Application of artificial intelligence TPS assisted film reading system combined with fluo-rescence in situ hybridization in diagnosis of urothelial carcinoma
Objective To investigate the application value of artificial intelligence(AI)urine cy-tology Paris Reporting System(TPS)assisted film reading system combined with fluorescence in situ hybrid-ization(FISH)in the diagnosis of urothelial carcinoma.Methods A total of 128 patients with suspected uro-thelial carcinoma were admitted to The First Hospital of China Medical University from March 2021 to March 2024 were retrospectively selected as the study subjects.The results of the urine cytology examination were in-terpreted using an artificial intelligence TPS assisted film reading system,and FISH examination was per-formed,with histopathological examination results serving as the gold standard.The aim of the study was to evaluate the diagnostic efficacy of the artificial intelligence TPS-assisted film reading system combined with FISH examination in detecting urothelial carcinoma.Results The histopathological examination showed the following results:negative 11.72%(15/128),atypical urothelial hyperplasia 7.03%(9/128),low-grade papil-lary urothelial tumor 27.34%(35/128),high-grade papillary urothelial carcinoma 53.12%(68/128),and meta-static carcinoma 0.78%(1/128).TPS diagnosis results indicated that TPS grade 1 accounted for 1.56%(2/128),TPS grade 2 accounted for 14.84%(19/128),TPS grade 3 accounted for 12.50%(16/128),and TPS grade 4 accounted for 68.75%(88/128),TPS Grade 5 accounted for 2.34%(3/128),TPS grade 6 accounted for 2.34%(3/128).The results of FISH diagnosis showed abnormalities in CEP3 in 61.72%(79/128),CSP7 in 62.50%(80/128),CEP17 34.38%(44/128),and GLP-pl6 in 19.53%(25/128).The accuracy of single and com-bined TPS and FISH in diagnosing urothelial carcinoma was 89.84%,88.28%,and 96.88%,respectively,and the combined diagnosis was superior to the single diagnosis(c2=4.870,P<0.05).Conclusion An artificial intel-ligence TPS-assisted film reading system,combined with FISH examination,can significantly improve the diag-nostic accuracy of patients with urothelial carcinoma.This innovative approach is deserving of clinical application.
NETL
NSTL
万方数据
