Capnography machine learning model provides high diagnostic accuracy for COPD

15th June 2023

Using capnography breath data, a machine learning algorithm diagnosed chronic obstructive pulmonary disease (COPD) with an accuracy of 91%.

In the study, published in Respiratory Research, UK researchers used the N-Tidal device and applied machine learning techniques to capnography data to help distinguish the CO₂ recordings (i.e. capnograms) of patients with and without COPD.

The team utilised capnography data from four clinical studies and developed machine learning algorithms to discriminate COPD from non-COPD, which comprised a group of patients who were either healthy or who had other conditions including asthma, heart failure, pneumonia, breathing pattern disorder and motor neurone disease.

The team developed three machine learning models and the predictability for COPD was assessed using receiver operator characteristic (ROC) curves and the subsequent estimates of sensitivity, specificity negative and positive predictive values.

Capnography machine learning accuracy

A total of 88,186 capnograms were collected from 295 patients, with each patient providing an average of 299 capnograms over 179 days.

The highest accuracy (91.3%) was provided by an XGBoost model with a corresponding sensitivity of 91.5% and a specificity of 91.4% for the diagnosis of COPD. Even on an unseen test data set, the XGBoost model still had an accuracy of 90%.

According to the manufacturer of the N-Tidal device, TidalSense, it takes under five minutes from the start of the breath test to diagnosis.

Based on the findings of the current study, the researchers concluded that the ability of the N-Tidal capnography device to accurately diagnose COPD in near-real-time lends support to its future use in a clinical setting.

COPD diagnostics in context

COPD led to 3.23 million deaths in 2019 and was the third leading cause of global deaths. Spirometry is generally considered to the be gold standard diagnostic tool for COPD, and its use is on the rise, yet it is also one of the major causes for misdiagnosis. Capnography is a widely used technique that could be used to diagnose COPD.

Diagnostic spirometry for COPD on the rise

8th June 2023

Diagnostic spirometry is increasingly used to confirm the presence of chronic obstructive pulmonary disease (COPD) but there are still existing barriers to more widespread use, according to a recent analysis.

Published in the journal NPJ Primary Care Respiratory Medicine, Swedish researchers examined whether the proportion of patients with diagnostic spirometry had increased over time. The team originally explored spirometry use in 2005 but re-assessed the level of use following the introduction of national guidelines in 2014. In the current study, they also set out to determine any factors associated with omitted or incorrectly interpreted spirometry.  

Using data from medical reviews and a questionnaire from primary and secondary care patients diagnosed with COPD between 2004 and 2010, the researchers compared the findings from a cohort diagnosed between 2000 and 2003. 

Changes in use of diagnostic spirometry

Among 703 patients with a COPD diagnosis between 2004 and 2010, 88% of these had diagnostic spirometry, compared with 59% (p < 0.001) in the previous cohort. Furthermore, the correct interpretation of spirometry results also increased between the two periods (75% vs 82%; p = 0.010).

In further analysis, it became clear that factors associated with not having diagnostic spirometry were: current smoking (Odds ratio, OR = 2.21, 95% CI 1.36 – 3.60), low educational level (OR = 1.81, 1.09 – 3.02) and being managed in primary care (OR = 2.28, 95% CI 1.02 – 5.14). The authors speculated that the lower use of spirometry in current smokers was largely because physicians probably felt the diagnosis was more likely and hence did not require confirmation.

While greater use of diagnostic spirometry was encouraging, the authors suggested that there was still a need for continuous medical educational activities to increase diagnostic accuracy.

Spirometry in context

The use of diagnostic spirometry has been advocated as a means to identify COPD in those with airflow obstruction and respiratory symptoms. However, spirometry is under-used in practice, with a real-world study finding that data from the technique was only used in 43.5% of nearly 60,000 COPD patients.

In fact, not using diagnostic spirometry potentially means that patients could be either under- or over-diagnosed with the condition. For example, it has been suggested that approximately 70% of COPD worldwide may be under-diagnosed and 30-60% of patients over-diagnosed. 

An inadequate assessment with diagnostic spirometry has important implications for patient management. For example, a late COPD diagnosis, can result in a higher exacerbation rate, increased comorbidities and costs compared with an early diagnosis.