Did you know that every breath we take holds the secret to our health? Nitric oxide (NO), a key biomarker, directly reflects respiratory inflammation. A recent study published in Analytical Chemistry has achieved precise detection of ultra-low NO concentrations in breath by reconstructing a neural network using multi-cycle spectral data, opening new possibilities for NO analyzer in non-invasive diagnosis of respiratory inflammatory diseases.
Breathing nitric oxide: A barometer of respiratory health
Exhaled breath acts as a mirror, accurately reflecting the physiological and pathological states of the human body. Among its many components, exhaled nitric oxide (FeNO) has become the most widely used biomarker in non-invasive diagnostics.
This compound is primarily produced by respiratory epithelial cells in response to inflammatory stimuli, with its concentration levels serving as a sensitive indicator of airway inflammation. In healthy individuals, exhaled NO concentration typically remains below 25ppb, whereas patients with respiratory conditions like asthma exhibit significantly elevated FeNO levels.
The bottleneck of the existing detection technology
The commonly used gas detection techniques include chemiluminescence, electrochemistry, photoacoustic spectroscopy, etc.
Chemiluminescence method is sensitive but easily disturbed
Electrochemical sensors are portable but require frequent calibration
High sensitivity of photoacoustic spectroscopy but sensitive to temperature and humidity
More importantly, exhaled breath contains a complex mixture of gases including nitrogen, carbon dioxide, water vapor, and ammonia. These interfering substances make it extremely challenging to detect nitric oxide (NO) at the ppb level with precision.
Technological innovation: multi-cycle spectral reconstruction neural network
To tackle this challenge, the research team developed a respiratory NO sensor based on the multi-cycle spectral reconstruction neural network (MSRNN), featuring three key breakthroughs in its core technology:
Multi-cycle spectral reconstruction: Hear the target sound from the “noisy party”
The team innovatively proposed a multi-cycle spectral reconstruction method. It’s like trying to hear someone speaking at a noisy party—where traditional methods try to filter out all the noise directly, the new method converts the target sound into a more recognizable form.
Specifically, this method transforms the spectrum from the wavelength domain to the intensity domain, enhancing the absorption characteristics of the target gas while discretizing noise and interference signals. Through a carefully designed mapping matrix, the system can accurately extract the characteristic “fingerprint” of NO from complex spectral signals.
UV segment fitting restoration: eliminate background interference
To address the discrete single-peak absorption characteristics of NO, the team employed a UV segment fitting restoration method, effectively eliminating the slow-changing absorption background in the spectrum without compromising the intensity values of the characteristic absorption peaks. This process is akin to removing background noise in photo editing to highlight the subject more prominently.
Convolutional neural network: intelligent concentration calculation
A special convolutional neural network model is constructed to establish a precise mapping relationship from spectral features to gas concentration through massive data training. The model contains two convolutional blocks, each with three convolutional layers and pooling layers, followed by a fully connected layer to output the concentration value.
Experimental verification: excellent performance
Remarkable detection accuracy: Experimental results show that the sensor achieves high-precision NO detection in the range of 1.63–846.68 ppb, with an average absolute error of only 0.31 ppb and an average absolute percentage error of 0.96%, and a detection accuracy of up to 0.63%.
Excellent stability: In the stability test, the sensor also performs well, the short-term detection coefficient of variation is only 0.40%, the long-term detection coefficient of variation is only 0.29%, the change of temperature and humidity has little effect.
Strong anti-interference capability: The sensor effectively removes ammonia interference in exhaled air through a SiO2 adsorption device, ensuring accurate and reliable detection results. Under varying humidity conditions, the coefficient of variation of detection results is only 0.66%, demonstrating excellent environmental adaptability.
Practical application: Distinguish between healthy and diseased states
In the actual exhalation test, 15 healthy volunteers participated in the experiment. The researchers compared the MSRNN-based sensor with two commercial electrochemical sensors, and the results showed a high degree of consistency among the three, demonstrating the reliability of the new sensor.
More significantly, when researchers added standard NO-simulated exhalation from patients with simulated airway inflammation, the new sensor successfully distinguished samples from healthy individuals from those of simulated patients. In ten consecutive tests, the coefficient of variation was only 1.06%, demonstrating excellent repeatability.
Technical Highlights Summary
Ultra-high sensitivity: Detects NO concentrations as low as 1.63ppb
Strong anti-interference ability: effectively deal with the interference of complex components in exhaled air
Good stability: Maintain reliable performance in changing temperature and humidity
Real-time Online Monitoring: Real-time Analysis of Exhaled Gas
look forward to the future
The successful development of this technology provides a powerful tool for the early screening and diagnosis of respiratory inflammatory diseases. Especially its non-invasive, rapid and accurate features make it have broad application prospects in clinical practice.
In the future, such sensors are expected to become routine testing equipment in hospital respiratory departments, and even developed into portable home devices, allowing patients to monitor their respiratory health at home. This will not only provide reliable basis for doctors to diagnose, but also provide convenience for patients to self-manage.
Conclusion
This study achieved precise detection of nitric oxide at ppb levels in respiratory air through multi-cycle spectral reconstruction neural network technology, breaking through the limitations of existing methods. With further refinement and broader application of this technology, we anticipate seeing its innovative use in various scenarios, revolutionizing respiratory health management.
Technological advances are making once complex medical tests simpler and more accurate, which will ultimately benefit the health of every one of us.
if you have any questions, please contact us directly!
reference material
Zhu, R., Gao, J., Tian, Q., Li, M., Xie, F., Li, C., … & Zhang, Y. (2025). Detection of Breath Nitric Oxide at Ppb Level Based on Multiperiodic Spectral Reconstruction Neural Network. Analytical Chemistry, 97(5), 3190-3197.
