From Voxels to Prognosis: AI-Driven Quantitative Chest CT Analysis Forecasts ICU Requirements in 78 COVID-19 Cases

Abstract

Background: The aim of the current study was to investigate the distribution and extent of lung involvement in patients with COVID-19 with AI-supported, automated computer analysis and to assess the relationship between lung involvement and the need for intensive care unit (ICU) admission. A secondary aim was to compare the performance of computer analysis with the judgment of radiological experts.

Methods: A total of 81 patients from an open-source COVID database with confirmed COVID-19 infection were included in the study. Three patients were excluded. Lung involvement was assessed in 78 patients using computed tomography (CT) scans, and the extent of infiltration and collapse was quantified across various lung lobes and regions. The associations between lung involvement and ICU admission were analyzed. Additionally, the computer analysis of COVID-19 involvement was compared against a human rating provided by radiological experts.

Results: The results showed a higher degree of infiltration and collapse in the lower lobes compared to the upper lobes (p < 0.05) No significant difference was detected in the COVID-19-related involvement of the left and right lower lobes. The right middle lobe demonstrated lower involvement compared to the right lower lobes (p < 0.05). When examining the regions, significantly more COVID-19 involvement was found when comparing the posterior vs. the anterior halves of the lungs and the lower vs. the upper half of the lungs. Patients, who required ICU admission during their treatment exhibited significantly higher COVID-19 involvement in their lung parenchyma according to computer analysis, compared to patients who remained in general wards. Patients with more than 40% COVID-19 involvement were almost exclusively treated in intensive care. A high correlation was observed between computer detection of COVID-19 affections and expert rating by radiological experts.

Conclusion: The findings suggest that the extent of lung involvement, particularly in the lower lobes, dorsal lungs, and lower half of the lungs, may be associated with the need for ICU admission in patients with COVID-19. Computer analysis showed a high correlation with expert rating, highlighting its potential utility in clinical settings for assessing lung involvement. This information may help guide clinical decision-making and resource allocation during ongoing or future pandemics. Further studies with larger sample sizes are warranted to validate these findings.

Keywords: COVID-19; Clinical decision-making; Computer analysis; Intensive care unit (ICU) admission; Lung involvement; Radiological expert.

Figure 1

Case example: Patient #78 Lung CT Segmentation of trachea, lungs, and lobes in calibrated scalar volume

Figure 2

Case example: Same Patient #78 after Lung CT Analysis of calibrated scalar volume with signs of severe SARS-CoV-2 affection (> 50 % in both lungs). Green: Emphysema. Blue: normal lung. Orange: Infiltration. Pink: Collapse. Orange + Pink = Affected. Red: Vessels.

Figure 3

This Box and Whisker plot shows the even distribution of COVID affecting lung CT with no difference between biological sexes.

Figure 4

Correlation between baseline clinical score (by an expert radiologist) and percentage of COVID-affected lung derived by f automatic computer analysis (R = 0.86, p < 2.2e-16)

Figure 5

This box-and-whisker Plot was created to visualize the distribution of percentages of COVID-affection in each expert score group. All scores showed significantly different values with the exception of score 1 vs. 2 (* = p < 0.05)

Figure 6

This scatter plot shows the percentage of COVID-affected samples under two conditions: uncalibrated and calibrated. The data points are color-coded based on their respective conditions, with green representing the calibrated condition and orange representing the uncalibrated condition. Additionally, lines connect the data points within each pairing, illustrating the change in affected percentage between the two conditions. The plot employs a white background theme, with the x-axis labeled “Condition” and the y-axis labeled “Affected (%)”. The x-axis displays the two conditions in a discrete manner, with the order of the categories set manually. The color legend is omitted from the plot for clarity.

Figure 7

This box-and-whisker plot shows the distribution of the percentage of affected samples among patients needing or not to be admitted to the ICU. This is represented on the x-axis, while the y-axis displays the percentage of COVID-affected lung volume. Within each ICU category, a boxplot summarizes the median, quartiles, and outliers of the affected percentage. Jittered individual data points are overlaid on the boxplots, visually representing the underlying data distribution. Patients who needed ICU support had significantly higher percentages of COVID alterations in both lungs. (p < 0.05)

Figure 8

This scatter plot shows the distribution of scores between the two reviewers. Jittered individual data points are above each reviewer and identical scores are represented by green, diverging scores by orange lines (31 of 81, 38%). The second reviewer felt unable to score and excluded 3 CT scans (“NA”: 1 pneumothorax, 2 preexistent lung diseases) In the Wilcoxon signed rank test with continuity correction there was a statistically significant interobserver bias (V = 356, p-value = 0.0001666).