COVID-19 pneumonia with incidentally detected pulmonary nodules

Campos, Rúbia, MD1; Bertolazzi, Pâmela, BS2; Rafael Campos Vitorino3

1 Radiology Department, Multimagem, Salvador, Bahia, Brazil

2 Siemens Healthineers, LAM

3 Escola Bahiana de Medicina e Saúde Pública, Salvador, Bahia, Brazil

25.02.2022

A 75-year-old woman, suffering from fever and progressive dyspnea for the past 48 hours, presented herself to the hospital. A RT-PCR test revealed a positive result, and a diagnosis of COVID-19 was confirmed. Her respiratory distress acutely worsened, and her blood oxygen saturation level dropped. Therefore, a chest CT examination was requested to investigate a suspected pneumonia.

CT images revealed bilateral, subpleural foci of ground-glass opacities (GGO) with a predominant distribution in the basal segments of both lower lobes, quantifying 8.52% of the total lung volume. According to the classification of the local medical community – mild (< 25% of parenchyma affected), moderate (between 25-50%) and severe (> 50% parenchyma affected) – this is considered as a mild degree of COVID lung infection. Two subpleural, well-delineated pulmonary nodules, measuring 12.7 mm and 8.7 mm in maximum 3D diameter, adjacent to the GGOs in the basal lateral segment of the left lower lobe, were identified and segmented by AI-Rad Companion Chest CT. A diagnosis of COVID-19 pneumonia was suggested, and the pulmonary nodules remained undefined. As the patient had no known neoplastic disease, the possibility of inflammatory nodules was considered and a follow-up CT scan in 3 months was recommended, according to the Fleischner Guideline. [1]

Two well-delineated subpleural pulmonary nodules, in the lateral basal segment of the left lower lobe, are visualized and segmented. The volume and the size of the nodules are automatically measured and summarized by AI-Rad Companion Chest CT.

Courtesy of Radiology Department, Multimagem, Salvador, Bahia, Brazil

Fig. 1: Two well-delineated subpleural pulmonary nodules, in the lateral basal segment of the left lower lobe, are visualized (Fig. 1a & 1c, arrows) and segmented (Fig. 1b & 1d). The volume and the size of the nodules are automatically measured and summarized (Fig. 1e) by AI-Rad Companion Chest CT.

The lungs and the areas of GGO are segmented and measured automatically by AI-Rad Companion Chest CT and the standardized results are shown.

Courtesy of Radiology Department, Multimagem, Salvador, Bahia, Brazil

Fig. 2: The lungs and the areas of GGO are segmented and measured automatically (Fig. 2a) by AI-Rad Companion Chest CT and the standardized results (Fig. 2b) are shown.

Coronal and sagittal MPR images show the two pulmonary nodules in the left lower lobe, which are highlighted by AI-Rad Companion Chest CT in a VRT image.

Courtesy of Radiology Department, Multimagem, Salvador, Bahia, Brazil

Fig. 3: Coronal (Fig. 3a) and sagittal (Fig. 3b) MPR images show the two pulmonary nodules (arrows) in the left lower lobe, which are highlighted by AI-Rad Companion Chest CT in a VRT image (Fig. 3c).

Incidental findings (IFs) are classified as secondary findings that are not related to the main complaint or to emergency patient care. [2] They are common in the radiologist's interpretation of CT scans and, once detected, cannot be ignored, as early imaging can significantly affect disease behavior and prognosis. IFs might need follow-up control or treatment, such as the small pulmonary nodules in this case. Although CT imaging can capture even small IFs, owing to its high resolution, there is a relevant risk of under-detection in an emergency setting when radiologists must report in a time-critical situation. [3] AI-Rad Companion Chest CT is a secondary reading device with integrated algorithms supported by Artificial Intelligence. It can assist the radiologists in analysing the lung parenchyma by segmenting the lung lobes, areas with predefined densities and lung nodules, performing automatic measurements of volume, size and density, and providing standardized results. This approach has the potential to reduce the number of missed secondary findings in clinical emergency settings that require a very time-critical radiological reporting. In this case, the areas of the GGOs, as well as the two small pulmonary nodules, are identified, segmented and quantified automatically by AI-Rad Companion Chest CT. This improves the workflow and the diagnostic confidence.

Scanner

Scan area

Thorax

Scan mode

Spiral

Scan length

359 mm

Scan direction

Caudo-cranial

Scan time

8.4 s

Tube voltage

110 kV

Effective mAs

28 mAs

Dose modulation

CARE Dose4D

CTDIvol

2.0 mGy

DLP

79 mGy*cm

Rotation time

0.8 s

Pitch

1.5

Slice collimation

32 x 0.7 mm

Slice width

1.0 mm

Reconstruction increment

0.7 mm

Reconstruction kernel

Hr64