In a recent study published in Frontiers in Public Health, researchers demonstrated that vasculature-like signals (N) in chest computed tomography (CT) scans could function robust imaging biomarkers (IBs) of early-stage coronavirus disease 2019 (COVID-19) screening. Additionally they validated these clinical-relevant IBs, thus, opening a latest possibility to help in screening COVID-19 patients. Most significantly, these IBs showed substantial potential to scale back the workload of clinicians and assist in distinguishing COVID-19 from other pulmonary diseases.

Study: Development and validation of chest CT-based imaging biomarkers for early stage COVID-19 screening. Image Credit: Tyler Olson/Shutterstock

Background                                                                                    

There may be a brief supply of nucleic acid (NA) detection kits and specialized professionals in lots of countries, limiting the potential for COVID-19 detection at an early stage via diagnostic testing. Relatively low viral load within the early stage of the disease also leads to false negatives or limited sensitivity of reverse transcription-polymerase chain response (RT-PCR) tests. Then, a substantial proportion of patients don’t even have typical clinical symptoms throughout the onset of the disease.

Pulmonary imaging, especially chest CT scanning, could play a singular role in early-stage COVID-19 diagnosis. It detects unifocal ground-glass opacities (GGOs) within the lungs of COVID-19 patients at an early stage of infection. Because the disease progresses, GGOs infiltrate the entire lung and appear as lesions. Lung CT images could also help track lung changes in patients with COVID-19 who’ve negative NA tests.

Nevertheless, what limits using most chest CT-based computational studies is the proven fact that there may be an absence of typical characteristics in early-stage COVID-19 patients. Furthermore, patients with community-acquired pneumonia (CAP) have misleading chest CT characteristics.

Concerning the study

In the current study, researchers screened 419 patients from two hospitals in China, combining artificial intelligence (AI) and clinical findings on vascular changes within the lung regions with a system biology approach. All COVID-19 patients had confirmed diagnoses of mild to moderate illness between January 2020 and March 2020 based on the National Health Commission of the People’s Republic of China criteria. The team recruited healthy patients and CAP patients randomly from the identical two hospitals and used them as controls in training and validation cohorts independently. The control patients also had lung infections diagnosed via CT imaging just a few months before the onset of the COVID-19 epidemic.

Further, the researchers used two different CT scanners, GE Optima 660 CT and uCT 530, with a tube voltage of 120 kilovoltage peak (kVp) and reconstruction thickness of 0.625 and 1.5 mm, respectively. They recognized vasculature-like signal(s) within the patients’ pre-segmented lung regions in three dimensions (3D) using iterative tangential voting (ITV). They resampled each 3D chest CT image into isotropic image space, followed by ITV.

The team invited two radiologists with greater than two months of intense and continuous diagnosis experience of COVID-19 in Wuhan, China, to independently and blindly assess the CT images within the validation cohort of the study. Finally, they used Mann-Whitney non-parametric test to find out the difference in vasculature-like signals and the abundance of individual IBs amongst different groups. Likewise, they used logistic regression to search out an association between lung signatures and COVID-19.

Study findings

Of the 419 study participants, 116 patients from Hospital A and 303 patients from Hospital B served because the training set and a double-blind validation set, respectively. The median ages (in years) of participants in these two cohorts were 42 and 51. The share of females and males in training and validation cohorts was 45.7% vs. 54.3% and 53.1% vs. 46.9%. The variety of COVID-19 patients, healthy participants, and CAP patients within the training cohort was 47, 20, and 49, respectively. Likewise, the validation cohort had 153, 60, and 90 COVID-19 patients, healthy participants, and CAP patients, respectively.

Compared with healthy and CAP patients, COVID-19 patients had significantly more vascular changes within the lung. Intriguingly, the typical intensity of vasculature-like structures recognized and enhanced by ITV within the lung region within the training cohort revealed statistically significant differences (p < 0.05) between healthy, CAP, and COVID-19 patients.

Applying Stacked Predictive Sparse Decomposition (Stacked PSD) on the vasculature-like signal space from the training cohort uncovered eight COVID-19-relevant IBs. Each had a significantly different abundance between COVID-19 patients and others, as assessed by principal component evaluation (PCA) and clustering. A random forest classification model for COVID-19 screening based on these IBs inside the training cohort showed that every IB contributed in a different way during screening. IB-163 gave the most effective single biomarker performance [ared under the curve (AUC) = 0.893].

Even within the validation cohort, those eight pre-identified IBs clearly separated the COVID-19 patients from others. Encouragingly, the random forest model based on pre-obtained IBs predicted COVID-19 with excellent sensitivity (0.941) and accuracy (0.931), competing well with two COVID-19-experienced chest radiologists. Apart from IB-88, all IBs provided perceptual and quantitative distinctions for 2 cases misdiagnosed by participating radiologists enabling accurate screening with over 96% confidence.

Conclusions

The double-blind validation across hospitals and CT scanners confirmed that the IBs identified within the study could work robustly and effectively in real-world clinical settings. They performed superior to COVID-19 experienced chest radiologists, especially for ambiguous cases, which is common during early-stage COVID-19 screening.

Many end-to-end AI models require large training cohorts and excessive computational resources. The study developed an unsupervised learning framework with a feed-forward IB extraction strategy that involved only element-wise non-linearity and matrix multiplication. Yet, it delivered superior, scalable, and stable screening performance using a small training cohort (n = 116).

To summarize, the authors developed a sturdy, accurate, and cost-effective COVID-19 screening method that provided clinical insights beyond existing clinical practice and the scope of many existing end-to-end AI systems. With improvisations, it could facilitate the prediction of COVID-19 patients’ prognosis and clinical outcomes at an early stage.