CZ Biohub 联合多家机构投入 18 亿美元建设开放数据集训练虚拟细胞模型
精选理由
CZ Biohub 带头砸 18 亿美元做开放数据集来训练虚拟细胞模型，作者还顺手喊话医学 AI 也需要同等级的临床数据投入。
由 CZ Biohub 牵头的多家合作机构将投入 18 亿美元，用于生成开放数据集以训练虚拟细胞模型。文章指出数据是生物 AI 的主要瓶颈，这一举措旨在直接解决该问题。作者同时强调，要攻克疾病仍需直接来自人体的临床数据，目前这一块仍然缺失，并呼吁在医学 AI 领域发起类似的投入计划。
原文 · Tanishq Abraham (论文推介)
Data is the main bottleneck for AI in bio
A bunch of partners, led by @biohub, are putting in $1.8 billion to generate open datasets to train virtual cell models!
That said, in order to tackle disease, we need to understand what's happening in humans directly and we need more CLINICAL DATA as well! This is something that is still missing. We need a similar effort for AI in medicine!!
Nevertheless, this is a very exciting initiative!
