An AI-Enabled Bias-Free Respiratory Disease Diagnosis Model Using Cough Audio.

Bioengineering (Basel)

AI4Networks Research Center, Department of Electrical & Computer Engineering, University of Oklahoma, Tulsa, OK 74135, USA.

Published: January 2024

AI Article Synopsis

  • Researchers are developing a new AI model called RBF-Net to improve cough-based diagnosis for respiratory diseases (RDs) by addressing confounding variables that can skew results.
  • RBF-Net combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to counteract the effects of confounders like gender, age, and smoking status while predicting RDs.
  • In tests, RBF-Net demonstrated superior accuracy over existing models, showcasing its effectiveness even in biased training scenarios using diverse COVID-19 datasets.

Article Abstract

Cough-based diagnosis for respiratory diseases (RDs) using artificial intelligence (AI) has attracted considerable attention, yet many existing studies overlook confounding variables in their predictive models. These variables can distort the relationship between cough recordings (input data) and RD status (output variable), leading to biased associations and unrealistic model performance. To address this gap, we propose the Bias-Free Network (RBF-Net), an end-to-end solution that effectively mitigates the impact of confounders in the training data distribution. RBF-Net ensures accurate and unbiased RD diagnosis features, emphasizing its relevance by incorporating a COVID-19 dataset in this study. This approach aims to enhance the reliability of AI-based RD diagnosis models by navigating the challenges posed by confounding variables. A hybrid of a Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks is proposed for the feature encoder module of RBF-Net. An additional bias predictor is incorporated in the classification scheme to formulate a conditional Generative Adversarial Network (c-GAN) that helps in decorrelating the impact of confounding variables from RD prediction. The merit of RBF-Net is demonstrated by comparing classification performance with a State-of-The-Art (SoTA) Deep Learning (DL) model (CNN-LSTM) after training on different unbalanced COVID-19 data sets, created by using a large-scale proprietary cough data set. RBF-Net proved its robustness against extremely biased training scenarios by achieving test set accuracies of 84.1%, 84.6%, and 80.5% for the following confounding variables-gender, age, and smoking status, respectively. RBF-Net outperforms the CNN-LSTM model test set accuracies by 5.5%, 7.7%, and 8.2%, respectively.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10813025PMC
http://dx.doi.org/10.3390/bioengineering11010055DOI Listing

Publication Analysis

Top Keywords

confounding variables
12
test set
8
set accuracies
8
rbf-net
6
ai-enabled bias-free
4
bias-free respiratory
4
respiratory disease
4
diagnosis
4
disease diagnosis
4
model
4

Similar Publications

The Stockholm Early Detection of Cancer Study (STEADY-CAN) cohort was established to investigate strategies for early cancer detection in a population-based context within Stockholm County, the capital region of Sweden. Utilising real-world data to explore cancer-related healthcare patterns and outcomes, the cohort links extensive clinical and laboratory data from both inpatient and outpatient care in the region. The dataset includes demographic information, detailed diagnostic codes, laboratory results, prescribed medications, and healthcare utilisation data.

View Article and Find Full Text PDF

Objective: Post-cesarean delivery (CD) acute pain may progress to chronic pain, which may impair maternal bonding and child development. In 2013, we compared the efficacy of versus on-demand oral analgesia for post-caesarean pain in a randomized-controlled-trial. The fixed-time-interval group had received scheduled paracetamol, tramadol, and diclofenac regardless of pain level, and the on-demand group received medication as needed, with oxycodone reserved for unrelieved pain in both groups.

View Article and Find Full Text PDF

Intraoperative FiO and risk of impaired postoperative oxygenation in lung resection: A propensity score-weighted analysis.

J Clin Anesth

January 2025

Department of Anesthesia, Critical Care and Pain Medicine, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Electronic address:

Study Objective: To assess whether, in a lung resection cohort with a low probability of confounding by indication, higher FiO is associated with an increased risk of impaired postoperative oxygenation - a clinical manifestation of lung injury/dysfunction.

Design: Pre-specified registry-based retrospective cohort study.

Setting: Two large academic hospitals in the United States.

View Article and Find Full Text PDF

Background: Tobacco use is a known modifiable risk factor for postoperative complications and revision surgery after anterior cruciate ligament reconstruction (ACLR). Previous studies focus on tobacco as a broad categorization of traditional smoking, smokeless tobacco, and other forms of nicotine use. It is unclear if differences in the type of nicotine used lead to similar adverse outcomes after ACLR.

View Article and Find Full Text PDF

Background: The skin microbiota, a complex community of microorganisms residing on the skin, plays a crucial role in maintaining skin health and overall homeostasis. Recent research has suggested that alterations in the composition and function of the skin microbiota may influence the aging process. However, the causal relationships between specific skin microbiota and biological aging remain unclear.

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!