⑤Meta将斥资千万组建加州超级政治行动委员会，推动AI友好政策； ⑥郭明錤：预计Meta AI眼镜Hypernova今年三季度量产，售价约800美元； ②字节跳动提出Jeddak AgentArmor智能体安全框架，为AI Agent行为立"规矩"； ③华为推出三款AI SSD新品，其中大容量系列的HUAWEI OceanDisk LC 560，单盘容量将最高可达 122/245 TB，实现业界最大单盘容量； ④谷歌正式发布图像生成模型Gemini 2.5 Flash Image； AI进化速递 | 谷歌发布图像生成模型Gemini 2.5 Flash Image ①浙大一院、阿里巴巴发布"平扫CT+AI"主动脉急诊模型iAorta； ⑦Anthropic宣布推出基于Claude模型的浏览器智能体研究预览版——Claude for Chrome。 ...

人民财讯8月27日电，阿里巴巴公众号消息，今天，浙江大学医学院附属第一医院、阿里巴巴达摩院发 布用于胸痛急诊场景的AI模型iAorta，可用常规平扫CT在几秒内识别急性主动脉综合征，将确诊时间缩 短至2小时内。 iAorta从1万多名胸痛患者中精准发现21例，帮助他们得到及时救治，目前已在浙江首批10家医院部 署，即将全国推广。 ...

Core Viewpoint - Acute Aortic Syndrome (AAS) remains a severe cardiovascular emergency with a high mortality rate, necessitating rapid and accurate diagnosis to improve patient outcomes [3][4][7] Group 1: AAS Overview - AAS has a mortality rate of approximately 40%-50% within 48 hours of onset, with an hourly increase of 1%-2% if untreated [3] - Clinical symptoms of AAS are often non-specific and variable, complicating timely diagnosis [3][4] - Traditional imaging methods like CTA are costly and have associated risks, while the actual incidence of AAS in patients undergoing CTA is only 2.7% [3][4] Group 2: AI-Based Diagnosis - Alibaba's Damo Academy developed an AI-based warning system called iAorta, which utilizes non-contrast CT scans to identify AAS with high accuracy [5][8] - iAorta demonstrated an area under the curve (AUC) of 0.958 in a multi-center retrospective study and showed sensitivity of 0.913-0.942 and specificity of 0.991-0.993 in large-scale real-world studies [10][20] - The AI system significantly reduced the time for correct diagnosis from an average of 219.7 minutes to 61.6 minutes [10] Group 3: Real-World Applications - iAorta has been tested in various hospitals, successfully identifying AAS in patients who were misdiagnosed or refused further testing [15][19] - In a pilot at Shanghai Chang Hai Hospital, iAorta accurately identified 21 out of 22 AAS patients, with an average diagnosis time of 102.1 minutes [19][20] - The AI system enhances diagnostic accuracy for doctors of all experience levels, with significant improvements noted in interns and general practitioners [20][23] Group 4: Conclusion - The integration of AI in diagnosing AAS represents a significant advancement in emergency medicine, potentially saving lives by preventing diagnostic delays and errors [24][23]