Artificial Intelligence‐driven Platform: Unveiling Critical Hepatic Molecular Alterations in Hepatocellular Carcinoma Development

Abstract Since most Hepatocellular Carcinoma (HCC) typically arises as a consequence of long‐term liver damage, the hepatic molecular characteristics are closely related to the occurrence of HCC. Gaining comprehensive information about the location, morphology, and hepatic molecular alterations related to HCC is essential for accurate diagnosis. However, there is a dearth of technological advancements capable of concurrently providing precise HCC diagnosis and discerning the accompanying hepatic molecular alterations. In this study, We have developed an integrated information system for the pathological‐level diagnosis of HCC and the revelation of critical molecular alterations in the liver. This system utilizes computed tomography/Surface‐enhanced Raman scattering combined with an artificial intelligence strategy to establish connections between the occurrence of HCC and alterations in hepatic biomolecules. Employing artificial intelligence techniques, the SERS spectra from both healthy and HCC groups were successfully classified into two distinct categories with a remarkable accuracy rate of 91.38%. Based on molecular profiling, we have identified that the nucleotide‐to‐lipid signal ratio holds significant potential as a reliable indicator for the occurrence of HCC, thereby serving as a promising tool for prevention and therapeutic surveillance. This article is protected by copyright. All rights reserved