基于人工智能-磁共振成像几何模型在经会阴前列腺靶向穿刺中的应用价值

Application value of artificial intelligence-magnetic resonance imaging geometric model in transperineal targeted prostate biopsy

目的探讨基于人工智能(artificial intelligence,AI)-磁共振成像(magnetic resonance imaging,MRI)几何模型在经会阴前列腺靶向穿刺中的应用价值。方法选取2022年6月至2023年6月芜湖市第二人民医院收治的204例符合穿刺指征的经会阴前列腺穿刺活检术患者作为研究对象,按照随机数字表法分为对照组与试验组,每组102例。对照组由影像科2名工作年限≥10年的副主任医师进行人工MRI阅片,行常规认知融合下经会阴前列腺穿刺活检术,试验组采用AI行MRI阅片,行AI-MRI几何模型下经会阴前列腺靶向穿刺。比较两组年龄、前列腺体积、总前列腺特异性抗原(total prostate specific antigen,tPSA)、穿刺针数、前列腺影像报告和数据系统(prostate imaging reporting and data system,PI-RADS)≥3分结节识别度、前列腺癌(prostate cancer,PCa)阳性检出率、靶点穿刺PCa阳性率、并发症发生率。结果两组年龄、前列腺体积、tPSA、穿刺针数比较,差异无统计学意义(P>0.05)。试验组PI-RADS≥3分MRI结节数高于对照组,差异有统计学意义(P<0.05)。两组PCa阳性检出率、靶点穿刺PCa阳性率比较,差异无统计学意义(P>0.05)。两组并发症总发生率比较,差异无统计学意义(P>0.05)。结论基于AI-MRI几何模型下的经会阴前列腺穿刺活检术较经会阴认知融合前列腺穿刺活检术可进一步提高穿刺准确度,且安全有效。AI阅片识别PCa结节优于人工阅片。

Objective To investigate the application value of artificial intelligence(AI)-magnetic resonance imaging(MRI)geometric model in transperineal targeted prostate biopsy.Methods A total of 204 patients who underwent transperineal prostate biopsy in the Second People's Hospital of Wuhu from June 2022 to June 2023 were selected as the research objects.According to the random number table method,they were divided into control group and experimental group,with 102 cases in each group.In the control group,two associate chief physicians in the imaging department with more than 10 years of working experience performed manual MRI reading and transperineal prostate biopsy under conventional cognitive fusion.In the experimental group,AI was used for MRI reading and transperineal targeted prostate biopsy under AI-MRI geometric model.The two groups were compared in terms of age,prostate volume,total prostate specific antigen(tPSA),number of biopsy cores,the recognition rate of prostate imaging reporting and data system(PI-RADS)≥3 nodules,the positive detection rate of prostate cancer(PCa),the positive rate of target biopsy PCa,and the incidence of complications.Results There were no significant differences in age,prostate volume,tPSA and number of puncture needles between the two groups(P>0.05).The number of MRI nodules with PI-RADS≥3 in the experimental group was higher than that in the control group,and the difference was statistically significant(P<0.05).There were no significant differences in the positive detection rate of PCa and the positive rate of target puncture PCa between the two groups(P>0.05).There was no significant difference in the total incidence of complications between the two groups(P>0.05).Conclusions Compared with transperineal cognitive fusion prostate biopsy,transperineal cognitive fusion prostate biopsy based on AI-MRI geometric model can further improve the accuracy of biopsy.Meannhile,it is safe and effective.AI imaging is superior to manual imaging in identifying PCa nodules.