脑梗死是全球排名第二大致死原因，在中国已成为第一大致残及死亡原因。急性缺血性脑梗死（acute ischemic stroke，AIS）则是最常见的脑梗死类型，约占全部脑梗死的80%。出血性转化（hemorrhagic transformation，HT）是AIS患者的自然转归过程之一，也是静脉溶栓治疗（intravenous thrombolysis，IVT）或血管内取栓治疗等治疗后最严重的并发症，给患者及其家庭和社会都带来沉重的负担。精准预测、评估AIS的HT具有重要临床意义。近年影像学研究集中于CT和MRI评估HT的价值。CT平扫诊断HT价值有限，AIS治疗前，阿尔伯塔脑梗死计划早期诊断评分≤7分与HT发生相关（P=0.033）,高密度大脑中动脉征是发生HT的独立危险因素（OR=10.334）；AIS取栓治疗后24 h复查双能CT，预测2~7 d内发生HT风险的效能较高（灵敏度82.5%、特异度100%）。CT血管造影中，血栓负荷高的患者HT发生概率上升（OR=1.28）；侧支循环良好的AIS患者HT发生率低。CT灌注成像参数包括表面渗透性、脑血容量、Tmax等,均有较好的HT预测价值。MRI平扫中，FLAIR上高信号可预测HT的发生；MRI弥散加权成像,高信号区域体积可预测HT（ROC曲线下面积0.78）；MRI磁敏感率加权成像，刷状征、微出血灶提示HT的发生；MRI增强T1加权成像,脑实质强化与HT发生显著相关（P<0.05）；MRI灌注加权成像中，梗死区脑血容量降低可预测HT。此外，CT及MRI图像后处理系统RAPID提升了评估HT的效能。建议根据各急救中心的硬件及当地医疗情况，设置个性化的影像学检测方式及流程，预测及管理HT。

Cerebral infarction is the second leading cause of death in the world, and has become the most serious cause of disability and death in China. Acute ischemic stroke (AIS) is the most common type of cerebral infarction, accounting for about 80% of all cerebral infarctions. Hemorrhagic transformation (HT) is one of the natural regression processes in patients with AIS, and is the most serious complication after treatments [such as intravenous thrombolysis (IVT) or endovascular thrombectomy (EVT)], which brings a heavy burden for patients and their families and even the all society. Accurate prediction and evaluation are of important clinical significance. In recent years, imaging research has focused on the value of CT and MRI in evaluating HT. The diagnostic value of CT plain scan is limited. Before AIS treatment, an early diagnostic score ≤ 7 points in the Alberta Cerebral Infarction Plan is associated with the occurrence of HT (P=0.033), and high-density middle cerebral artery sign is an independent risk factor for the occurrence of HT (OR=10.334). For AIS patients treated with thrombectomy within 2-7 days, dual energy CT scanning at 24 hours after therapy had a high efficacy for prediction of HT occurrence, with a sensitivity of 82.5% and specificity of 100%. CT angiography suggests that patients with high thrombus burden had a higher probability of developing HT (OR=1.28). The incidence of HT in AIS patients with good collateral circulation is low. CT perfusion imaging parameters, including surface permeability, cerebral blood volume, Tmax, etc., have good predictive value for predicting HT occurence. In MRI plain scan, high signal on FLAIR can predict the occurrence of HT. The volume of high signal areas on MRI diffusion-weighted imaging can predict HT, with area under the ROC curve of 0.78.Brush like sign, and micro bleeding lesions on sensitivity weighted MRI indicate the occurrence of HT. Enhanced T1 weighted imaging of MRI shows a significant correlation between brain parenchymal enhancement and HT occurrence (P<0.05). Perfusion weighted imaging of MRI shows a decrease in cerebral blood flow (CBV) in the infarcted area may predict HT occurence. In addition, the CT and MRI image post-processing system RAPID has improved the evaluation efficiency for HT occurrence. In the future, personalized imaging detection methods and processes should set up based on the hardware and local medical conditions of each emergency center for HT management.