Objective: We retrospectively evaluated the levels of pre-diagnostic blood indicators in patients with HCC in order to develop a non-invasive predictive model that can precisely anticipate the onset of HCC and possibly enhance early clinical identification and prognostic assessment. Methodology: Between June 2021 and December 2022, total of150 HCC patients were admitted toHospital, 70 of whom had complete prognostic data. Twenty five factors that could indicate whether or not HCC would develop early were retrieved. Propensity score matching, ROC curve, logistic regression, and decision curve analyses were performed using R (version 3.6.1) software. All the tests were two sided and p value less than 0.05 was considered as for statistical significance. Results: The scoring model included two common patient characteristics (age and gender) as well as five independent predictor variables for the start of HCC. With an area under the curve (AUC) of 0.890 (95% CI 0.856-0.925). When compared to single variables or other score systems, the score model had greater predictive performance in discriminating and clinical net benefit, according to ROC analysis. The score model, however, exhibited strong predictive values for patients with early tumour stages (AJCC stage I) or small tumours ( 2 cm), according to stratified study. Additionally, the HCC patient's score started to rise 30 months before the clinical diagnosis and peaked at 6 months. Conclusion: By adopting this model, we may be able to make early adjustments to the current risk categorization system and take into account more extensive surveillance programmes for high-risk individuals. Additionally, it can assist doctors in evaluating the prognosis and development of HCC patients.