Deep learning CT reconstruction improves liver metastases detection

Abstract

Objectives: Detection of liver metastases is crucial for guiding oncological management. Computed tomography through iterative reconstructions is widely used in this indication but has certain limitations. Deep learning image reconstructions (DLIR) use deep neural networks to achieve a significant noise reduction compared to iterative reconstructions. While reports have demonstrated improvements in image quality, their impact on liver metastases detection remains unclear. Our main objective was to determine whether DLIR affects the number of detected liver metastasis. Our secondary objective was to compare metastases conspicuity between the two reconstruction methods.

Methods: CT images of 121 patients with liver metastases were reconstructed using a 50% adaptive statistical iterative reconstruction (50%-ASiR-V), and three levels of DLIR (DLIR-low, DLIR-medium, and DLIR-high). For each reconstruction, two double-blinded radiologists counted up to a maximum of ten metastases. Visibility and contour definitions were also assessed. Comparisons between methods for continuous parameters were performed using mixed models.

Results: A higher number of metastases was detected by one reader with DLIR-high: 7 (2-10) (median (Q₁-Q₃); total 733) versus 5 (2-10), respectively for DLIR-medium, DLIR-low, and ASiR-V (p < 0.001). Ten patents were detected with more metastases with DLIR-high simultaneously by both readers and a third reader for confirmation. Metastases visibility and contour definition were better with DLIR than ASiR-V.

Conclusion: DLIR-high enhanced the detection and visibility of liver metastases compared to ASiR-V, and also increased the number of liver metastases detected.

Critical relevance statement: Deep learning-based reconstruction at high strength allowed an increase in liver metastases detection compared to hybrid iterative reconstruction and can be used in clinical oncology imaging to help overcome the limitations of CT.

Key points: Detection of liver metastases is crucial but limited with standard CT reconstructions. More liver metastases were detected with deep-learning CT reconstruction compared to iterative reconstruction. Deep learning reconstructions are suitable for hepatic metastases staging and follow-up.

Keywords: Artificial intelligence; Computed tomography; Deep learning; Image reconstruction; Liver neoplasm.

Fig. 1

Patient flow-chart. (1) Loss of raw data, at least one reconstruction not saved on PACS

Fig. 2

A contrast-enhanced CT image obtained with ASiR-V (a) and DLIR-H (b) showing the same hypoattenuating metastasis, magnified in the right lower corner (white arrows). Both readers detected the lesion on DLIR-H and missed the diagnosis on ASiR-V, as did a third independent reader. CT image of the same patient two months later showing the growth of the lesion and confirming its malignancy (white arrow) (c)

Fig. 3

A contrast-enhanced CT image obtained with ASiR-V (a) and DLIR-H (b) showing the same hypoattenuating metastasis, magnified in the right lower corner (white arrows). Both readers detected the lesion on DLIR-H and missed the diagnosis on ASiR-V, as did a third independent reader. The artifact reduction provided by DLIR can be seen in this example with osteosynthesis material artifact near the lesion significantly reduced. CT image of the same patient 18 months earlier before systemic treatment confirms lesion malignancy (c)

Fig. 4

A contrast-enhanced CT image obtained with ASiR-V (a) and DLIR-H (b) showing the same hypoattenuating metastasis of 11 mm, magnified in the right lower corner (white arrows). Both readers missed the lesion on DLIR-H but not on ASiR-V. MRI of the same patient six weeks later in portal venous phase T1-weigthed image showing lesion growth and confirming its malignancy (white arrow) (c)

Fig. 5

a Subjective evaluation of CT image quality and noise. Statistically significant differences were obtained for all pairwise comparisons. b Subjective evaluation of hepatic metastases. All statistically significant differences of pairwise comparisons are displayed along with their p-values. Values are given as mean score (bars) ± standard deviation (error bars) of a five-point rating scale