Related Concept Videos
Pulmonary Hypertension: Classification and Pathogenesis
There are various classifications for PH, each relating to different underlying causes and also...
Chronic Obstructive Pulmonary Disease-I: Introduction
General Characteristics of Pipe Flow I
The classification of fluid...
Chronic Obstructive Pulmonary Disease-IV: Assessement and Diagnostic Studies
Medical History
Overview of Pulmonary Circulation
The process begins with the right ventricle of the heart pumping deoxygenated blood into the pulmonary trunk. This large vessel extends about 5 centimeters before splitting into the left and right pulmonary arteries. These arteries...
Chronic Obstructive Pulmonary Disease IV: Clinical Manifestations
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Automated orbital wall fracture detection and severity classification of ocular injuries using deep learning.
Intelligent extraction of CT image landmarks for improving cam-type femoroacetabular impingement assessment.
Utilizing Deep Convolutional Neural Networks and Hybrid Classification for Gastrointestinal Disease Diagnosis from Capsule Endoscopy Images.
Related Experiment Video
Updated: Jun 4, 2026

Characterization of the Isolated, Ventilated, and Instrumented Mouse Lung Perfused with Pulsatile Flow
Published on: April 29, 2011
Classification of pulmonary system diseases patterns using flow-volume curve.
Hossein Arabalibeik1, Samaneh Jafari, Khosro Agin
1Research Center for Science and Technology in Medicine (RCSTIM), Tehran University of Medical Sciences, Tehran, Iran. arabalibeik@tums.ac.ir
This study introduces a novel method using polynomial functions and neural networks to analyze spirometry flow-volume curves, improving respiratory disease pattern detection beyond traditional parameters.
Area of Science:
- Pulmonary Medicine
- Biomedical Engineering
- Data Science
Background:
- Spirometry is a key pulmonary function test for detecting respiratory abnormalities.
- Current diagnostic systems rely on standard parameters like FEV1, FVC, and FEV1%.
- Physicians also consider the flow-volume curve's pattern for diagnosis.
Purpose of the Study:
- To develop an advanced method for respiratory disease pattern classification using spirometry.
- To evaluate the effectiveness of polynomial function coefficients from flow-volume curves in disease detection.
Main Methods:
- Fitted simple polynomial functions to spirometry flow-volume curves.
- Employed Fisher discriminant analysis to identify informative curve coefficients.
- Utilized a neural network for classifying respiratory abnormality patterns (restrictive, obstructive, mixed, normal).
Main Results:
- Polynomial function coefficients captured disease pattern information more effectively than single-point parameters.
- The neural network achieved high diagnostic performance.
- Total accuracy was 97.6%, with sensitivity at 97.5% and specificity at 98.8% for four categories.
Conclusions:
- This novel approach enhances the diagnostic accuracy of spirometry for respiratory diseases.
- Analyzing flow-volume curve patterns with polynomial functions and neural networks offers superior insights.
- The method shows significant potential for early and accurate detection of respiratory conditions.

