Using a single abdominal computed tomography image to differentiate five contrast-enhancement phases: A machine-learning algorithm for radiomics-based precision medicine

• Published in: European journal of radiology Vol. 125; p. 108850
• Main Authors: Dercle, Laurent, Ma, Jingchen, Xie, Chuanmiao, Chen, Ai-ping, Wang, Deling, Luk, Lyndon, Revel-Mouroz, Paul, Otal, Philippe, Peron, Jean-Marie, Rousseau, Hervé, Lu, Lin, Schwartz, Lawrence H., Mokrane, Fatima-Zohra, Zhao, Binsheng
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.04.2020
• Subjects: Aged; Algorithms; Carcinoma, Hepatocellular - diagnostic imaging; Carcinoma, Hepatocellular - pathology; Cohort Studies; Contrast Media; Diagnosis, Differential; Female; Humans; Image Interpretation, Computer-Assisted - methods; Liver - diagnostic imaging; Liver - pathology; Liver neoplasms; Liver Neoplasms - diagnostic imaging; Liver Neoplasms - pathology; Machine Learning; Male; Middle Aged; Precision Medicine - methods; Quality control; Radiographic Image Enhancement - methods; Radiomics; Retrospective Studies; Tomography, X-Ray Computed - methods; CT; CE; QC; Machine learning; Quality control; HCC; NCP; PVP; Contrast media; Liver neoplasms; AP; Radiomics
• ISSN: 0720-048X; 1872-7727; 1872-7727
• DOI: 10.1016/j.ejrad.2020.108850

[Display omitted] •A single abdominal CT image can differentiate five contrast-enhancement phases.•Performance was validated over a decade in multiple institutions across vendors.•The tool was used for radiomics-based precision medicine in liver neoplasms.•The portal venous phase was optimal in half of patients with liver neoplasm.•Contrast-enhancement was suboptimal in cirrhotic patients. The clinical adoption of quantitative imaging biomarkers (radiomics) has established the need for high quality contrast-enhancement in medical images. We aimed to develop a machine-learning algorithm for Quality Control of Contrast-Enhancement on CT-scan (CECT-QC). Multicenter data from four independent cohorts [A, B, C, D] of patients with measurable liver lesions were analyzed retrospectively (patients:time-points; 503:3397): [A] dynamic CTs from primary liver cancer (60:2359); [B] triphasic CTs from primary liver cancer (31:93); [C] triphasic CTs from hepatocellular carcinoma (121:363); [D] portal venous phase CTs of liver metastasis from colorectal cancer (291:582). Patients from cohort A were randomized to training-set (48:1884) and test-set (12:475). A random forest classifier was trained and tested to identify five contrast-enhancement phases. The input was the mean intensity of the abdominal aorta and the portal vein measured on a single abdominal CT scan image at a single time-point. The output to be predicted was: non-contrast [NCP], early-arterial [E-AP], optimal-arterial [O-AP], optimal-portal [O-PVP], and late-portal [L-PVP]. Clinical utility was assessed in cohorts B, C, and D. The CECT-QC algorithm showed performances of 98 %, 90 %, and 84 % for predicting NCP, O-AP, and O-PVP, respectively. O-PVP was reached in half of patients and was associated with a peak in liver malignancy density. Contrast-enhancement quality significantly influenced radiomics features deciphering the phenotype of liver neoplasms. A single CT-image can be used to differentiate five contrast-enhancement phases for radiomics-based precision medicine in the most common liver neoplasms occurring in patients with or without liver cirrhosis.