Deep Learning CT-based Quantitative Visualization Tool for Liver Volume Estimation: Defining Normal and Hepatomegaly

Abstract

Background Imaging assessment for hepatomegaly is not well defined and currently uses suboptimal, unidimensional measures. Liver volume provides a more direct measure for organ enlargement. Purpose To determine organ volume and to establish thresholds for hepatomegaly with use of a validated deep learning artificial intelligence tool that automatically segments the liver. Materials and Methods In this retrospective study, liver volumes were successfully derived with use of a deep learning tool for asymptomatic outpatient adults who underwent multidetector CT for colorectal cancer screening (unenhanced) or renal donor evaluation (contrast-enhanced) at a single medical center between April 2004 and December 2016. The performance of the craniocaudal and maximal three-dimensional (3D) linear measures was assessed. The manual liver volume results were compared with the automated results in a subset of renal donors in which the entire liver was included at both precontrast and postcontrast CT. Unenhanced liver volumes were standardized to a postcontrast equivalent, reflecting a correction of 3.6%. Linear regression analysis was performed to assess the major patient-specific determinant or determinants of liver volume among age, sex, height, weight, and body surface area. Results A total of 3065 patients (mean age ± standard deviation, 54 years ± 12; 1639 women) underwent multidetector CT for colorectal screening (n = 1960) or renal donor evaluation (n = 1105). The mean standardized automated liver volume ± standard deviation was 1533 mL ± 375 and demonstrated a normal distribution. Patient weight was the major determinant of liver volume and demonstrated a linear relationship. From this result, a linear weight-based upper limit of normal hepatomegaly threshold volume was derived: hepatomegaly (mL) = 14.0 × (weight [kg]) + 979. A craniocaudal threshold of 19 cm was 71% sensitive (49 of 69 patients) and 86% specific (887 of 1030 patients) for hepatomegaly, and a maximal 3D linear threshold of 24 cm was 78% sensitive (54 of 69) and 66% specific (678 of 1030). In the subset of 189 patients, the median difference in hepatic volume between the deep learning tool and the semiautomated or manual method was 2.3% (38 mL). Conclusion A simple weight-based threshold for hepatomegaly derived by using a fully automated CT-based liver volume segmentation based on deep learning provided an objective and more accurate assessment of liver size than linear measures. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Sosna in this issue.

Graphical abstract

Figure 1:

Example output from the automated tool in a 48-year-old woman weighing 89.6 kg. (A) Coronal postcontrast CT and (B) maximum-intensity projection images show fused automated liver segmentation. As seen in (C) the original postcontrast CT image, the liver has a somewhat vertical orientation but does not appear enlarged. The automated liver volume was 2142 mL, which is below the weight-based hepatomegaly threshold of 2231 mL for this patient. However, the automated linear craniocaudal measurement was elongated at 22.5 cm. This discordant liver shape has been referred to as a Riedel lobe configuration and is a normal variant.

Figure 2:

Example output from the automated tool in a 59-year-old woman weighing 85.3 kg. (A) Axial (transverse), (B) coronal, and (C) maximum-intensity projection postcontrast CT images show fused automated liver segmentation. As seen in (D) the original axial (transverse) and (E) coronal postcontrast CT images, the liver has a somewhat bulbous configuration, suggesting morphologic enlargement. The automated liver volume was 2573 mL, which is well above the weight-based hepatomegaly threshold of 2173 mL for this patient. However, the automated linear craniocaudal measurement was only 17.4 cm, thereby appearing to underestimate the liver size in this discordant case.

Figure 3:

Density plot of automated liver volumes shows the relatively normal distribution for this generally healthy adult sample. Results for noncontrast scans have been normalized and combined with postcontrast scans. Density is a unitless measure, representing fraction of cases, where the total area under the curve sums to 1.

Figure 4:

Graph shows automated CT-based liver volume according to patient weight (Wt). Weight was the dominant patient factor affecting liver volume. The solid red line represents the derived weight-based threshold for hepatomegaly based on two standard deviations above the mean (dashed line).

Figure 5:

The density plot of the subanalysis comparing automated (Auto) liver volume (orange) and manual or semiautomated (Semi) liver volume (green) show good overlap in the distributions. The median volume difference was less than 3%.