Artificial intelligence model on chest imaging to diagnose COVID-19 and other pneumonias: A systematic review and meta-analysis

Objectives

When diagnosing Coronavirus disease 2019(COVID‐19), radiologists cannot make an accurate judgments because the image characteristics of COVID‐19 and other pneumonia are similar. As machine learning advances, artificial intelligence(AI) models show promise in diagnosing COVID-19 and other pneumonias. We performed a systematic review and meta-analysis to assess the diagnostic accuracy and methodological quality of the models.

Methods

We searched PubMed, Cochrane Library, Web of Science, and Embase, preprints from medRxiv and bioRxiv to locate studies published before December 2021, with no language restrictions. And a quality assessment (QUADAS-2), Radiomics Quality Score (RQS) tools and CLAIM checklist were used to assess the quality of each study. We used random-effects models to calculate pooled sensitivity and specificity, I² values to assess heterogeneity, and Deeks' test to assess publication bias.

Results

We screened 32 studies from the 2001 retrieved articles for inclusion in the meta-analysis. We included 6737 participants in the test or validation group. The meta-analysis revealed that AI models based on chest imaging distinguishes COVID-19 from other pneumonias: pooled area under the curve (AUC) 0.96 (95 % CI, 0.94–0.98), sensitivity 0.92 (95 % CI, 0.88–0.94), pooled specificity 0.91 (95 % CI, 0.87–0.93). The average RQS score of 13 studies using radiomics was 7.8, accounting for 22 % of the total score. The 19 studies using deep learning methods had an average CLAIM score of 20, slightly less than half (48.24 %) the ideal score of 42.00.

Conclusions

The AI model for chest imaging could well diagnose COVID-19 and other pneumonias. However, it has not been implemented as a clinical decision-making tool. Future researchers should pay more attention to the quality of research methodology and further improve the generalizability of the developed predictive models.

Interstitial lung abnormality (ILA) and nonspecific interstitial pneumonia (NSIP)

This review article aims to address mysteries existing between Interstitial Lung Abnormality (ILA) and Nonspecific Interstitial Pneumonia (NSIP). The concept and definition of ILA are based upon CT scans from multiple large-scale cohort studies, whereas the concept and definition of NSIP originally derived from pathology with evolution to multi-disciplinary diagnosis. NSIP is the diagnosis as Interstitial Lung Disease (ILD) with clinical significance, whereas only a part of subjects with ILA have clinically significant ILD. Eventually, both ILA and NSIP must be understood in the context of chronic fibrosing ILD and progressive ILD, which remains to be further investigated.

Pneumomediastinum while using mechanical insufflation-exsufflation after recovery from riluzole-induced interstitial lung disease

We, herein, report a 61-year-old male patient with amyotrophic lateral sclerosis (ALS) complicated pneumomediastinum while using mechanical insufflation-exsufflation (MI-E) after recovery from riluzole (RZ)-induced interstitial lung disease (RZ-ILD). After the treatment of RZ-ILD, he required non-invasive mechanical ventilation (NIV) at minimal pressure settings and MI-E to manage ALS-related breathing and airway-clearance issues, respectively. After a while, he developed progressive worsening dyspnoea, and chest computed tomography revealed extensive pneumomediastinum that had spread to the area surrounding the oesophagus, the retrosternal space, and the pericardial space. He was treated with immediate discontinuation of MI-E; however, he had to keep using NIV to support his severe respiratory muscle involvement. Pneumomediastinum gradually reduced in size and no recurrence of pneumomediastinum occurred. The clinical course of our patient suggests that excessive coughing associated with MI-E combined with his previous RZ-ILD, which potentially renders his lungs vulnerable to airway pressure, may have been the aetiological factors for secondary pneumomediastinum, i.e. barotrauma. Clinicians should be aware of the risk of pneumomediastinum while using MI-E in patients with ALS, who have other pre-existing risk factors for pneumomediastinum, such as drug-induced ILD in our case.

IgG4-related disease presenting with combined pulmonary fibrosis and emphysema (CPFE)

A 64-year-old man was admitted to our hospital with an abnormal chest shadow. The patient was a current-smoker and had a past illness of autoimmune pancreatitis with a high serum level of IgG4, 348 mg/dL. Chest CT showed upper-lobe emphysema, and lower-lobe reticulation with honeycombing, suggestive of combined pulmonary fibrosis with emphysema (CPFE). Surgical lung biopsy was revealed a usual interstitial pneumonia pattern with marked infiltration of IgG4-positive plasma cells. The patient was diagnosed with IgG4 related disease (IgG4-RD) presenting with CPFE. Pulmonary manifestation was improved by corticosteroid therapy. IgG4-RD may be an underlying condition in patient with CPFE.