Colorectal cancer stands as a leading cause of cancer-related morbidity and mortality globally,with liver metastases being a significant determinant of patient prognosis.Conventional diagnostic methods,includ-ing imaging studies and biomarker testing,frequently exhibit inadequate sensitivity and specificity,underscoring the necessity for more advanced technologies.Recent advancements in genomics,transcriptomics,proteomics,me-tabolomics,and epigenomics have revolutionized our understanding of the biological mechanisms driving colorectal cancer.These methodologies enable comprehensive analyses of genetic mutations,gene expression profiles,protein modifications,and metabolic reprogramming,all of which are pivotal to the metastatic process.This article high-lights the advanced capabilities of artificial intelligence(AI)technologies in processing complex multi-omics data,thereby enhancing diagnostic accuracy and supporting personalized treatment strategies.It also addresses the challenges AI encounters in multi-omics analyses,such as ensuring data quality,improving model interpretability,and facilitating clinical translation.Additionally,it explores the potential integration of emerging technologies like single-cell sequencing and spatial omics into large-scale,multicenter studies to further enhance the clinical utility of these tools.

Colorectal cancer stands as a leading cause of cancer-related morbidity and mortality globally,with liver metastases being a significant determinant of patient prognosis.Conventional diagnostic methods,includ-ing imaging studies and biomarker testing,frequently exhibit inadequate sensitivity and specificity,underscoring the necessity for more advanced technologies.Recent advancements in genomics,transcriptomics,proteomics,me-tabolomics,and epigenomics have revolutionized our understanding of the biological mechanisms driving colorectal cancer.These methodologies enable comprehensive analyses of genetic mutations,gene expression profiles,protein modifications,and metabolic reprogramming,all of which are pivotal to the metastatic process.This article high-lights the advanced capabilities of artificial intelligence(AI)technologies in processing complex multi-omics data,thereby enhancing diagnostic accuracy and supporting personalized treatment strategies.It also addresses the challenges AI encounters in multi-omics analyses,such as ensuring data quality,improving model interpretability,and facilitating clinical translation.Additionally,it explores the potential integration of emerging technologies like single-cell sequencing and spatial omics into large-scale,multicenter studies to further enhance the clinical utility of these tools.

Objective To discuss the correlation of apolipoprotein E (ApoE) gene polymorphism with cerebral infarction (CI)and intracerebral hemorrhage (ICH) among Tibetan nationality with cerebrovascular diseases in Qinghai Province, seek the differences in each allele of ApoE in Tibetan nationality. Methods The data from a total of 94 patients with cerebrovascular diseases was collected from the people's hospital of qinghai province, the people's hospital of guoluo prefecture , and the people's hospital of yushu prefecture as the cerebrovascular disease group, including 48 cases of cerebral infarction. There were 46 cases of cerebral hemorrhage. A total of 96 healthy Tibetan subjects were selected as the control group. DNA was extracted from all subjects. Real－time PCR was used to detect ApoE. The correlation between ApoE genotype and cerebral infarction and cerebral hemorrhage was analyzed. Results E2/E3 gene was common in Tibetan nationality with cerebrovascular diseases. E2/E3 genotype accounted for 50% in cerebral infarction group. E2/E3 (65.2%) was the most common in intracerebral hemorrhage group. E2/E4 (64.6%) was the most common in the control group. There was a statistically significant difference between the two groups (P＜0.01). In the Tibetan population, ε3 allele genome (48.0%) was the most common in cerebral infarction group and ε2(43.5%) were the most common alleles in intracerebral hemorrhage group. In the normal control group, ε4 (49.0%) was the most common allele. Conclusion E2/E3 genotype may be related to cerebrovascular diseases. ε3 allele may be the susceptible factor of cerebral infarction wherase ε4 may be the protective factor of cerebrovascular diseases in Tibetan population.

Objective To study the imaging characteristics of intracranial dysembryoplastic neuroepithelial tumors(DNT),and to evaluate the role of CT and MRI in their diagnosis.Methods We presented 8 DNTs cases confirmed histopathologically after operation.CT and MRI features were retrospectively analyzed(3 cases received MR scan only and 5 had both studies),compared with the surgical findings and pathological appearance.Results Eight patients had totally 11 tumors with multifoci in 2 cases.Six DNTs were located in the temporal lobe,1 lesion involved the parietal lobe,1 in the frontal lobe,1 in the brain stem,1 in the cerebellum,and 1 in the frontal and temporal lobe simultaneously.All lesions were observed intracortically,and partial of them extended into the subcortical location.On 5 non-contrast CT scans,3 DNTs were homogeneously hypodense;1 case contained isodense nodules within the hypodense focus.Peripheral spotty calcification was found in 1 case.On pre-contrast MR images,all lesions showed hypointense on T_1-weighted images,moreover,multicystic change with more markedly decreased signal intensity was observed.The DNT had a well-demarcated,multinodular gyriform configuration,or a soap bubble appearance at the cortical margin.On T_2-weighted images,high signal intensity of the tumors was seen except for hypointense calcification.The bone remodeling of the adjacent calvaria was noticed in(1 case.) The tumors had slightly increased signal intensity on FLAIR(fluid attenuated inversion recovery) in 2 cases,and homogeneously mild hyperintense on PDW(proton density weighted imaging),while the isointense focus was observed in one case on DWI(diffusion weighted imaging).No obvious enhancement was identified except 1 lesion with mild enhancement following intravenous administration of contrast material.3 tumors had ill-defined contours,whereas the others showed well-demarcated margins.No surrounding vasogenic edema was present except 2 foci with slight edema.Conclusion DNT usually shows characteristic radiologic findings,in combination with the clinical presentation and history,the diagnostic accuracy should be improved,and,as such,unnecessary radiation and/or chemotherapy may be avoided.