借助AI手术超能力普及复杂手术——安德洛墨达外科

大家好，欢迎回到医疗创新播客。我是布兰诺夫，今天和我的搭档阿尔文德一起。我们要采访一位非常酷的人物——安德罗梅达外科公司的CEO兼创始人尼克·达米亚诺。

Hey, everyone. Welcome back to the medical innovation podcast. My name is Branov. I'm here with my cohost, Arvind. And today, we're speaking to someone really cool, Nick Damiano, the CEO and founder of Andromeda Surgical.

安德罗梅达正在研发自主手术机器人，我们迫不及待想了解更多。尼克，能请您做个自我介绍吗？

Andromeda is building autonomous surgical robots, we're so excited to learn more. Nick, would you mind introducing yourself?

好的，感谢邀请。很荣幸来到这里。我是安德罗梅达外科的联合创始人兼CEO，我们致力于开发自主手术机器人。

Yeah. Thanks for having me. Great to be here. So, yeah, I'm cofounder and CEO of Andromeda Surgical. We're developing autonomous surgical robots.

公司成立大约两年，2023年起步，现在已接近成为上市最快的手术机器人。我们已提交510k申请，并获得了良好的临床数据。在此之前，我是斯坦福毕业的工程师。原本想当医生，后来发现住院医师培训和医学院的流程太繁琐，于是决定通过构建产品来规模化地帮助病患。

We're about two years old as a company. We started in 2023 and are pretty close now to being the fastest surgical robot to market. We've submitted our five ten k, gotten good clinical data already. And before that, I started as an engineer out of Stanford. I was gonna be a doctor and then decided it was, too much trouble to go through residency and med school and all of that and decided I could basically build things to help patients at scale right now instead of going through all that.

于是转型成为工程师，在创业前曾任职于几家医疗科技初创公司。在创立安德罗梅达之前，还创办过两家医疗领域的公司。

So became an engineer and then worked for a few med tech startups before being a founder and had two companies before this, both in the the medical space before starting Andromeda.

听到这些背景真是令人振奋。尼克，或许可以多聊聊你在医疗技术领域的经历和成果。我和普拉纳夫现在就读医学院，走完这条路后，发现其实在这个领域通过技术构建解决方案也是条充满可能性的路径——毕竟有太多亟待解决的精彩问题。很想听听你本科时的经历，考虑医学院又转向的心路历程，以及当时解决了哪些具体问题。

It's it's really exciting to hear that background. I think it it could be worth talking a little bit more about your background, Nick, in terms of the health care technology that you've been part of and things that you've built. Pranav and I both have some thoughts about, you know, now that we're in medical school and have gone through this whole path, have an alternative path of, like, building in this space because of all the, really exciting, you know, problems that that there are to solve. So we'd love to hear a little bit about what it was like, you know, going through undergrad, thinking about, you know, potentially going to medical school, but then pivoting, realizing that you could solve these other issues and what what problems you you solve were able to solve at that at that time.

确实有几个关键因素促成了这个决定。一方面在斯坦福和硅谷的经历让我深入科技圈，参加了学校的创业领袖研讨会，那些创始人白手起家的故事让我热血沸腾——我意识到这本质上就是我的天性。从小到大我一直在折腾各种小项目，小时候还组织过体育联赛。

Yeah. I think it was a couple of things that really led to that decision in the end. One was mean, being at Stanford, being in Silicon Valley, I really got into the the tech scene and started going to this entrepreneurial thought leader seminar that Stanford had, and it was, just really exciting stories from founders that had built things and realized that was kind of who I was by nature. I'd always started little little companies or little organizations. As I had gone, I had some sports leagues when I was a a kid.

我在大学时创办了一个小T恤生意，制作仿制的斯坦福运动衫和一些搞笑T恤卖给同学。我对此非常投入。同时，我记得大概是大二前后的那个暑假——确切说应该是在大二结束后——我在一个医学实验室里，亲眼目睹了住院医师们的工作状态。那是个等级森严且充满限制的职业环境，至少作为住院医师时，他们必须完全服从上级指令。

I had started a little T shirt business in in college that was making fake, Stanford athletic shirts and, like, funny funny shirts that I would sell to people. And so, really was drawn into that. And then also at the same time, I think the summer before, or before or after sophomore year. I think it was after sophomore year that I was in a medical lab and just saw what it was like for residents in the lab to to do what they did. And it was a really kind of hierarchical and limiting career where I I felt at least when you're a resident, it was they really had to do what?

就是日复一日地执行主治医师的要求。于是我认定自己不适合这种高压的工作模式。我骨子里就是个叛逆者，觉得这种生活不适合我。再加上想到要熬这么多年才能熬出头，再看看当时许多科技公司的创业故事——比如刚成立不久的Facebook，创始人辍学搬到帕洛阿尔托就直接开干了。

Take take orders and just kind of work, you know, day and night to just, like, do what the attendings wanted. And so I basically decided I didn't wanna live in that kind of hard work of a world. I was always a bit of an iconoclast, and I just didn't think that was that was for me. And then just thinking how many years it was to get there and then just seeing how a lot of these companies were built. Like, Facebook had just started, and had dropped out of school and moved to Palo Alto and just, like, done it.

这种'想到就能做'的思维方式深深吸引了我。我决定要开发能同时造福数百、数千甚至数百万人的产品，而不是一次只能帮助一个病人。那一刻我清楚地意识到这才是我该走的路。

The kind of the you can just build things or you can just do things mentality was basically at the the heart of that. And I just decided I could go and build products now that could help not just one patient at a time, but hundreds, thousands, millions even. And that was pretty clear at that point that was the way I should go.

感谢分享，这些经历非常精彩。我和Arvind都有同感——通过医疗创新来开发产品工具和服务，确实能比传统医疗方式影响更多人群，这正是我们对您的事业如此感兴趣的原因。如果没记错的话，这是您创立的第三家公司？

Thank you for sharing that. It's really great to hear. I think Arvind and I are both on similar pages. We have both realized that you can make so much of a larger impact and help so many more people by being on the innovation side of health care, by building products and tools and services that are able to help not just one patient at a time, but hundreds or thousands at a time, which is why we're so excited about what you're doing. If I recall correctly, this is your third company.

您最初创立了Neurof，然后是Flow，现在正在运作Andromeda Surgical。我们很想了解，是什么促使您从工程师角色转型，踏出创业第一步的？

You started off with Neurof, then then Flow, and are currently working on Andromeda Surgical. We'd love to hear more about what prompted you to initially jump from your initial engineering roles to starting your own company, your first company?

说来话长。其实最初我总觉得医疗科技门槛太高，刚毕业就当创始人不太现实——现在回想这是个错误认知。当时我先后经历两家创业公司失败，反而促使我决定自己创业。

Yeah. I mean, when I it was, I guess, two startups I worked for not making it that led me into the to being a founder, maybe make the jump. At the beginning, I I think for some reason, I didn't believe. Maybe I thought that med tech was was gonna be too hard to just go out of school and be a founder. I think that was a a flawed assumption now that I look back at it.

医疗创业确实比普通科技公司更难，但并非不可为。人们常常低估自己的学习能力。真正的转折点是我供职的初创公司经历资本重组，员工从40人裁到只剩几个，我突然失业了。那时我就想：如果由我来掌舵，结果会完全不同。所以下次，我不想再做听天由命的工程师，而是要把命运掌握在自己手中。

It is harder than just building a tech company, but you can you can do it. I mean, I think people underestimate their ability to learn. And so the I think the big impetus was when I was working for a start up, and it got recapped and, you know, basically went down from, I don't know, like, 40 employees to, like, a a few, and I was suddenly out on the the street and had to figure out what was next and, just basically thought that if I had been leading the company, I would have done things differently. And so next time, I don't wanna be an engineer at the whim of what does the leadership team do. I wanted to take matters into my own hands.

正是这一点最终促使我走出去，寻求创立一家公司。我接触了许多人，思考了各种想法，最终创建了后来被称为Avail的新项目。确实，在这个过程中我们犯了很多错误。我和我的联合创始人们都是第一次创业，对于创办公司意味着什么几乎没有参考依据。

And that's what led me to eventually just go out and and seek to found a company. And I went and talked to lots of people and thought about different ideas and ended up doing what became a new rep and then eventually was called Avail later on. And, yeah, I mean, definitely made a lot of mistakes coming into that one. I mean, I think my myself and my cofounders both were were all first time founders. It didn't really have much reference point for what it's like to found a company.

虽然公司最终发展得很好，但我们不得不通过艰难的方式吸取每一个教训。ZenFlow是我在完全不同领域创立的第二家公司。Avel（或称NewRep）提供远程手术支持，通过视频在手术室提供支持，而非需要代表或另一位外科医生亲临现场。

The company ended up doing really well in the end, though we we had to learn every lesson the hard way. And then, ZenFlow, yeah, I mean, that was a second company I founded in a very different space. It was I mean, so Avel or NewRep was remote surgical support. So you were over a video supporting cases in the Operating Room instead of having a rep or another surgeon physically there. You could just do it remotely.

而ZenFlow是一款三级植入式设备，这在某些方面要困难得多，需要克服重重障碍才能获得FDA批准、推向市场并销售给医院。我花了八年时间领导这个项目，学到了很多。现在我觉得自己终于积累了足够经验，有望创建一家能实现宏伟目标的大公司，这正是我们现在投身Andromeda的原因。作为多年医疗科技从业者，我深刻认识到设备易用性和人机交互是影响其效能的主要限制因素。

And ZenFlow was a class three implantable device. So very different and much harder in certain ways to get that through all the hurdles it takes to go to get to FDA clearance, get to market, sell to hospitals, all all of that kind of kind of stuff. And so, yeah, spent eight years leading that, learned learned a lot. And now I feel like I'm finally at the point that I I know enough to really have a good shot at building a a huge company that can do something really ambitious, which is why we went into this now. So with Andromeda, I mean, we I had basically from my from what I had seen as a med tech operator for many years, saw that usability and just human interaction with devices was the big a big limiting factor for how well they worked.

ZenFlow和其他几乎所有由人工操作的导管设备都存在这个问题。我曾考虑开发自主机器人来部署设备，但对单一产品公司而言构建整个系统并不现实。这时我的联合创始人从另一个角度切入——他曾在Starsky Robotics公司开发了首辆自动驾驶卡车，一直在思考这项技术除驾驶外还能创造高价值的应用场景，手术就是其中之一。当我们相遇时，发现彼此都在思考自主手术机器人，随即兴奋地决定：我们应该实现它。

That was certainly the case with ZenFlow as well as almost every other catheter based device that's delivered by a a human. And I thought about developing autonomous robots to deploy that, though it didn't make sense for us as a single product company to build this whole system. You'd have to have a robotic platform, and then you guys build different devices on top of that, to make that work. And then my cofounder actually had come at it from a different angle where he was he'd built the first autonomous truck with a company called Starsky Robotics, and he had been thinking about just what you could use that technology for besides driving that would add a lot of value, and then surgery is one thing he had thought of. So we met, and we had both been thinking about autonomous surgical robots and then got really excited just thinking, you know, we should build this.

我们应该共同探索这个方向，验证其可行性。所有迹象都极其乐观地表明这必将成为未来趋势。尽管多数人认为这还需要二十年，但他坚信现在就能实现——毕竟自动驾驶技术已经实现，而这在许多方面其实是比手术更复杂的难题。当时我们就意识到：如果我们不做，未来十年内也必定会有人做这件事。

We should look into building this together and see if this really could be a thing. And the the signs were all super positive from everything we saw that this was definitely gonna be the way that things went. And most people thought it was twenty years away, but we especially especially him. Like, he he knew that we could build this now, that it was it already been built for driving, which is actually a a harder problem than surgery in many ways. And so it just seemed like at that point that someone was gonna do this in the next ten years if if we didn't do it.

总会有人创办这样一家公司，而且必将成为行业巨头。我们都被这个极具挑战性的问题所吸引，最终决定必须付诸行动。当时我手头有五个其他项目，他也有其他正在进行的工作，但我们一致认定：这个方向比其他任何项目都令人振奋。

Someone was gonna start a company. It was gonna be big that was gonna do this. And we both found this to be a really exciting problem, so we just said we we have to do this. I had five other projects, and he had other other projects too he was working on at the same time. And we both came to the definitive conclusion that this is way more exciting than any of that stuff.

我们必须立即行动。这至少在该领域是我们能打造的最卓越的公司。

We have to do this right now. This is the the best company we could possibly build in at least in this space.

是的。当谈到手术机器人时，我不禁回想起大三临床轮转时，我们用达芬奇机器人进行子宫切除术等操作。那台机器人基本就是我们的认知标准——它通过机械臂进入手术区域，医生在远端操控台用类似游戏手柄的方式控制机器人进行微创手术，手术通道只有针孔大小，医生通过独立显示屏完成操作。

Yeah. I think when you talk about surgical robots, I just think back to my third year wrote clinical rotations where we would do, like, hysterectomies and things with the da Vinci. And so that was, like, the robot that I think was what we thought. And for listeners, it's basically this robot where they would attach the arms basically, into the operating field, and then you do, like, laparoscopic surgery, which is through pinholes. And the surgeon would sit at a at a remote area to, basically use almost like video game controls to control the, the robot in the, surgery and would have their own display to do it.

这就是我对机器人潜力的最初认知。而说到‘自主’手术机器人，事情就更有趣了。我们之前讨论过，提到机器人总会联想到那些机械臂——它们会不会自己走动呢？

And and that was my first kind of exposure of what could potentially be a robot. Now when you say, like, autonomous, surgical robots, that makes things even more interesting. And I think, you know, we were talking a little bit about this before, but when we think about robots, we think about, okay. Then there's probably these arms. They're kinda like, are they walking around?

它们能独立完成手术吗？但您提到关键不在于硬件本身，而更多在于控制系统而非物理执行部分。我们很想听您详细阐述对自主机器人的定义，特别是在手术领域，以及安德洛墨达项目在这个方向上的定位。

Like, are they doing the surgery by themselves? But you mentioned it part of it or, you know, the majority of it is is more so the brains than the than necessarily the the hardware itself that's doing the physical work. We'd love to hear more about how you define autonomous, robots, particularly in surgery and from what angle you're taking Andromeda, in that in that regard.

确实。我们认为未来十年机器人领域——至少在外科机器人领域——最大价值增量将来自软件层面，整个机器人行业可能都是如此。所以我们本质上更像是软件公司，与其他手术机器人厂商截然不同。虽然我们也有实体机器人，但基本采用现成部件。

Yeah. And we we feel like that most value is gonna be added in robotics, in surgical robotics, at least, with software, probably in robotics as a whole also in the next decade. So we are we are much more of a software company than everybody else in surgical robots. We do have a physical robot. It's a it's mostly off the shelf parts.

机械臂来自库卡，是许多其他手术机器人的标准配置。但我们的核心几乎完全聚焦在软件上。关于自主性，目前有些公司自称实现了手术机器人的自主操作，但与我们理解的、或大众认知的自主性并不相同。

The arm is from KUKA. It's a standard robotic arm that's used in a lot of other surgical robots. But then yeah. I mean, so we're really focused almost entirely on the the software. And, yeah, as far as autonomous goes, so there are companies that claim to be autonomous in surgical robots now, but they're not autonomous in the sense that we mean it or that most people mean it.

人们常常混淆自动化和自主化。在我看来，自主性更强调人工智能的非确定性特征——不只是预设程序让机器人执行固定动作，而是机器人能自主决策，不局限于预定路径。我们产品的核心就是通过AI驱动手术，但依然会保持人类在回路的模式。

A lot of times, people, I guess, conflate things that are automated with things that are autonomous. Autonomous to me has more of an an AI nondeterministic connotation to it where it's not just you tell the robot to do a specific thing, and then it just does that thing without you intervening during it, that you completely define what it's gonna do, and then it does it. It's actually the robot making decisions for itself and not just in a predetermined path. So that's how we mean autonomous is that we're using AI at the core of the product, and that's gonna drive surgery. It's gonna be, though, human in the loop.

需要特别说明的是，我们并非要取代外科医生。医生仍会在控制台操作，更像是特斯拉自动驾驶模式而非Waymo全无人驾驶——可能这个类比太硅谷了。总之医生会在控制台监督，机器人并非扮演外科医生角色。

It's one thing worth mentioning that we're we're not dis intermediating the surgeon from the procedure. They're gonna be at the console. It's it's more like a Tesla self driving than Waymo, if you can imagine those two. Maybe that's too San Francisco of a of an analogy. But it's it's at they're at the console, so it's not acting as a surgeon.

这更像是赋予外科医生超能力，让他们成为可能达到的最佳状态。

It's more giving the surgeon superpowers and making them the best surgeon they could possibly be.

我喜欢这个比喻，特斯拉与Waymo的对比。不像Waymo那样连方向盘都没有，遇到前方有情况时也没有刹车踏板可以紧急制动。这更像是有人类参与的Tesla模式——外科医生仍在确保一切正常进行，运用自己的判断力，在发现比机器人或AI更好的手术路径时可以介入并改变手术方向。

I love that analogy, Tesla versus Waymo. It's not like a Waymo where there's not even a steering wheel that you can touch or brake pedal you can use to urgently stop if there's something that's coming in front of you. It's more like a Tesla where there's a human in the loop. The surgeon's still making sure everything's going okay. The surgeon's using their own judgment and can step in and alter the direction of the surgery if they see a path that's better than what the robot or the AI is doing.

是的，这很重要。研究这个领域时你会发现，如果要开发自主手术机器人，会面临很多敏感问题，这很合理。外科医生、患者、FDA或其他监管机构都会对这种新前沿技术感到不安。因此必须建立正确的保障措施，并采取适当步骤逐步实现完全的五级自主。

Yeah. It's important. Looking into this space is that if you're gonna build autonomous surgical robots, there's a lot of sensitivity around that and for good reason. So when you look at surgeons, patients, FDA, or other regulatory bodies, they're gonna be a little bit uneasy about this new frontier becoming a thing. And so you have to build the right safeguards in and take the right steps along the way to get all the way to full level five autonomy.

不能一蹴而就，我们正从基础做起——就像使用谷歌地图导航的机器人，先提供定位引导功能，告诉你当前位置、手术中所有关键点的位置。然后在此基础上逐步增加自主层级，通过发布新功能并确保其安全性。同时进行影子测试，在当前产品上测试下一代自主功能但不实际执行，只是观察算法在特定情况下的决策。通过设置正确保障措施、保持人类参与、在关键环节设置防护栏防止可能伤害患者的操作，我们才能稳步推进。

You can't just go all the way there right away, and that's what we're doing is starting with the robot that's basically like driving with Google Maps where you have mapping guidance that's kind of telling you where you are, where everything else you care about, every landmark in the surgery is. And then it's going on top of that with different with increasing layers of autonomy where we're gonna be releasing features and making sure that they're safe. Also doing this shadow testing in the background where we're testing the next generation's autonomous features on the current product, but not actually having them take action. Just looking at what the algorithms would have done in a certain case. And that's how we're gonna be able to build plus by by just having the right safeguards, having humans in the loop, having guardrails in certain places where it's not able to do certain things that might be, harmful to the patient.

所以必须采取极其审慎的策略来实现完全自主。

So just taking a really deliberate deliberate approach in terms of getting all the way to full autonomy.

没错。没人会立即批准完全自主的、无人监督的机器人独立完成所有手术。必须循序渐进地获得公众认可和信任，同时提升AI能力。换个话题谈谈外科医生的使用体验——我读到与达芬奇机器人不同，你们的外科医生没有可以控制机器人每个动作的完整控制台。

Right. No one's going to approve fully autonomous or, I guess, no surgeon or loop surgeries where the robot is just doing everything by itself without any oversight immediately. You have to do it step by step and gain public approval, gain trust, but also build up the AI's capabilities. Switching directions a little bit into the surgeon experience. I've read online that unlike da Vinci robots, the surgeon in your case does not have a full console where they can control the robot's every move.

而是使用类似平板电脑的设备进行校正、调整手术路径或做其他决策。能否详细说明外科医生使用你们机器人或软件时的具体体验？

Instead, it's more like a tablet where they can make corrections or alter the course and perhaps make other decisions. Could you go into more detail about what the surgeon experience is using one of your robots or, I guess, using your software?

是的，它本质上是一个平板控制器，现在基本上就是一台iPad。这是我们临床病例中一直在使用的设备，也是我们认为最适合该产品长期发展的控制器方案。

Yeah. So it's it's a tablet controller. It's basically an iPad now. That's what we've been using for our clinical cases. And that's what we see as the best controller for the long term way this product is going.

在目前已完成的手术案例中，它的表现也非常出色。已有六位不同的泌尿科医生在真实患者身上操作过，我们发现医生们能很快上手。毕竟这是大多数人已经习惯使用的工具——即便没用过iPad或平板，至少也用过智能手机，这大大降低了学习门槛。

It also has worked really well in the cases we've done so far. We've had six different urologists operate it in, live human patients and really seen that it's it's something people take pretty quickly to. It's a tool that, of course, most people are used to using. I mean, they at least at least use a smartphone, if not an iPad or a a tablet. And so that's really reduced the learning curve.

我们还意识到平板可以用于培训，这对所有医疗技术产品（尤其是达芬奇这类手术机器人）都是个大难题。我们可以把原本需要长时间高强度训练的内容做成iPad应用，让医生通过应用程序学习操作流程和控制系统，而不必非得在手术室里用尸体、模型或真人练习。他们可以自主安排时间在家学习，这非常便利。随着公司发展，当我们为系统增加自主操作等级后，医生就不再需要操控机器人的每个动作，而是启动不同级别的自动驾驶模式后按需调整。这种场景下，iPad平板界面就特别适合——不像达芬奇系统那样需要时刻保持高强度操控。

One thing we're also gonna do that we realize we can do with a tablet is for training, which is a big problem with, well, every med tech product and especially robots with DaVinci, for example, there's a I mean, intentionally a very long and intensive training process is we can we can put that onto an iPad app and have them learn procedures and our controls on the app and not have to be in the Operating Room on, you know, cadavers or models or patients. They can just do it on their own time at home, which is is really nice. And then as we so as we grow as a company, as we add autonomy levels to this, then it's just it's gonna be a lot less of you're doing every motion that the robot makes. It's gonna be you're basically launching different levels of autopilot, then you can adjust as needed. And for that, I mean, the the iPad tablet interface really works well that you're not gonna need to be as actively at the controls as you would with the da Vinci, for example.

没错。我认为令人兴奋的另一点是，正如你所说，作为以软件为核心的公司，通过兼容iPad和现有手术硬件系统，你们实现了高度模块化，能很好地整合进新医疗体系和其他医疗系统。这是有意为之的设计吗？

Yeah. And I think what's what's exciting about that too is is, like you said, being a predominantly software company, integrating to iPads and existing surgical, hardware as well, you become really modular and can integrate really well into into new systems and other health care systems. Is that was that a pretty intentional design?

绝对是。模块化设计非常重要，我们要确保产品不会打乱手术室原有工作流程，不需要外科医生从头学习全新系统，同时能适配不同手术类型。我们当前主攻的是名为HOLAP的前列腺剜除术，之后会快速扩展到其他术式，比如明年就会推出第二个适应症。

Definitely. I mean, the the modular aspect is really big, that we wanna something that's not gonna disrupt the workflow in the operating room. It's not gonna be something entirely new that the surgeon has to learn, and then it can be adapted to different procedures. So we're starting with this procedure called HOLAP, this prostate enucleation procedure, And then we're gonna be quickly growing into other procedures beyond that. So the second indication is gonna launch next year, for example.

今年我们推出HOLEP手术方案后，明年将发布下一个适应症。我们的目标是迅速将其发展为多功能平台——初期每年至少新增一种术式，后期实现每年多种扩展。这将是一个长期发展、高度通用的手术平台，因此必须确保基础架构从一开始就支持这种扩展性，避免后期为适配10-20种术式而对硬件进行彻底改造。

So we're launching HOLEP this year, and then, the next the next one will be next year. And then we want to expand this into a broad platform very quickly where in the beginning, we're launching at least one new procedure per year, and then it becomes multiple per year. And so this is gonna be a a long term, a really, versatile surgical platform. And we wanna make sure that that it's built from the beginning to be able to do that, that we're not gonna have to totally revamp the hardware to be able to be a be a, you know, 10 or 20 procedure platform versus a a single procedure.

是的。我想更深入探讨下HOLAP技术。如果你不介意，我想试着阐述我对该术式和技术结合的理解，当然需要你随时纠正。据我理解这是用于治疗BPH（良性前列腺增生）的病症对吧？

Yeah. And I kinda wanna delve a little more into Holub. If if you entertain me, I'm gonna try to understand or try to explain my understanding of of the procedure and how the technology loops into it, would love for you to to to correct me as you definitely will have to. But, my understanding is it's for this condition called BPH, benign prostate hyperplasia. Yep.

这种病症常见于老年男性，会导致排尿困难等问题。基本上，前列腺增大会阻碍尿流通过。这项手术采用激光技术，实质上是缩小压迫尿道、阻碍排尿的前列腺体积。而Andromeda能与现有平台集成，从而更精准地引导激光进行治疗——这对缺乏该平台或手术经验的外科医生而言原本相当困难。这样的理解对吗？还是说...

This condition that happens often in older men, and it can cause, you know, urinary, difficulty. And so, essentially, the the prostate's enlarged preventing, like, a urinary stream from going going through. And so the idea is this procedure, takes a laser and will will basically decrease the size of the prostate that's that's, compressing the, the urethra and preventing urine from going through. And so, Andromeda being able to use, integrate with existing platforms enables better guidance of that laser to, provide more precise treatment that otherwise would be pretty difficult by a surgeon that maybe doesn't have as much experience with that platform or that procedure. Is that some kind of understanding, or or, how how would

你会怎么概括？简而言之，没错，HOLEP是治疗前列腺增生（BPH）的最佳手术方案。目前已有数百篇论文充分证实，它在安全性和疗效上达到了最佳平衡，而且是唯一能彻底切除所有增生组织的根治性疗法。

you say that was? Basically. So, yeah, HOLEP is the best procedure for BPH. It's pretty well established in hundreds of papers now. This is the one that has the best combination of safety and effectiveness, and it's doing it's it's the only one that's curative as in far as far as it removes all the enlarged tissues.

患者术后无需再接受其他治疗，同时手术侵入性较低。虽然也可以用达芬奇腹腔镜机器人进行手术，但相比经尿道自然腔道的HOLEP术式，前者创伤大得多。这个手术就像从内部挖除橙子果肉——只不过在这里，需要去除的'果肉'是增生组织，你要沿着这个近似球形的腺体外科平面将其完全剥离。

You're never gonna need anything else, but then it's not overly invasive. You could do a procedure with the da Vinci with, you know, abdominal laparoscopic robot, but that's much more invasive than doing a a whole up, which is transurethral natural orifice. And it's, yeah, basically, like coring out an orange from the inside where you need to separate the pulp from the peel and the pulp unlike an orange, the pulp is what you don't want in this case. It's the enlarged tissue, and you wanna get that all out. And so you're following the surgical plane around this roughly spherical gland, and then you wanna remove all that enlarged tissue.

对于像新西兰的Peter Gilling这样的专家——他是HOLEP的发明者，可能也是该领域最顶尖的术者之一——完成手术就像本能反应，三十分钟就能完美搞定。但普通术者通常需要两小时以上，操作吃力，有时还会导致患者出现并发症。

And it's something that when you're an expert, when you're some of the people we work with now, like, you know, Peter Gilling in New Zealand who's the inventor of whole app and maybe one of, if not, the best operators in it. I mean, it's it's, like, second nature. They can do it in thirty minutes, and it's it's perfect every time. But then your average operator is taking two hours or more and really struggling with it. Sometimes they have issues that result in in, problems for the patient.

因此我们的目标是让每位医生都能像世界顶级专家那样高效自信地完成手术，确保最佳疗效且零并发症。我们要在这个已经处于临床最高水准的手术基础上再进一步——不仅针对HOLEP，对其他术式也是如此：选择现有最佳方案，然后精益求精，始终为患者提供最优临床结果。

So we wanna basically make everybody as efficient as the best in the world and as confident and able to deliver the best possible outcomes and not have any complications at all. So taking the the best procedure, which is already on the highest clinical ground you can be on, then raising that up higher. So we really wanna do that. I mean, not just for this procedure, but for others is take the best option that's out there and then make that better so that this is always gonna be the clinically best outcome. If you wanna get the optimal result for the patient, that this is gonna be what, what's gonna get you there.

这正是我们从HOLEP开始的目标——该术式本身已有惊人数据支撑，无需大规模研究来证明其有效性。我们可以直接认定这就是目前最好的治疗方案。

And that's that's our goal for any procedure starting with whole app, which, you know, already has phenomenal data to work with so we don't have to go and do huge studies to show that whole app works. We already can, you know, just take that as a given that this procedure is is already the best.

所以你们本质上是在让每位外科医生都能以最高水平完成高度复杂的专科手术，或者说，实现复杂手术（乃至所有手术）的普及化。

So you're essentially allowing every surgeon to perform highly complex specialized procedures at the highest level, or in other words, democratizing complex or any procedure.

我很好奇。这正是将手术流程民主化，让任何地方的人——无论是世界哪个角落，甚至未来某天在太空，比如如果你在火星上成为首批定居者之一——都能获得这种治疗，得到可能的最佳护理，尽管那里没有理想的外科医生。手术结果的差异之大令人惊讶。很多时候，不幸的是，患者遇到的不一定是最有经验的外科医生，意外高比例的手术效果未能达到最佳状态，这正是我们想要解决的问题。

I'm curious. That's exactly democratizing the the procedure and giving everybody anywhere, you know, even in any parts of the world or even in space someday, maybe if you're on Mars and you're one of the first people to to live there, that you can you can have this and you get the the best possible care you could possibly get despite not having, you know, the ideal surgeon there. It's it's amazing how much surgical outcomes can vary. I mean and a lot of times, you you aren't getting the most experienced surgeon, unfortunately, and and things go not as well as they could on a surprisingly large percentage of the time, so we wanna solve that.

这确实令人振奋。我们一定会密切关注你们的后续动态。我想稍微转换下话题，谈到自主手术机器人时，我们需要考虑两种训练：首先是培训外科医生使用你们的机器人，让他们适应仅凭iPad操作而非传统控制台，不时刻保持对机器人的完全掌控。不过你们已经部分涉及这方面了。

That's really exciting. We'll definitely keep an eye out for your upcoming news. I just to shift gears a little bit, there's two types of training we have to think about when it comes to autonomous surgical robots. First is training the surgeons to use your robots, getting comfortable with just having an iPad without having a full console, without having full control of the robot at all times. But you've touched on that a little bit.

另一种训练是关于模型训练。模型如何真正区分健康组织与病变组织？哪些该切除哪些该保留？你们是如何解决这个问题并训练模型的？

Let's the other type of training we wanna discuss is training the models. How does a model actually know what is good tissue, what is bad tissue, what should be excised versus what should not be touched? How did you go about approaching that and training the models?

是的。我们的做法是与全球收集大量手术视频（如HOLAP）的外科医生合作。通过整合拥有最大数据集的外科医生资源，我们建立了目前全球最大的HOLAP视频数据库。我们希望基于数千例临床案例的专业知识来构建这个系统。当然，随着机器人投入使用并积累案例数据，它会持续学习，未来将达到数万甚至更多手术量级。

Yeah. I mean, the way we started to do this is we partnered with surgeons around the world that are collecting a lot of videos of procedures like HOLAP. And now we've built the biggest HOLAP video dataset, I think, in the world right now by partnering with the surgeons that had the biggest datasets and then combining those into one. So we wanna build this on the back of thousands many thousands of cases of expertise. Then, of course, as we launch the robot and it does cases, it learns from that data, and then it'll be someday tens of thousands or more procedures.

某种意义上它将比世界上任何单一外科医生都经验丰富。至于解决最复杂的问题，有时需要将其拆解为更简单的问题。比如在整体切除手术中，追踪组织平面就是主要挑战——这个平面并不总是清晰可辨。我用橙子作过比喻：用内窥镜和激光分离橙瓣与果皮相对容易，边界清晰可见。

So it'll have way more experience in a way than any single surgeon in the world. And then as far as solving the hardest problems, you really have to break them down into simpler problems sometimes. So one thing we've the main challenge of whole up is following this plane, and it's not always as distinct. I mentioned the orange analogy where, it'd be pretty easy, I think, with a scope and laser to separate the an orange pulp from appeal. You probably could see where that was where the delineation was.

但在HOLEP手术中就没那么简单了。尽管视觉线索不总是明确，我们发现了其他判断依据——具体细节可能涉及太多商业机密不便透露。不过我们可以将这些额外线索输入算法，通过多维度数据实现精确的平面识别。机器人的优势之一在于它们具有完美的空间定位能力。

But in HOLEP, it's not as easy. But despite the visuals not always being clear, there are other clues we figured out. I'd probably be giving away too many trade secrets to say what those what those are. But there are other clues we can feed into the algorithms that build up to very accurate plane recognition from these different inputs that we have. Robots one thing good about robots is that they they have perfect sense for where they are.

人类医生在这边进行剜除后又移回原位时，很难精确记得之前操作的具体位置。但机器人能以亚毫米级精度记录每个动作，我们可以利用这些历史位置数据来指导后续操作。

When you're a human and you're doing some enucleation over here and then you move back over here and you wanted to know where you were when you were on this side, it's you're not gonna have a perfect sense. You're gonna have to basically approximate where you were. But the robot knows exactly where it was to the submillimeter precision, so we can track that and then use those positions in the history that of what it's done already to be able to inform what it's gonna do next.

确实很有道理。那种精确度是常规技术无法达到的，而且确实需要很长时间来掌握。我认为医疗保健领域，尤其是临床应用中最有趣的一点就是整个监管环节。这常常会成为一道门槛——而且往往是合理的——以确保我们提供给患者的治疗方案是安全的，并且符合应有的标准。

Yeah. That makes a lot of sense. That precision, that is something you just don't get with normal techniques, and it it definitely takes a long time to learn. I think one thing that's always been interesting with people building in health care and especially clinical applications is the regulatory piece of it all. I think that can often be a barrier, often rightfully so, to make sure that the things that we're, you know, like, letting patients have in terms of treatments are safe and, up to stand the standard that they should be.

有意思的是，正如你提到的HOLAP技术已被确立为治疗BPH的最佳护理标准之一。但当你结合仙女座技术进行增强时，情况就有些不同了。我很想了解更多关于将其作为既定手术流程使用时，如何通过仙女座技术以不同方式操作所涉及的监管流程，以及这个流程具体是怎样的。

It's interesting because, like, HOLAP, like you mentioned, has already been established as one of the best standards of care for BPH. But then when you augment it with the, Andromeda technology, it becomes a little bit different. I'd love to hear a little bit more about the the regulatory process of using it as established procedure, but then doing it in a different way with Andromeda and how what what that process even looks like.

是的。在考虑监管方面时，关键是我们不把这视为针对BPH的干预措施，而是一种整体提升手术水平的方式。我们不需要证明长期效果更优——这已是既定事实。因此这至少减轻了我们的举证负担，因为我们并非在尝试完全创新的东西。

Yeah. And it's it's really important when we think about the regulatory side of things that we're not really framing this as a BPH intervention. It's a way to do surgery in general better. We're not trying to show that there's better effectiveness in the long run that's already taken as a given. So it does reduce the burden of proof on us at least because we're not trying to do anything entirely new.

我们本质上是在以世界顶尖专家的水准实施整体提升，而非完全不同的手术流程。如果是全新手术，我们就必须进行长期研究并观察效果。但从监管角度看，我们已与FDA打过很多交道，虽然在美国以外其他地区的审批经验相对较少。有趣的是，科技圈尤其那些没有医疗背景的人总认为FDA是个黑箱，只会不断抛出法规要求而你无法影响或预测他们的决定。

We're we're basically trying to do whole up as well as the world expert, not not entirely different procedure. If we were doing a different procedure, we'd have to do these long term studies and, you know, look at how that performed. But, yeah, I mean, from a regulatory standpoint, we've dealt with FDA a lot now. We haven't really done as much with other bodies outside The US besides getting trials started in different countries. It's it's amazing how I I think there's this belief among people in the tech world, especially, that that haven't lived in the medical field that FDA is is a black box where they're just gonna throw regulations at you or requirements at you, and you can't really influence that at all or you can't predict what they're gonna say.

这完全不符合事实。我们发现这是个高度协作的过程：我们举行了多次预提交会议，现在已向FDA提交了510k文件，整个沟通过程非常顺畅。我们基于基本原则制定了产品开发和测试方案来证明安全性。

And that's that's really not true. I think I mean, we found it to be a really collaborative experience. We had multiple presub meetings. Now we've submitted a five ten k to FDA, and our interactions have been phenomenal. I mean, we've we've come up with these first principle with ways of how we're gonna develop and test this product and show that it's safe.

FDA与我们的理念高度一致。只要你诚实地用数据证明产品的安全有效性，FDA很可能会认同你的评估。只要你采取正确的安全策略，通过合理方式降低风险或证明产品的风险收益比，在99.95%的情况下都不会遇到大麻烦。偶尔会有特殊情况，但通常没那么糟，实际过程远比人们想象的轻松。我认为这种风险被夸大了——至少我们的体验一直是非常积极且具有建设性的。

And FDA has been incredibly well aligned with that, that it really is if you if you're honest with yourself and you come up with a way with with the data that's required to show this thing is gonna be safe and effective, then it's very likely that FDA will agree with that assessment. And if you are taking the right approach to safety and, just doing the things that legitimately will derisk or show what the risk risk benefit is of the product, then you're not gonna have much trouble with FDA in ninety nine ninety five or more percent of cases. There's weird things happen occasionally, but it's almost always not that bad, and and it's a lot less scary than people think it's gonna be. So, yeah, I I think that that risk is is overblown. And, I mean, so far for us, the experience has been really, really positive and and collaborative.

没错。当你试图进行渐进式改良或对现有操作方式进行微调时，获得FDA或其他监管机构的批准要比全盘推翻重来容易得多。

Right. It's much easier to get approval from the FDA or any regulatory body when you are trying to make an incremental change or trying to just slightly modify the way things are already done instead of throwing out the entire book and starting from scratch.

是的。我还想说，应该将其表述为收益与风险的权衡，并明确指出我们正努力为患者带来这些益处。当然，风险确实存在。我们会坦诚透明地对待风险，但必须将其与收益进行权衡。我认为FDA和其他监管机构如CE标志、日本PMDA等，都希望为公共健康做出正确决策。

Yeah. And I would say also to frame it as benefits versus risk and really say, we're trying to deliver this this benefit or these benefits for patients. And, yes, there are risks. We're gonna be honest and transparent about the risks, but then we have to weigh those against against these benefits. And, I think FDA also wants to do what's right for public health as well as any other body, you know, CE Mark or PMDA in Japan or whatever else you wanna look at.

但确实，这本质上也是他们的目标。如果你们能基于共同使命达成一致，效果会远比将其视为对立面要好得多。

But, yeah, I mean, that that's really what their goal is too. And so if you align yourselves on a shared mission, then you'll do much better than thinking of them as an adversary.

没错。归根结底，尽管通过FDA审批存在诸多障碍和挑战，但你们的目标是一致的——都希望帮助患者并降低风险或负担。说到获得监管批准和进行试验，如果我没记错的话，你们在2024年12月作为ASTROT试验的一部分完成了首次全实验室手术。目前最新进展是，我读到你们已在智利和新西兰完成了约15例手术。

Right. I mean, at the end of the day, despite the hurdles and challenges of going through FDA approval, your and their goals are aligned. You want to help patients and minimize risks or burdens. But speaking of getting regulatory approval and just going through the trials, if I recall correctly, you had your first whole lab surgery as part of the ASTROT trial back in December 2024. You've currently had it last, I read about 15 procedures done in Chile and New Zealand.

能否详细谈谈进展如何？以及为何在190多个国家中，最终选择这两个国家作为试验地点？

Can you tell us more about how that's been going, and how did you come to the conclusion that these would be the two best countries to pursue trials in compared to the other 190 plus countries out there?

是的，之后我们又完成了几例。但最初是12月从智利开始的。我们启动时就想设立两个不同的手术地点——因为在临床试验中我见过太多次这样的情况：当向某个监管机构提交材料时，意外状况不一定是不可预测的，而往往是导致延误。

Yeah. We've done more a few more since then, but, yeah, that's where we started. We started in Chile initially in December. And we basically, when we started, we wanted to have two different places we could go and do cases. We because I've seen in I've seen so many times in clinical trials that something goes wrong when you're submitting to some regulatory body that, things not necessarily are are unpredictable as much as they get delayed.

各种原因都可能导致进度推迟。这次我们想设置双重保障。后来新西兰因某些特殊原因延迟了审批，而智利方面进展更快。我们最终在一月下旬进入新西兰，时间上也没差太多。

I mean, things things get pushed back for all different reasons. And in this case, we wanted to have two shots on goal. We ended up getting delayed for some weird reason in in New Zealand, and then, Chile was able to move ahead sooner. We ended up going to New Zealand in late January. It's not not that much later.

选择新西兰的原因在于——HOLAP技术的发明者暨该领域顶尖专家就在新西兰。这项技术诞生于奥克兰三小时车程的小城陶朗加，最早发表于1997年。而在智利，我们与圣地亚哥的泌尿科医生莱德斯马建立了合作，他对参与项目表现出极大热情。

But then yeah. I mean, the reason why we went there so in New Zealand, I mentioned the inventor of HOLAP and the top thought leader in the field is in New Zealand. It was invented in a smaller city called Tauranga in three hours from Auckland down in New Zealand. And I think 1997 is when it was first published, and he's been kind of the top expert since then. And then in Chile so we had we connect with the urologist there, doctor Ledesma, down in Santiago, who, was just really enthusiastic about being part of this.

首先，他告诉我们整个过程相当直接明了。他还有许多其他中心的朋友也可以参与这项研究。于是我们前往调查，在圣地亚哥市与他及其他泌尿科医生会面，对医生、医院以及整个流程有了很好的了解，一切看起来都非常顺利。这就是我们决定推进的原因。关键在于与合适的外科医生和机构合作。

And he told us he had a lot of it was a pretty straightforward process, first of all. Then he had a lot of other centers that he was people he was friends with there that could also be part of the study. And so we went down there to investigate, and we met with him and other urologists in the in the city of Santiago and, just really got a good feel for the the doctors and the the hospitals, and the process seemed really straightforward. So that's why we did it. It was it's really mostly about working with the right surgeons and sites.

此外，现场团队也非常重要，比如研究协调员和其他参与研究的人员。你必须对所有利益相关者进行筛选，确保流程高效畅通。这两个国家的审批流程都很直接，耗时不长。对于初创公司来说，时间成本极其昂贵——当你准备好获取数据时，就希望能立即得到，而不是等待六个月。因此我们也希望快速推进。

Also, the site teams matter a lot, like the research coordinators, Whoever else is involved in the study, you have to you have to kind of vet all the stakeholders, all the people that are that are part of it, and then the efficiency of just getting things cleared. So, both those countries have pretty straightforward processes that don't take that long. So when you're a start up company, the cost of time is I mean, time is so expensive also that when you're ready to get the data, you wanna get the data. You don't have to wait six months to get the data. So we we also wanted to move it move it fast.

我们还想谈谈你们如何让外科医生使用这个平台。看起来很多人对你们开发的产品及其在手术中的应用感到兴奋。根据我的临床经验，许多医生对使用新技术往往持怀疑或犹豫态度，特别是面对不熟悉的技术时。当然不是所有人如此，但我很想知道你们联系医生或寻找潜在用户时的经历？他们的学习曲线有多快？整体接受度如何？

We also wanted to talk about you know, you've gotten these surgeons using the platform, and, seems like a lot of people are really excited for what you're building and how it's being integrated into their procedures. When I think about lots of doctors that I've interacted with, at least during my clinical experience, there's often a lot of, like, skepticism or hesitancy about using new things, especially when they're maybe not what they're used to. And, yeah, there's just this strong skepticism towards particularly things like technology. Of course, that's not everyone, but I I would love to hear about what has it been like, you know, contacting doctors or finding people that have, you know, potential to use the platform? And how quick has that learning curve been and reception in general?

你们需要筛选出那些愿意使用新技术且态度务实的医生。在初期病例中，绝不能选择那些——

You've gotta select for the doctors that want to use new tech and are Yeah. Realistic about what it's gonna be like. You don't want somebody in your first cases that

——指望使用经过二十年验证的达芬奇机器人系统的医生。这类人选显然不合适。无论是临床试验还是商业销售，选择首批用户时必须精挑细选。他们需要具备早期采用者的特质，清楚自己参与的项目性质，且不会有脱离实际的期望。

are that's gonna expect a da Vinci robot that's that's been tried and tested for twenty years. I mean, that's that's the wrong person. And so you have to be selective too with a with a trial or with commercial sales for that matter when you're looking at your first commercial adopters. They need to be the early adopter profile. They need to understand what they're getting into and not have unrealistic expectations.

即使某些医院或外科医生可能达成合作，也未必是合适人选。因此我们会有意暂缓吸纳部分人员。在研究过程中，我们严格筛选了具有正确思维模式的研究者——他们理解这是早期阶段，并愿意协助我们完善产品。初期我们就知道软件需要迭代，也确实这样做了：在病例间隙进行软件修改，通过台架测试甚至尸体实验验证，几周后带着更新版本继续开展更多病例。

So even if you can make a sale to certain hospitals or surgeons, it may not be the right person. And so there's maybe people you intentionally wait to to bring into the the fold. And so for that, we we really did filter the the investigators in our study for people that were definitely of the right mindset, that they were they understood this as early, and they were gonna help us develop it. So that in the in the beginning, we knew there were gonna have to be changes made to the software, and we did that. In between cases, we would, you know, in in some cases, go to some cases, go and, you know, come back, make software changes, do testing on the bench or even cadavers, and then come back a couple weeks later and do more cases with the the changes implemented.

我们需要的是能最高效协助产品改进的外科医生，他们提供的反馈将帮助我们做出正确的产品优化决策。

So we wanted surgeons who were gonna help us most efficiently improve the product and, you know, also give us input that would help us make the right changes to the product.

你几分钟前提到的一点是希望尽快获得数据，不想等待六个月、一年或更长时间才能拿到可用数据并推进工作。如果我没记错的话，安德洛墨达外科从构想到实际执行第一例手术只用了大约十八个月。这速度快得惊人——对任何初创企业都算神速，尤其在医疗科技领域，或者更具体地说在外科手术领域。

One of the points you mentioned a few minutes ago was that you want data as fast as possible. You don't wanna wait six months or a year or a long time to get usable data and move forward. If I recall correctly, Andromeda Surgical went from being idea to actually performing the first procedure in around eighteen months. That's incredibly fast. It's in it's fast for any start up, but especially when it comes to health tech or even more specifically when it comes to surgery.

这个领域的创新通常非常缓慢。要突破重重障碍、跨越各种门槛，最终能在患者身上进行测试极其困难。你们是如何做到如此快速的？如何在短短十八个月内完成这一切？我知道两位都有丰富的经验。

Innovation in this space is very slow. It's very hard to jump through the barriers, jump through the hoops, and actually get to test on patients. How did you go about moving so fast? How did you do every all this in just eighteen months? I know both of you have a lot of experience.

你们是经验丰富的创始人，能否详细谈谈团队组建过程？在这个变革阻力大、发展缓慢的行业里，你们如何打造出能如此快速推进的工作环境？

You you're, experienced founders, but can you tell us a little more about how did you form your team? How did you go about forming an environment where you can move this fast in an industry that's so resistant or slow to change?

我认为关键在于公司的构建方式和战略选择。创业初期我们就明确：基于我在医疗科技行业的经验，最困难的是复杂硬件的研发、验证和测试。复杂度每增加一级，所需时间几乎呈几何级增长——因为涉及大量零部件、组件、子系统及其交互关系，任何环节都可能出错。比如包装、灭菌这些环节都可能出问题。

I think it starts in the way you build the company and that the battles you choose to fight. So the beginning, we said, okay. We're not gonna having been in med tech, I realized the hardest things are building and validating and testing complex hardware. The more complex it gets, the time it takes is almost like the square of the complexity or something that it's it's way above linear because there's just so many different parts, components, subassemblies, interactions between those that can go wrong. And I just think there's things like, you know, packaging, sterilization, all this stuff that can that can go wrong.

感觉总会有意外发生。所以我们尽可能规避这些风险。其次我们决定不承担任何临床风险，只开展现有成熟术式。第一阶段目标仅仅是达到人工操作的水平——用机器人复现医生的手动操作。

Something always does, it feels like. So we tried to minimize that as much as possible. And then, also, the second thing we decided was not to take any clinical risk where we were gonna do procedures that already existed. And our first goal was just to do those as well as the person as someone does manually. So we just wanted to replicate what they're doing manually with a robot.

因此，精准界定要解决的问题范围，明确做与不做的边界，是能够快速推进的最重要因素。我们创立公司的目标就是在一年半内进入临床阶段。接下来就是执行力的问题——组建的团队工作极其刻苦，大多数人完全没有医疗科技背景，团队里几乎没人来自这个领域。

And so just scoping the problem we wanted to solve and the things that we chose to do or not to do is probably the most important factor to being able to move fast. We set up we set up to build a company that could get to the clinic in in a year and a half, and then it comes to executing and actually doing that, which is, again, a a a very nontrivial thing to do. We we build a team that I think was just incredibly hard work. Most of them are not from med tech at all. Almost nobody in our team is from med tech.

我是少数有相关背景的成员之一。团队对公司愿景高度认同，大家拼命工作、严守期限、保质保量完成任务。幸运的是我们都有优质人脉网，能从之前共事过的优秀人才中招募成员——这对首次创业者恐怕很难复制。当然并行推进多项工作始终是明智之举。

I'm I'm one of the few that that is. And it's just really well aligned on what we wanna be as a company and just pushing super hard, people that you can rely on to meet the deadlines and just get the stuff done and and deliver high quality work. Fortunately, both of us that had had great networks and prior companies where we had worked with great people and were able to pull from from some of that, and, that really helped a lot. That's something that a first time founder wouldn't probably be able to do as as well. And then, yeah, I mean, just being smart, like, parallel tracking things is always a good idea.

正如我提到的，我们同时去了两个国家，如果只去新西兰的话可能会耽误几个月。这就是那种情况。然后对于供应商来说，如果你要加急，就必须考虑时间成本。假设每月消耗40万，那么每周就是大约10万，对吧？

Like I mentioned, we we went to two countries at once, and we would have gotten delayed by a couple months in New Zealand if we had just gone there. And so that's that kind of thing. And then then also for for vendors, I mean, if you expedite, you have to factor in how much how much does time cost us. How much does if you're burning, let's say, 400 k per per month, then every week is about a 100 k. Right?

所以要把这个因素考虑进去。如果在关键路径上花5万就能节省一周时间，而这个环节正是最大的瓶颈，那就值得做。如果真能节省一周，那就很划算。这种思维方式几乎没人具备——有人看到能节省一个月的1万加急费却不去做，我觉得这很疯狂。

So then just factoring that in. And if you can pay 50 k to save a week on the critical path where this is the thing that's most holding you up, then then you do it. If if that's if that's, you know, actually gonna save you a week, then then it's worth it. So that's a mindset almost what nobody has where people look at, like, a 10 k expedite fee that'll save them a month and don't do it, which I think is crazy.

选择你的战场、组建合适的团队、不要孤注一掷，这些建议非常精辟。关于你的团队，我很好奇你的第一位员工是谁？当初决定扩大团队并加快进度时，你是如何确定需要招聘什么职位以及选择人选的？

Choosing your battles, having the right team, and not putting everything in one basket. It's phenomenal advice. Going more into your team, I'm curious to hear about who was your first hire, how to go about deciding what role you wanted to hire for and who you wanted to hire when you first decided to expand your team and try to move a little bit faster.

我正在回想具体是谁，因为最初是承包商后来转正的人。我想第一个可能是拉米。我是通过斯坦福生物设计项目认识的一位共同朋友结识他的，他曾在强生AI手术小组工作过，是受过专业训练的外科医生，做过实际手术，这非常有意思。

I'm trying to think who was the actual because the people that were contractors at first and they became hires, I think the first was probably, Rami. So he was I met Rami through a mutual friend who I knew from Stanford Biodesign, and he had been at J and J in their AI surgery group. And he's a trained physician surgeon. So he's done he's done surgery. And he it was really interesting.

我们发现——其实是他和不久后加入的米歇尔（她刚从UCSF住院医师项目毕业）都是外科医生——这类人有个独特的角色定位：他们既能提供手术背景和经验，又能为我们做临床工作，还能从外科角度为产品功能提供建议。对于AI标注这类工作，虽然当前方案扩展性不强需要改进，但...

We we figured out, and it was both both him and actually Michelle who joined soon after that. We're both surgeons. Michelle came from UCSF residency and and joined us full time, that we realized there was an interesting role for people like that where they could really be have the surgical context, have that experience, and be able to to both do clinical work for us and then also be able to kind of advise on product features from a surgeon context. And then also for things like AI annotation, I mean, we don't wanna this doesn't scale that well. We're gonna have to find a a different solution eventually.

让外科医生标注我们使用的图像或视频，这是构建手术AI的核心难题之一，因为外行无法完成，必须由懂行的人操作。早期他们确实亲力亲为标注了成百上千的视频，这个我们新创设的角色发挥了重要作用。

But having surgeons annotate the images or the videos, what we're doing, it's one of the core problems of building surgical AI is that the layman can't do it. You need somebody who knows what they're looking at. So that also really helped. In the beginning, they were actually putting in the the legwork to annotate these hundreds or thousands of videos. So, yeah, really interesting role that we now created.

我认为我们会把这个模式复制到其他医学博士和外科医生身上，那些有临床经验的人现在能为我们做很多不同种类的辅助工作。

I think we'll replicate that across other MDs, surgeons that, you know, came from practice and now can can do a lot of different things that are helpful to us.

确实。你们组建了如此多元化的团队令人振奋。正如你所说，早期引入临床专家来掌握手术流程的专业知识至关重要。我觉得最有趣的是，就像Pranav之前提到的，你们在技术临床应用这个通常进展缓慢的领域，如此迅速地取得了这么多成就。那么对你来说，展望仙女座公司未来五年，你认为怎样的发展才算成功？

Yeah. It's really exciting you've built such a diverse team. I think, like you said, having the clinical people early on to have that expertise of how procedures need to be done is is so important. I think what's really interesting, like you like Pranav said earlier, is how much you've done so quickly, especially in a space that's typically slower moving in terms of, like, clinical applications of, of technology. So for you, when you're looking at Andromeda, what do you think, like, what do you think success looks like for the company, in the next five years?

你之前提到一个非常有趣的观点：即使在审视临床试验或验证技术时，也从不局限于比如仙女座与Holip的组合，而是着眼于仙女座如何赋能整个外科领域——我们如何全面提升外科手术？这种宏观视角令人振奋。但你们将如何规划仙女座未来几年的发展路径？

You said something really interesting earlier where, even when you're looking at clinical trials or validating your technology, it's never looking at, for example, like, Andromeda with Holip. It's looking at, like, Andromeda with surgery in general. Like, how are we augmenting surgery in general? And so I think taking that big picture approach is really exciting. But, how do you how do you approach the next few years of how you're building Andromeda?

我们的使命是让每台手术都达到最佳效果，这正是我们努力的方向。我们坚信这个目标可以实现，希望有朝一日我们的产品能成为几乎每台介入手术的核心，为每位患者带来最优结果。我甚至设想过在太空等极端环境实施手术的可能性，本质上就是让最优医疗方案在任何场景下都能普及。我们要尽可能接近这个理想。

I mean, the mission is to make every surgery go as well as it possibly could, and that's we wanna be building towards. So we we believe this is we believe this is possible. We hope to someday have the product that's the center of almost every interventional procedure and brings the optimal outcomes to every patient. And, you know, I've mentioned other crazy possibilities like doing this in space or other places, basically democratizing the best possible care across every setting you can can get to. And I think we wanna we wanna get as close as possible to that goal.

我们会持续建设直到达成目标。同时，也希望激励更多人朝这个方向努力。有意思的是，我的第一家公司就存在类似轨迹：通过New Rep或Avail，我们开创了远程手术支持的概念。虽然公司最终被收购时估值不算惊艳，也不能算巨大成功，但它确实催生了现在由多家企业主导的整个产业，最终成为了我们预期的主流模式。

We're gonna keep building until we get there. And, you know, also kind of inspire others to to move in this direction. Also, my my first company is interesting as a parallel that, with new rep or Avail, we created this concept of having remote surgical support. And the company itself got pretty big, but then it got got acquired, but not for an amazing deal. And, it didn't it isn't really considered a it wasn't considered a huge success in the end, but it did start this whole industry that now is a big thing with multiple companies is is becoming the the dominant paradigm we thought it would be.

某种意义上，我们成功创造了想要的世界变革。就当前技术发展阶段而言，我们认为这是对外科领域最具影响力的事业，而我们希望成为实现这一变革的推动者。当然，我们并非抱着硅谷那种'不想活在别人比我们更能改善世界的世界'的心态——我们真心希望世界变得更好，只是更希望自己成为那个推动者。

So in a way, we succeeded in creating the change in the world we wanted to create. In this, in this sense, in this case, I we we wanna create that change. We think it's this is the most impactful thing you could do for surgery in at this current time in in terms of where we are technologically, and we wanna make that happen. We wanna be the ones that do it, but then we're not like if you know the Silicon Valley quote of, I don't wanna live in a world where somebody makes the world a better place better than we do. We, we want the the world to be a better place, and we hope we're the ones who do it.

但即便不是我们，我们也乐见其成。

But if if not, we hope that it happens.

这就像超前的理念最终催生出受你启发的产品，而现在你们显然站在这项技术的最前沿。顺便问个创始人常被问到的问题：仙女座这个名字的由来？我自然会联想到银河系，但很想听听你们命名背后的故事。

Yeah. It's like that concept of being ahead of your time and then seeing the the products that come out after that be kind of inspirations of yours, and and now it seems like you're really at the forefront of this technology. And I think, you know, saying saying the company name multiple times, we also love asking founders where your company name comes from, Andromeda. I mean, I think about the galaxy, but would love to hear if there's any kind of backstory behind how you guys came up with the name.

是的。这是银河（galaxy），不是神话（myth）。当初命名时，我确实不知道有个相关神话。虽然我知道那是个希腊人物，但不知道具体故事。真正重要的是我们要将技术推向银河级的高度。

Yeah. It's it's the galaxy, not the myth. There's there's some there's a myth that I wasn't really aware of when we came up with the name. I I knew that it was a Greek figure, though I didn't know the story. And it's really about that we're taking things to the next galaxy.

我的意思是，我们团队对太空主题情有独钟。市面上有很多太空主题的名字，但这个不止是月球、火星或海王星之类的概念。它代表着颠覆性的宏大目标——事实上银河系最终会与仙女座星系碰撞融合，就像这项技术将与现有外科手术结合，最终创造出更强大的未来医疗体系。

I mean, this is something that there we we like space a lot. There's a lot of space theme names out there, but with this is, like, not just it's not just the moon or the Mars or Neptune or whatever. This is something that's, you know, big goal way out there that's totally paradigm shifting. And, actually, it's gonna collide with the Milky Way eventually and merge into one where this is it's gonna gonna be the future of surgery, and it's gonna it's gonna merge. This technology is gonna merge with what surgery is now and create something bigger and better in the end.

我很喜欢这个关于'仙女座外科'命名的故事。在结束前，您对听众有什么建议？特别是那些想进入医疗行业、开发自己产品并产生影响的人们。

I love that. That's a good story of the Andromeda Surgical name. As we begin to wrap up, what advice do you have for our listeners? People who are interested in health care and who may want to build their own product and make their own impact in this industry.

确实。我特别认同萨姆·奥尔特曼最近说的'只管去做，不要害怕'。虽然我自认初期还算大胆，但依然觉得自己没准备好直接创业，尤其是在这个领域。关键别被那些限制性思维束缚——只要你有资源整合能力和学习意愿，在行动中不断摸索，惊人的事情就可能实现。

Yeah. I would say I mean, I I really like the Sam Altman's recent quote of you can just do things that this is to not be afraid. I think I was a little bit I mean, I was definitely on the the bolder side as far as just doing things in the beginning, but I still felt like I wasn't ready to go and and be a founder right out of the gate, especially in this space. I would just don't be constrained by that stuff that you can always you can always figure out how to do things. If you're somebody who's resourceful and willing to learn and just figure things out as you go, that crazy things are possible.

现在回想起来，我觉得自己好像必须先创建两家公司才能做好现在这个项目。但其实职业生涯早期也可以尝试，可能过程不会这么顺利，但确实不该畏惧挑战大事。别听那些'先从小规模开始逐步发展'的建议——你完全可以从大处着眼，做自己真正热爱的事。对事物投入更多热情，结果就会更好。

So, yeah, I I mean, now I'm I I had I felt like I had to build two companies to get to the point I could build this company. But now I think I probably could have done it early in my career too that maybe it wouldn't have gone as smoothly, but I think you shouldn't be afraid to try that kind of thing and and go for big things. Don't listen to the advice of you need to do something small and incremental that, to start, and then you can build to a bigger company. I think you can you can do the big company in the beginning and just do things that you're you're passionate about. If you care about something more, you're gonna do a better job at it.

如果让我做完全不关心的领域——比如帮保险经纪人提高利润之类的——我根本不会有动力。即使具备创业者技能，也做不出好成绩，因为缺乏真正的热忱。所以要选择你真正在乎的领域。如果发现值得解决的需求，就找个切入点直接动手创造。

If if I had built something about I don't know what I what's something I don't care about, like like, making insurance brokers more profitable or something like that, that I wouldn't have been my my heart wouldn't have been in it. And even if I'm, you know, skilled as a founder, I probably couldn't have done a great job building that company because I wouldn't have cared enough to to really do it. So pick the thing that you really care about. If there's a need you really wanna go after, then try try to find an angle on that and just just build stuff.

这建议非常实用。尤其是对走临床医学道路的人——就像您之前提到的等级制度和僵化体系，容易让人产生'这些问题存在这么久，我能有什么贡献'的思维定式。但正如您所说，意识到有多少人同样想过却未行动，这个认知本身就很宝贵。这对临床背景的人特别有启发。

Yeah. That's that's really good advice. I feel like especially going down the clinical medicine route, like, you mentioned this before, but the hierarchical nature of it and the the rigidness that often makes you feel like, oh, well, if these problems have existed for so long, like, what am I gonna be able to do to contribute to those and get kind of, like, tunneled into that thinking of, like, well, if people have with more experience than me, like, haven't solved this issue, then how do I? But like you said, I think just being able to go and do it and and thinking about how many other people probably had that same mindset and didn't do it is is is huge. So, because that's that's really good advice, especially for people with a clinical background.

尼克，我们非常感谢你今天抽空参与。与你交谈非常愉快。在我们结束之前，关于你自己或Andromeda还有什么想分享的吗？

We really appreciate your time today, Nick. It's been it's been great talking to you. Is there anything else that you wanna share about yourself, about Andromeda before we wrap up?

目前我能想到的大概就是这些了。关于Andromeda，我想说的是——我们正全力打造一款将彻底改变外科手术范式产品，同时也在尽可能加快产品扩展并推向市场。

I think that's probably most of it for now that I can think of. Yeah. I mean, I'd say just, like, look out for I think one thing about Andromeda. So we're we're trying of course, we're focused on building something that we think is gonna be very paradigm shifting in surgery. We're also trying to do this to expand this product as as fast as we can and get this to market.

我们几乎肯定会成为上市速度最快的外科手术机器人，之后会快速发展。所以我想说，请保持关注，有很多令人兴奋的进展即将到来。如果有人想加入团队，我们当然会随着发展持续招聘。我们需要那些敢于打破常规、有动力在这个领域构建非凡事物的优秀人才。

We will be almost certainly the fastest surgical robot to market and then growing fast. So, yeah, I mean, I would just say, look out. There's a lot of interesting stuff that's coming down the pipe, so stay tuned for that. If anybody wants to be part of the the team, of course, we're we're gonna be hiring as that all happens. We want great people that are, I guess, contrarian and, you know, more motivated to build big things that are unconventional, in this space.

如果你希望参与其中，请务必联系我们。希望在不久的将来能分享更多有趣的新进展。

So definitely reach out if if you wanna be be part of that. And, yeah, hopefully, we'll have some interesting additional news to share before too long.

我们非常期待。非常感谢你尼克，感谢你的参与。认识你并聆听你的故事是我们的荣幸。

We look forward to it. Thank you so much, Nick. Thank you for joining us. It was a pleasure meeting you and hearing your story.

是的，很高兴认识你们两位。

Yeah. Great to meet you both.