Loading...
Loading...
Contributors:
Dataset URI:
10.5281/zenodo.23724862A curated dataset of mathematically generated synthetic spirometric curves designed to support research on respiratory pattern recognition, spirometric quality assessment, medical education, and machine learning. About the Dataset This repository presents a systematically labelled dataset comprising 3,600 synthetic forced expiratory manoeuvres generated through a demographic-conditioned mathematical modelling framework. The generation process incorporates sex, age, and height to estimate reference spirometric parameters, including Peak Expiratory Flow (PEF), Forced Vital Capacity (FVC), and Forced Expiratory Volume in one second (FEV1). Reference PEF values are obtained through demographic interpolation, while FVC and FEV1 predictions are derived from the Global Lung Function Initiative (GLI-2012) reference equations for the Caucasian population group. The proposed mathematical framework combines demographic reference values, category-dependent morphological characteristics, and stochastic variability to generate paired flow-volume (F-V) and volume-time (V-T) curves. The model was developed through an iterative refinement process involving three experienced pulmonologists, whose expert assessments contributed to improving the physiological coherence and morphological consistency of the simulated respiratory patterns. Dataset Composition The dataset comprises nine predefined spirometric categories, representing normal respiratory patterns, different degrees of airflow obstruction, and common technical errors occurring during forced expiratory manoeuvres. Each category is identified by a unique label in the dataset: Normal spirometric pattern (h): Synthetic manoeuvres representing normal spirometric behaviour. Mild obstructive pattern (pomi): Synthetic manoeuvres representing mild airflow obstruction. Moderate obstructive pattern (pomo): Synthetic manoeuvres representing moderate airflow obstruction. Severe obstructive pattern (pose): Synthetic manoeuvres representing severe airflow obstruction. Slow expiratory start (wss): Manoeuvres characterised by a delayed or insufficiently explosive onset of forced expiration. Insufficient expiratory effort (wie): Manoeuvres representing inadequate expiratory effort during the test. Variable expiratory effort (wve): Manoeuvres characterised by irregular variations in expiratory effort. Cough during forced expiration (wc): Manoeuvres incorporating simulated cough artefacts during expiration. Premature termination of expiration (wat): Manoeuvres in which forced expiration is terminated prematurely. The nine categories are organised into three broader groups: normal spirometric patterns (400 cases), obstructive patterns (1,200 cases), and technically incorrect manoeuvres (2,000 cases). Each category contains 400 synthetic manoeuvres, equally distributed between female and male demographic profiles (200 per sex). The complete dataset therefore comprises 1,800 female and 1,800 male cases, covering ages from 10 to 80 years and heights from 100 to 228 cm. The dataset is organised by target category, with each manoeuvre linked to its corresponding F-V curve, V-T curve, PNG image, and metadata record through consistent file identifiers. Data Representation and File Format Each synthetic manoeuvre is represented by two numerical arrays stored in NumPy (.npy) format: Flow–volume (F-V): A 2 × 200 array containing exhaled volume (L) and expiratory flow (L/s). Volume–time (V-T): A 2 × 200 array containing time (s) and exhaled volume (L). All arrays contain 64-bit floating-point values. The V-T representation consists of 200 uniformly spaced time samples over a six-second interval. The complete dataset therefore contains 7,200 numerical curve files, corresponding to the paired F-V and V-T representations of the 3,600 synthetic manoeuvres. Additionally, 3,600 PNG images provide graphical representations of the generated spirometric curves, including their corresponding demographic reference profiles and relevant spirometric descriptors. A central CSV metadata file contains 14 fields describing each synthetic manoeuvre, including: File identifier and target category. Demographic characteristics: sex, age, and height. Reference and generated PEF values, together with the PEF lower limit of normality (LLN). Reference and generated FVC values. Reference and generated FEV1 values. FEV1 z-score and its corresponding lower limit of normality (LLN). Additional spirometric descriptors, including the FEV1/FVC ratio, percentage-predicted FEV1, and time to peak expiratory flow, can be derived from the provided metadata and numerical curves but are not stored as independent metadata fields. License Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Copyright (c) 2026 QHPC & SP Research Lab This dataset is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License. You are free to: - Share — copy and redistribute the material in any medium or format - Adapt — remix, transform, and build upon the material Under the following terms: - Attribution — You must give appropriate credit. - NonCommercial — You may not use the material for commercial purposes. Full license text: https://creativecommons.org/licenses/by-nc/4.0/ Contact Alejandro Antonio Salvador NavarroUniversity of JaénEmail: salvador@ujaen.es Funding This work was supported by Grant PID2023-146520OB-{C21,C22}, funded by MICIU/AEI/10.13039/501100011033 and, where applicable, by “ERDF A way of making Europe”, “ERDF/EU”, “the European Union”, or “the European Union NextGenerationEU/PRTR”. Additional support was provided by grant R1.A_2025_005 from the University of Jaen under Programme R.1 UJA INVESTIGA 2025.
Keyword(s):
MeSH Term(s):
Grants and Funding:
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)