Search Results for author: Rossella Arcucci

Found 31 papers, 10 papers with code

Zero-Shot ECG Classification with Multimodal Learning and Test-time Clinical Knowledge Enhancement

no code implementations11 Mar 2024 Che Liu, Zhongwei Wan, Cheng Ouyang, Anand Shah, Wenjia Bai, Rossella Arcucci

Through multimodal learning on ECG records and associated reports, MERL is capable of performing zero-shot ECG classification with text prompts, eliminating the need for training data in downstream tasks.

Clinical Knowledge Descriptive +5

Electrocardiogram Instruction Tuning for Report Generation

no code implementations7 Mar 2024 Zhongwei Wan, Che Liu, Xin Wang, Chaofan Tao, Hui Shen, Zhenwu Peng, Jie Fu, Rossella Arcucci, Huaxiu Yao, Mi Zhang

Electrocardiogram (ECG) serves as the primary non-invasive diagnostic tool for cardiac conditions monitoring, are crucial in assisting clinicians.

Multi-fidelity physics constrained neural networks for dynamical systems

no code implementations3 Feb 2024 Hao Zhou, Sibo Cheng, Rossella Arcucci

As a result, during the training of predictive models, physical constraints can be evaluated within low-fidelity spaces, yielding a trade-off between training efficiency and accuracy.

Data Assimilation using ERA5, ASOS, and the U-STN model for Weather Forecasting over the UK

1 code implementation15 Jan 2024 Wenqi Wang, Jacob Bieker, Rossella Arcucci, César Quilodrán-Casas

In recent years, the convergence of data-driven machine learning models with Data Assimilation (DA) offers a promising avenue for enhancing weather forecasting.

Weather Forecasting

Freeze the backbones: A Parameter-Efficient Contrastive Approach to Robust Medical Vision-Language Pre-training

no code implementations2 Jan 2024 Jiuming Qin, Che Liu, Sibo Cheng, Yike Guo, Rossella Arcucci

Modern healthcare often utilises radiographic images alongside textual reports for diagnostics, encouraging the use of Vision-Language Self-Supervised Learning (VL-SSL) with large pre-trained models to learn versatile medical vision representations.

Image Classification Image Segmentation +5

G2D: From Global to Dense Radiography Representation Learning via Vision-Language Pre-training

no code implementations3 Dec 2023 Che Liu, Cheng Ouyang, Sibo Cheng, Anand Shah, Wenjia Bai, Rossella Arcucci

G2D achieves superior performance across 6 medical imaging tasks and 25 diseases, particularly in semantic segmentation, which necessitates fine-grained, semantically-grounded image features.

object-detection Object Detection +5

Efficient deep data assimilation with sparse observations and time-varying sensors

1 code implementation24 Oct 2023 Sibo Cheng, Che Liu, Yike Guo, Rossella Arcucci

We introduce a novel variational DA scheme, named Voronoi-tessellation Inverse operator for VariatIonal Data assimilation (VIVID), that incorporates a DL inverse operator into the assimilation objective function.

IMITATE: Clinical Prior Guided Hierarchical Vision-Language Pre-training

no code implementations11 Oct 2023 Che Liu, Sibo Cheng, Miaojing Shi, Anand Shah, Wenjia Bai, Rossella Arcucci

The framework derives multi-level visual features from the chest X-ray (CXR) images and separately aligns these features with the descriptive and the conclusive text encoded in the hierarchical medical report.

Contrastive Learning Descriptive

Utilizing Synthetic Data for Medical Vision-Language Pre-training: Bypassing the Need for Real Images

no code implementations10 Oct 2023 Che Liu, Anand Shah, Wenjia Bai, Rossella Arcucci

The advent of text-guided generative models raises a compelling question: Can VLP be implemented solely with synthetic images generated from genuine radiology reports, thereby mitigating the need for extensively pairing and curating image-text datasets?

Image Classification object-detection +2

ETP: Learning Transferable ECG Representations via ECG-Text Pre-training

no code implementations6 Sep 2023 Che Liu, Zhongwei Wan, Sibo Cheng, Mi Zhang, Rossella Arcucci

In the domain of cardiovascular healthcare, the Electrocardiogram (ECG) serves as a critical, non-invasive diagnostic tool.

Language Modelling Representation Learning +2

A generative model for surrogates of spatial-temporal wildfire nowcasting

no code implementations5 Aug 2023 Sibo Cheng, Yike Guo, Rossella Arcucci

The model is tested in the ecoregion of a recent massive wildfire event in California, known as the Chimney fire.

Temporal Sequences

M-FLAG: Medical Vision-Language Pre-training with Frozen Language Models and Latent Space Geometry Optimization

1 code implementation17 Jul 2023 Che Liu, Sibo Cheng, Chen Chen, Mengyun Qiao, Weitong Zhang, Anand Shah, Wenjia Bai, Rossella Arcucci

The proposed method, named Medical vision-language pre-training with Frozen language models and Latent spAce Geometry optimization (M-FLAG), leverages a frozen language model for training stability and efficiency and introduces a novel orthogonality loss to harmonize the latent space geometry.

Image Classification Language Modelling +3

A novel approach for predicting epidemiological forecasting parameters based on real-time signals and Data Assimilation

no code implementations3 Jul 2023 Romain Molinas, César Quilodrán Casas, Rossella Arcucci, Ovidiu Şerban

This paper proposes a novel approach to predict epidemiological parameters by integrating new real-time signals from various sources of information, such as novel social media-based population density maps and Air Quality data.

Decision Making

Generative Text-Guided 3D Vision-Language Pretraining for Unified Medical Image Segmentation

no code implementations7 Jun 2023 Yinda Chen, Che Liu, Wei Huang, Sibo Cheng, Rossella Arcucci, Zhiwei Xiong

To address these challenges, we present Generative Text-Guided 3D Vision-Language Pretraining for Unified Medical Image Segmentation (GTGM), a framework that extends of VLP to 3D medical images without relying on paired textual descriptions.

Computed Tomography (CT) Contrastive Learning +4

Med-UniC: Unifying Cross-Lingual Medical Vision-Language Pre-Training by Diminishing Bias

1 code implementation NeurIPS 2023 Zhongwei Wan, Che Liu, Mi Zhang, Jie Fu, Benyou Wang, Sibo Cheng, Lei Ma, César Quilodrán-Casas, Rossella Arcucci

Med-UniC reaches superior performance across 5 medical image tasks and 10 datasets encompassing over 30 diseases, offering a versatile framework for unifying multi-modal medical data within diverse linguistic communities.

Disentanglement

Frozen Language Model Helps ECG Zero-Shot Learning

no code implementations22 Mar 2023 Jun Li, Che Liu, Sibo Cheng, Rossella Arcucci, Shenda Hong

In downstream classification tasks, METS achieves around 10% improvement in performance without using any annotated data via zero-shot classification, compared to other supervised and SSL baselines that rely on annotated data.

Language Modelling Self-Supervised Learning +1

Spectral Cross-Domain Neural Network with Soft-adaptive Threshold Spectral Enhancement

1 code implementation10 Jan 2023 Che Liu, Sibo Cheng, Weiping Ding, Rossella Arcucci

The robust performance of SCDNN provides a new perspective to exploit knowledge across deep learning models from time and spectral domains.

Electrocardiography (ECG) Feature Engineering +1

Generalised Latent Assimilation in Heterogeneous Reduced Spaces with Machine Learning Surrogate Models

no code implementations7 Apr 2022 Sibo Cheng, Jianhua Chen, Charitos Anastasiou, Panagiota Angeli, Omar K. Matar, Yi-Ke Guo, Christopher C. Pain, Rossella Arcucci

The new approach is tested on a high-dimensional CFD application of a two-phase liquid flow with non-linear observation operators that current Latent Assimilation methods can not handle.

BIG-bench Machine Learning

Forecasting emissions through Kaya identity using Neural Ordinary Differential Equations

no code implementations7 Jan 2022 Pierre Browne, Aranildo Lima, Rossella Arcucci, César Quilodrán-Casas

Starting from the Kaya identity, we used a Neural ODE model to predict the evolution of several indicators related to carbon emissions, on a country-level: population, GDP per capita, energy intensity of GDP, carbon intensity of energy.

Correcting public opinion trends through Bayesian data assimilation

no code implementations29 May 2021 Robin Hendrickx, Rossella Arcucci, Julio Amador Dıaz Lopez, Yi-Ke Guo, Mark Kennedy

Traditional survey polling remains the most popular estimation technique, despite its cost and time intensity, measurement errors, lack of real-time capabilities and lagged representation of public opinion.

Opinion Mining

Adversarial autoencoders and adversarial LSTM for improved forecasts of urban air pollution simulations

1 code implementation13 Apr 2021 César Quilodrán-Casas, Rossella Arcucci, Laetitia Mottet, Yike Guo, Christopher Pain

Our two-step method integrates a Principal Components Analysis (PCA) based adversarial autoencoder (PC-AAE) with adversarial Long short-term memory (LSTM) networks.

Time Series Time Series Analysis

Attention-based Convolutional Autoencoders for 3D-Variational Data Assimilation

1 code implementation6 Jan 2021 Julian Mack, Rossella Arcucci, Miguel Molina-Solana, Yi-Ke Guo

We propose a new 'Bi-Reduced Space' approach to solving 3D Variational Data Assimilation using Convolutional Autoencoders.

Adversarially trained LSTMs on reduced order models of urban air pollution simulations

no code implementations5 Jan 2021 César Quilodrán-Casas, Rossella Arcucci, Christopher Pain, Yike Guo

This adversarially trained LSTM-based approach is used on the ROM in order to produce faster forecasts of the air pollution tracer.

An Epidemiological Modelling Approach for Covid19 via Data Assimilation

1 code implementation25 Apr 2020 Philip Nadler, Shuo Wang, Rossella Arcucci, Xian Yang, Yike Guo

We compare and discuss model results which conducts updates as new observations become available.

Cannot find the paper you are looking for? You can Submit a new open access paper.