used in ship emission estimation when the ship registry is missing. Our approach was able to detect 45\% of the otherwise undetected emissions if the baseline method was to be used. As ship navigational status is prone to error, the behaviors found are proposed as an alternative variable based in robust data. These contributions build a framework that can distribute the learning processes and that resists network failures in low-powered devices. (Español) En los últimos años, IoT y las Smart Cities se han convertido en un paradigma popular de computación que se basa en dispositivos conectados a la red que proporcionan diferentes funcionalidades, desde medidas de sensores hasta acciones domóticas. Con este paradigma, es posible tener información en casi tiempo real, como por ejemplo la contaminación actual de la ciudad. Junto con los paradigmas mencionados, Fog Computing permite computar cerca de donde se producen los datos, es decir, los nodos Edge. Este paradigma proporciona baja latencia y tolerancia a fallos dada la posible independencia de los dispositivos sensores. Esta posibilidad puede ser beneficiosa en muchas situaciones, sin embargo, requiere incluir en el Edge los procesos de preparación de datos que aseguran la idoneidad para su uso, ya que los datos entrantes pueden ser erróneos. Ante esta situación, el Machine Learning es útil para corregir datos y también para producir predicciones de los valores futuros. A pesar de que se han realizado estudios sobre los usos de los datos en el Edge, hasta donde sabemos, no hay una evaluación de las diferentes situaciones de modelado y la viabilidad del enfoque. Por lo tanto, esta tesis tiene como objetivo evaluar la posibilidad de construir un sistema distribuido que garantice que los datos sean correctos a través de su preparación con Machine Learning. También el sistema deberá estimar las emisiones y predecir el estado futuro de la ciudad de una manera casi en tiempo real. La viabilidad se evalúa a través a través de tres contribuciones. La primera contribución se centra en escenario distribuido con un conjunto de datos de tráfico vial que proporciona una solución robusta para construir un modelo central. Este enfoque se basa en Federated Learning, que permite entrenar modelos en los nodos Edge y luego fusionarlos de forma centralizada. De esta manera, los modelos en el Edge pueden ser independientes, pero también se pueden sincronizar. Los resultados muestran la comparación de la precisión con un modelo central y uno distribuido y una comparación con dispositivos de bajo consumos contra servidores. Estos análisis muestran que es viable utilizar el Machine Learning en este paradigma. La segunda contribución se centra en un caso de uso particular de estimación de las emisiones de barcos. Para estimar las emisiones, los datos deben ser correctos, cosa que no siempre pasa. Esta contribución explora las diferentes técnicas disponibles para corregir los datos del registro de barcos y propone el uso de técnicas simples de Machine Learning para hacer imputación de valores faltantes o erróneos. Esta contribución analiza las diferentes variables y su relación para proporcionar a los profesionales pautas para la corrección y el tratamiento de datos. Los resultados muestran que con el Machine Learning clásico es posible mejorar los resultados frente a métodos del estado del arte. Además, como estos algoritmos son lo suficientemente simples como para poder utilizarse en dispositivos Edge. La tercera contribución se centra en generar nuevas variables a partir de las disponibles con un conjunto de datos de trazabilidad de barcos obtenido del Sistema AIS. Esto se hace utilizando en conjunto una red neuronal y un algoritmo de agrupación para agrupar los movimientos en patrones de movimiento o comportamientos. Se evalúa su funcionamiento para predecir el tipo de barco, la potencia del motor principal y el estado de navegación. Con esta predicción, nuestro sistema es capaz de detectar el 45% de las emisiones que no se detectan con métodos standard. Como el estado de navegación del barco es propenso a errores, los comportamientos encontrados se proponen como una variable alternativa basada en datos robustos. Estas contribuciones constituyen un marco para distribuir los procesos de aprendizaje y que resiste errores en la red con dispositivos de bajo consumo.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.