# Non-contact lung disease classification via orthogonal frequency division multiplexing-based passive 6G integrated sensing and communication

## TL;DR

This study explores using 6G/WiFi radio signals to screen for five lung diseases non-invasively, with high accuracy using machine learning.

## Contribution

A novel contactless diagnostic method using 6G/WiFi signals and deep learning for multi-disease respiratory screening.

## Key findings

- A convolutional neural network achieved 98% accuracy in classifying five respiratory diseases from radio signal data.
- Reliable screening is possible using only 12.5% of the total bandwidth, enabling dual sensing and communication.
- The method shows promise for low-cost, real-time diagnostics in resource-limited settings.

## Abstract

The screening tools for respiratory diseases typically involve spirometry (for asthma and COPD), CT scans (for interstitial lung disease), chest X-rays (for pneumonia and tuberculosis), and sputum analysis (for tuberculosis).

This work examines a diagnostic approach whereby a subject’s chest is radio-exposed to non-ionizing 6G/WiFi multi-carrier radio signals at a frequency of 5.23 GHz. The fact that each respiratory disease modulates the amplitude, frequency, and phase of each radio frequency differently allows us to screen for five respiratory diseases: asthma, chronic obstructive pulmonary disease, interstitial lung disease, pneumonia, and tuberculosis. We collect a new dataset (OFDM-Breathe) from 220 individuals in a hospital setting, including 190 patients and 30 healthy controls. The dataset contains over 26,000 s of radio signal recordings across 64 frequencies. Several machine learning and deep learning models are evaluated to classify disease type based on the discriminatory signatures of radio signals.

We learn that a vanilla convolutional neural network achieves 98% accuracy in differentiating between the five respiratory diseases, along with strong performance in precision, recall, and F1-score. An ablation study demonstrates that reliable screening with up to 96% accuracy is possible using only eight frequencies, representing just 12.5% of the total bandwidth and leaving 87.5% available for 6G/WiFi data communication.

The proposed method could enable real-time respiratory disease screening, could help realize the health equity in developing countries, and lays the groundwork for 6G/WiFi-enabled integrated sensing and communication platforms for healthcare systems of the future.

This study explores a non-ionizing, contactless screening method for five prevalent lung diseases using 6G/WiFi radio signals (to complement the conventional methods such as chest X-rays, spirometry, CT scans). It leverages the fact that each lung disease leads to an abnormal breathing pattern which in turn alters the reflected signals in a unique way. Data from 220 individuals (both healthy and patients) were collected in a clinical setting, and AI models were trained to distinguish between disease-specific patterns. The approach demonstrated high accuracy in detecting respiratory conditions and holds promise as a low-cost diagnostic tool, especially in remote or resource-limited settings (such as in developing countries). The findings also highlight the potential of 6G/WiFi signals for broader health diagnostics.

Buttar et al. explore a contactless method for screening five respiratory diseases using 6G Integrated Sensing And Communication Orthogonal Frequency Division Multiplexing (ISAC OFDM) signals from software-defined radios. Using deep learning, the approach enables accurate, real-time screening for Asthma, Chronic Obstructive Pulmonary Disease, Interstitial Lung Disease, Pneumonia and Tuberculosis.