Background: Hepatocellular carcinoma (HCC) is the fourth most common cause of cancer-related death worldwide. Early detection of HCC patients is related to favorable survival. Nodules in liver cirrhosis (LC) are at higher risk of developing into HCC. There is an urgent clinical need for development of an accurate and affordable non-invasive method for early HCC diagnosis among LC patients with nodules. Methods: We have previously shown that the genome-wide HIFI (5-Hydroxymethylcytosine/motIf/ Fragmentation/nucleosome footprInt) method held a solid diagnostic value in differentiating HCC from LC. In this study, we leverage this knowledge to diagnose early HCC (BCLC 0/A) in cirrhotic nodules using copy number variation (CNV) to replace 5-Hydroxymethylcytosine to simplify NGS protocol and decrease cost. The updated method generates a score that reflects the presence of tumor-derived cfDNA in 10 ml blood via low coverage (2x) whole-genome sequencing (WGS). We applied it in a retrospective cohort (validation set 1, n = 171) and a prospective cohort (validation set 2, n = 156), both of which involved patients with newly diagnosed early stage HCC (BCLC 0/A) as well as individuals with cirrhotic nodules. Results: The updated method showed excellent performance for early HCC detection both in the validation set 1 (84 HCC and 87 cirrhotic nodules; AUC: 0.951, 82.1% sensitivity at 90.8% specificity) and validation set 2 (71 HCC and 85 cirrhotic nodules; AUC: 0.958, 81.7% sensitivity at 91.8% specificity) (Table). The AUC values for distinguishing early HCC from cirrhotic nodules could reach 0.951 (80.9% sensitivity at 91.0% specificity) and 0.947 (74.7% sensitivity at 92.1% specificity) among patients with AFP < 400 μg/L and those with nodular size < = 2 cm, respectively. More importantly, our model also maintained consistent performance in detecting very early stage HCC (BCLC 0) with AUC of 0.941 (74.7% sensitivity at 91.3% specificity). Conclusions: These findings provide an accurate, affordable, excellent clinical potential model integrating four cfDNA molecular signatures for detecting early stage HCC from cirrhotic nodules using low-coverage WGS of plasma cfDNA samples.