Background: The high prevalence, morbidity and mortality, and disease heterogeneity of chronic obstructive pulmonary disease (COPD) result in the scattered data derived from patient visits in different medical units. The huge cost of integrating the scattered data for analysis and modeling, as well as the legal demand for patient privacy protection lead to the emergence of data island.

Objectives: On the premise of protecting patient privacy, integrating scattered data of patients from different medical units for high-quality modeling is beneficial to promoting the development of digital health. Based on this, we develop a distributed COPD disease diagnosis system termed COPD average federated learning (COPD_AVG_FL) using FedAvg.

Methods: First, to build the COPD_AVG_FL, the clinical data of COPD patients from the real world is collected and the data pre-processing is performed to clean the incorrect data, outlier samples and missing values. Then, a classical federated learning architecture is designed as COPD_AVG_FL. Finally, to evaluate the established COPD_AVG_FL system, we develop Centralized Machine Learning (CML).

Conclusions: Our results suggest that, with the assistance of COPD_AVG_FL, the absolute improvement rates are 13.4% (accuracy), 13.3% (precision), 12.8% (recall), 13.1% (F1-Score) and 12.9% (AUC) on the test data, respectively. The decoupling between model training and raw training data protects the patients' privacy, and helps to securely integrate more COPD data from different medical units to generate a more comprehensive model COPD_AVG_FL. This approach promotes the landing of wise information technology of medicine for COPD in the real clinical world. Code for our model will be made available at https://github.com/Cczhh/COPD_AVG_FL/tree/master.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11283156PMC
http://dx.doi.org/10.1016/j.heliyon.2024.e33566DOI Listing

Publication Analysis

Top Keywords

scattered data
12
medical units
12
data
10
machine learning
8
chronic obstructive
8
obstructive pulmonary
8
pulmonary disease
8
integrating scattered
8
patient privacy
8
federated learning
8

Similar Publications

Objectives: Hospital staff are expected to report adverse events to safety leaders and risk managers. At our institution, staff report adverse events via an incident reporting software system, known as "RL." These "RL reports" are kept separate from patient medical records.

View Article and Find Full Text PDF

During the 2016-2017 Canada Basin Acoustic Propagation Experiment, an ocean acoustic tomography array with a radius of 150 km measured the impulse responses of the ocean every 4 hr at a variety of ranges and bearings using broadband signals with center frequencies from 172.5 to 275 Hz. Ice-profiling sonar data showed a gradual increase in ice draft over the winter with daily median ice drafts reaching maxima of about 1.

View Article and Find Full Text PDF

Experimental and Theoretical Study on Electron Interactions with Acetic Acid Molecules.

J Phys Chem A

December 2024

Division of Electron Collisions Physics, Institute of Physics and Applied Computer Science, Faculty of Applied Physics and Mathematics, Gdańsk University of Technology, ul. Gabriela Narutowicza 11/12, 80-233 Gdańsk, Poland.

The absolute total cross section for electron collisions with acetic acid has been measured using an electrostatic electron spectrometer and linear transmission method for collision energies ranging from 0.4 to 300 eV. Elastic electron scattering from acetic acid within a low-energy range has also been studied theoretically using the Schwinger multichannel and R-matrix methods, in the static-exchange and static-exchange plus polarization levels of approximation for energies up to 15 eV.

View Article and Find Full Text PDF

AntiBody Sequence Database.

NAR Genom Bioinform

December 2024

Institut Pasteur, Université Paris Cité, Bioinformatics and Biostatistics Hub, 28 rue du Dr Roux, F-75015 Paris, France.

Antibodies play a crucial role in the humoral immune response against health threats, such as viral infections. Although the theoretical number of human immunoglobulins is well over a trillion, the total number of unique antibody protein sequences accessible in databases is much lower than the number found in a single individual. Training AI (Artificial Intelligence) models, for example to assist in developing serodiagnoses or antibody-based therapies, requires building datasets according to strict criteria to include as many standardized antibody sequences as possible.

View Article and Find Full Text PDF

Objective: Calcaneal fracture fixation remains a challenging procedure in orthopedics, with computational tools increasingly aiding in the optimization of preoperative planning. To compare the biomechanical stability of intramedullary fixation and locking plate fixation for Sanders II and III calcaneal fractures by three-dimensional (3D) finite element analysis and to provide a theoretical basis for clinical application.

Methods: The Computed Tomography (CT) images were segmented using Mimics software (Materialise NV, Belgium) to identify the region of interest based on threshold segmentation.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!