I had the absolute pleasure during my engineering undergraduate (Oxford) to take a biomedical module. One of my 'labs' was on nonlinear acoustics, specifically ultrasound applied for therapeutic uses. It was very captivating seeing a very focused point within a block of gel become ablated. A part I found particularly exciting was realising that it was a phased array of ultrasonic emitters, so that the point where the ablation occurred could in fact be placed anywhere you desired in the gel.
They showed us results of HIFU applied to real patients to non-invasively ablate tumours and treat prostate issues. As far as I can tell the probe creating the ultrasonic waves needs to be relatively close.
A thought I had at the time was if you knew all of the material properties of all of the tissues inside someone and their locations (say with an MRI) you could in theory apply this even deeper in someone than is currently possible - with a larger stick-on patch of actuators as a phased array.
Finally, another memorable thing that was discussed was what another researcher was doing with ultrasonics.
Stride (who I am delighted to say was a fantastic lecturer) was very interested in bubbles. She would construct tiny bubbles where the surface (or interior?) was made of a chemotherapy drug. These bubbles could then be injected into someone's blood stream and would be ruptured using ultrasound to allow for extremely targeted application of chemotherapy (the jet formed from rupture would be so strong it would inject the drug into nearby tissue).
Fascinating, fascinating stuff but of course developed over many years of hard work.
> "Cancer is awful," Xu says. "What's making it even worse is cancer treatment."
Well said. And it's either terrible or expensive (and sometimes also terrible as well).
Proton therapy for instance is amazing at targeting hard to reach tumors like those in the eye, but costs close to fix figures as it requires a team of people to design the treatment.
At the intersection of ultrasound and startups (since this is HN), does anyone have any thoughts about that Openwater project? They are apparently working on open source ultrasonic medical devices.
I don’t actually know much about them, I just heard of them because their CEO (Mary Lou Jepsen, she’s quite famous, right?) was on the AMC podcast (months ago, actually, I was just going randomly though the back catalogue).
Tech folks pivoting to medical always throws off some alarm bells to me, but she was fairly compelling on the podcast and the basic idea seemed to make sense. Ultrasonic treatments, using diagnostic-level energies, using focusing and resonance based tricks, I guess. (It is way outside my wheelhouse, sorry if the description is inaccurate).
> Tech folks pivoting to medical always throws off some alarm bells to me, but she was fairly compelling on the podcast and the basic idea seemed to make sense.
The best way to evaluate biotech startups from the outside is to look at their investors. If they’re full of VCs specializing in biotech, chances are someone did the bare minimum due diligence on the science.
Theranos for example didn’t have a single one because biotech VCs steered clear of that mess entirely.
> Tech folks pivoting to medical always throws off some alarm bells to me
Same for me. I've been in the medical device industry for 15+ years now and came from "tech". What a lot of techies under/don't appreciate is that the medical device industry is heavily regulated and moves at a muuuch slower pace than other technologies.
There are lots of regulatory and quality/testing hurdles that you must clear (namely verification and validation testing, in addition to your 510(k) clearance or approval, if PMA) before you can market and sell your device.
I tell customers, on average, a Class II medical device project can take 18-24 months and cost $3M to 4M, minimum.
Yeah, it seems that their pitch is that they want to move at consumer electronics speed, I mean, their website explicitly says
“Our tech-driven approach leverages software, hardware and AI […]
That means we can iterate at the speed of consumer electronics”
Which is kind of scary but also a bit interesting.
How would you go about regulating an open source medical device? The user can just plop whatever software on there that they want, and ultrasound themselves wherever… play with resonance and focusing, right?
There was a game in the early '80s called Microsurgeon where you piloted a robot probe into a body to cure it of diseases. It was armed with an array of tools, one of which was ultrasound that you could use to destroy cancer. I wonder how long this idea has been around for.
The article mentions that this is a different type of ultrasound treatment than the one that has been in use for prostate cancer treatment for some time.
>But you cannot use this anywhere where the ultrasound would be blocked by other organs.
Yes you can. If you had an array of ultrasonic transducers around the body you could have each of them in phase targeting a single spot. Beamforming is a thing we've been doing for years with RF. It's even more trivial with sound.
We were privy to a lab that accidentally cooked mice with gold nanoparticles in the late 90s with multiple IR lasers. After they figured the power side, it turns out that gold nanoparticles are wildly cytotoxic on a number of axes.
How is it more trivial with sound? Sound is just a wave just like ultrasound. In fact, ultrasound has the word sound in it making it sound. So your conclusion is not sound.
Can’t the surgery be then with a small probe just to get the ultra sound tip near the cancer? I don’t know the size of the ultrasound tip but seems to me it can be smaller then a hand or tweezers.
Often constructive and destructive interference of waves can be used to focus the ultrasound through tissue without any incisions at all. Kidney stones are sometimes broken up this way.
> Fun fact: using this ultrasound for prostate cancer treatment reduces the risk of erectile disfunction
I’m not aware of strong evidence in this area (not saying you’re incorrect).
For the liver indications, several elite radiology departments have had very poor outcomes with their patients, despite the strong public data. I would not, with my own prostate, try a new technology until at least a decade out, at least.
Urinary and erectile function are a major issue with partial and radical prostatectomy. These ultrasound treatments are showing significant improvements in both areas.
This technology is also now used to treat non-cancerous prostate enlargement (BPH).
True, but currently prostatectomy is the most common intervention and second is radiotherapy which as mentioned in a comment above is a very expensive alternative with known side effects.
Histotripsy is early in its clinical life but I wouldn't say untested.
I remember seeing a demo from people who could slap a raw steak into one of these machines, and with ultrasound, sear their logo into the meat at sub mm precision. But that was long ago & not ready for medical usage yet. Cool that it seems to be used for actually treating people now.
I just spoke to a oncologist surgeon about this. Even though their facially doesn't have one (they are expensive machines) he said it's looking like it'll be standard care in the future.
He did an evaluation about getting one for my local hospital.
“Lithotripsy” is the name of the kidney stone treatment. My understanding is it’s based on vibration, not ultrasound (I know, vibration is sound - my understanding is the method on the linked article uses higher frequency + intensity + shorter pulses than the kidney stone method - so sorta like microwaving tumors vs using a massage gun on kidney stones?)
Having had kidney stones, they're both used. I think for the sonic one they put you in a water bath because it conducts better. But as I understand it, the docs can pick whichever one is more optimal, be it shattering the stone sonically or zapping it with a laser.
AFIKR two facilities do this kind of treatment. One in Canada and one in China. There already was a HN threads with some reporting to have been treated in Canada.
Apparently, only some tumors have a distinct and unique shape / size. The “trick” is to calibrate the resonance exactly to the size of the cancer cell. So that resonance would “hurt” only that kind of shape / size cell. Which was much harder to do than it sounds. Sadly not all cancer cells are unique and not that “easily” distinguishable by size
But I am not in the medical field and just repeating what I’ve read.
Watching Hank Green's YouTube video where he found out that his cloudy pee was cancer leaving his body, he was surprised that doctors don't tell you to expect it. It can be such a morale boost.
If I interpret the article correctly, the ultrasound energy does two things: it effectively destroys the cancer cells by overheating them, and it physically breaks apart the tumour. Your immune system can further break up and get rid of dead cells the way it deals with normal dead cells.
> Some researchers have raised concerns about histotripsy potentially seeding new cancer growths as tumours are broken up inside the body, meaning they can be transported to other areas. That fear, however, hasn't borne out in animal studies so far.
I doubt ultrasound would trigger apoptosis in cancer cells, one of the reasons they're cancerous is that they refuse to commit suicide when they should.
The only thing the article fails to mention is the use of more than one transducer used to focus multiple ultrasound beams to an intersection point in the body, increasing the heating power of all beams
There was a startup in Shanghai in the early 2000. Their device used multiple transducers. The probe was at least 40 cm in diameter. They did trials on uterine fibroids, among other diseases. One of the difficulties was while it looks good in theory, but the path ultrasound travels in the body is more complicated than, say x-ray or gamma ray. They expected a fine focal zone, but sometimes the focal zone was much larger than expected. This new wave of ultrasound equipment may have discovered better ways to control the sound beam.
Don't know about mRNA but individualized remedies based on CAR-T technology have been making significant strides in this area, with major commercialisation expected in the next 1-2 years
I had the absolute pleasure during my engineering undergraduate (Oxford) to take a biomedical module. One of my 'labs' was on nonlinear acoustics, specifically ultrasound applied for therapeutic uses. It was very captivating seeing a very focused point within a block of gel become ablated. A part I found particularly exciting was realising that it was a phased array of ultrasonic emitters, so that the point where the ablation occurred could in fact be placed anywhere you desired in the gel.
They showed us results of HIFU applied to real patients to non-invasively ablate tumours and treat prostate issues. As far as I can tell the probe creating the ultrasonic waves needs to be relatively close.
A thought I had at the time was if you knew all of the material properties of all of the tissues inside someone and their locations (say with an MRI) you could in theory apply this even deeper in someone than is currently possible - with a larger stick-on patch of actuators as a phased array.
Finally, another memorable thing that was discussed was what another researcher was doing with ultrasonics. Stride (who I am delighted to say was a fantastic lecturer) was very interested in bubbles. She would construct tiny bubbles where the surface (or interior?) was made of a chemotherapy drug. These bubbles could then be injected into someone's blood stream and would be ruptured using ultrasound to allow for extremely targeted application of chemotherapy (the jet formed from rupture would be so strong it would inject the drug into nearby tissue).
Fascinating, fascinating stuff but of course developed over many years of hard work.
> "Cancer is awful," Xu says. "What's making it even worse is cancer treatment."
Well said. And it's either terrible or expensive (and sometimes also terrible as well).
Proton therapy for instance is amazing at targeting hard to reach tumors like those in the eye, but costs close to fix figures as it requires a team of people to design the treatment.
For comparison, a liver histotripsy costs $17.5k:
https://histosonics.com/news/histosonics-notches-significant...
Not a bad deal for a non-invasive life-saving surgery.
At the intersection of ultrasound and startups (since this is HN), does anyone have any thoughts about that Openwater project? They are apparently working on open source ultrasonic medical devices.
I don’t actually know much about them, I just heard of them because their CEO (Mary Lou Jepsen, she’s quite famous, right?) was on the AMC podcast (months ago, actually, I was just going randomly though the back catalogue).
Tech folks pivoting to medical always throws off some alarm bells to me, but she was fairly compelling on the podcast and the basic idea seemed to make sense. Ultrasonic treatments, using diagnostic-level energies, using focusing and resonance based tricks, I guess. (It is way outside my wheelhouse, sorry if the description is inaccurate).
> Tech folks pivoting to medical always throws off some alarm bells to me, but she was fairly compelling on the podcast and the basic idea seemed to make sense.
The best way to evaluate biotech startups from the outside is to look at their investors. If they’re full of VCs specializing in biotech, chances are someone did the bare minimum due diligence on the science.
Theranos for example didn’t have a single one because biotech VCs steered clear of that mess entirely.
> Tech folks pivoting to medical always throws off some alarm bells to me
Same for me. I've been in the medical device industry for 15+ years now and came from "tech". What a lot of techies under/don't appreciate is that the medical device industry is heavily regulated and moves at a muuuch slower pace than other technologies.
There are lots of regulatory and quality/testing hurdles that you must clear (namely verification and validation testing, in addition to your 510(k) clearance or approval, if PMA) before you can market and sell your device.
I tell customers, on average, a Class II medical device project can take 18-24 months and cost $3M to 4M, minimum.
Yeah, it seems that their pitch is that they want to move at consumer electronics speed, I mean, their website explicitly says
“Our tech-driven approach leverages software, hardware and AI […]
That means we can iterate at the speed of consumer electronics”
Which is kind of scary but also a bit interesting.
How would you go about regulating an open source medical device? The user can just plop whatever software on there that they want, and ultrasound themselves wherever… play with resonance and focusing, right?
She has a couple of TED talks on this tech from several years ago.
I was aware of her from the OLPC project and the cool Pixel Qi screen tech from that, but haven't watched the talks.
There was a game in the early '80s called Microsurgeon where you piloted a robot probe into a body to cure it of diseases. It was armed with an array of tools, one of which was ultrasound that you could use to destroy cancer. I wonder how long this idea has been around for.
https://en.wikipedia.org/wiki/Microsurgeon_(video_game)
This can also be used for prostate, it's nothing new. But you cannot use this anywhere where the ultrasound would be blocked by other organs.
Fun fact: using this ultrasound for prostate cancer treatment reduces the risk of erectile disfunction
The article mentions that this is a different type of ultrasound treatment than the one that has been in use for prostate cancer treatment for some time.
>But you cannot use this anywhere where the ultrasound would be blocked by other organs.
Yes you can. If you had an array of ultrasonic transducers around the body you could have each of them in phase targeting a single spot. Beamforming is a thing we've been doing for years with RF. It's even more trivial with sound.
We were privy to a lab that accidentally cooked mice with gold nanoparticles in the late 90s with multiple IR lasers. After they figured the power side, it turns out that gold nanoparticles are wildly cytotoxic on a number of axes.
https://house.fandom.com/wiki/Clueless
How is it more trivial with sound? Sound is just a wave just like ultrasound. In fact, ultrasound has the word sound in it making it sound. So your conclusion is not sound.
IN fact, they do this today to break up kidney stones. Multiple beams.
Can’t the surgery be then with a small probe just to get the ultra sound tip near the cancer? I don’t know the size of the ultrasound tip but seems to me it can be smaller then a hand or tweezers.
Often constructive and destructive interference of waves can be used to focus the ultrasound through tissue without any incisions at all. Kidney stones are sometimes broken up this way.
See currentsurgical.com
> Fun fact: using this ultrasound for prostate cancer treatment reduces the risk of erectile disfunction
I’m not aware of strong evidence in this area (not saying you’re incorrect).
For the liver indications, several elite radiology departments have had very poor outcomes with their patients, despite the strong public data. I would not, with my own prostate, try a new technology until at least a decade out, at least.
Urinary and erectile function are a major issue with partial and radical prostatectomy. These ultrasound treatments are showing significant improvements in both areas.
This technology is also now used to treat non-cancerous prostate enlargement (BPH).
> Urinary and erectile function are a major issue with partial and radical prostatectomy
There are other options besides prostatectomy or the untested histotripsy.
True, but currently prostatectomy is the most common intervention and second is radiotherapy which as mentioned in a comment above is a very expensive alternative with known side effects.
Histotripsy is early in its clinical life but I wouldn't say untested.
Here is a trial (2022): https://pubmed.ncbi.nlm.nih.gov/35714666/
And a review: https://pubmed.ncbi.nlm.nih.gov/36686753/
I remember seeing a demo from people who could slap a raw steak into one of these machines, and with ultrasound, sear their logo into the meat at sub mm precision. But that was long ago & not ready for medical usage yet. Cool that it seems to be used for actually treating people now.
I just spoke to a oncologist surgeon about this. Even though their facially doesn't have one (they are expensive machines) he said it's looking like it'll be standard care in the future.
He did an evaluation about getting one for my local hospital.
You can get a ultrasonic fat cavitation machine off Ali Express for a few hundred bucks. The technology has gotten surprisingly cheap.
Don't they break up kidney stones using ultrasound as well? Or is that a different type of "ultrasound"?
“Lithotripsy” is the name of the kidney stone treatment. My understanding is it’s based on vibration, not ultrasound (I know, vibration is sound - my understanding is the method on the linked article uses higher frequency + intensity + shorter pulses than the kidney stone method - so sorta like microwaving tumors vs using a massage gun on kidney stones?)
I think parent is thinking of "Ultrasonic lithotripsy" which does use ultrasound.
I’ve had it, it’s ultrasound but it’s not always effective against hard stones.
I think that's traditionally done with lasers.
Having had kidney stones, they're both used. I think for the sonic one they put you in a water bath because it conducts better. But as I understand it, the docs can pick whichever one is more optimal, be it shattering the stone sonically or zapping it with a laser.
That’s transurethral lithotripsy.
AFIKR two facilities do this kind of treatment. One in Canada and one in China. There already was a HN threads with some reporting to have been treated in Canada.
https://news.ycombinator.com/item?id=31630679
Apparently, only some tumors have a distinct and unique shape / size. The “trick” is to calibrate the resonance exactly to the size of the cancer cell. So that resonance would “hurt” only that kind of shape / size cell. Which was much harder to do than it sounds. Sadly not all cancer cells are unique and not that “easily” distinguishable by size
But I am not in the medical field and just repeating what I’ve read.
It's amazing how we're turning sound waves into healing tools.
Once you've destroyed the tumor, how do you get it out of the body?
https://www.statnews.com/2024/06/22/hank-green-pissing-out-c...
Watching Hank Green's YouTube video where he found out that his cloudy pee was cancer leaving his body, he was surprised that doctors don't tell you to expect it. It can be such a morale boost.
The recycling of dead cells is a normal biological process the same thing happens when they use radiation to kill cancer cells
If I interpret the article correctly, the ultrasound energy does two things: it effectively destroys the cancer cells by overheating them, and it physically breaks apart the tumour. Your immune system can further break up and get rid of dead cells the way it deals with normal dead cells.
Won't there still be some broken up live cells that can now migrate around the body and cause cancer in other areas?
This was an issue with uterine morcellation; https://en.wikipedia.org/wiki/Morcellator. Per the article, though:
> Some researchers have raised concerns about histotripsy potentially seeding new cancer growths as tumours are broken up inside the body, meaning they can be transported to other areas. That fear, however, hasn't borne out in animal studies so far.
Usually, it suffices to initiate apoptosis, the self-destruction mechanisms of the cells.
I doubt ultrasound would trigger apoptosis in cancer cells, one of the reasons they're cancerous is that they refuse to commit suicide when they should.
It heats them until enough damage is done that they die regardless.
So more like necrosis, not apoptosis. Maybe non-biologists are not aware, but apoptosis is not just cell death.
Or just tears them apart in the case of Histosonics.
Also very very interesting for brain modulation!
It's a highly promising direction for many diseases, I specifically remember Alzheimer's as one
https://www.fusfoundation.org/diseases-and-conditions/
I hope for a great future of this therapy.
I really hope she didn't damage her (or her colleague's) hearing while doing these experiments!
The only thing the article fails to mention is the use of more than one transducer used to focus multiple ultrasound beams to an intersection point in the body, increasing the heating power of all beams
There was a startup in Shanghai in the early 2000. Their device used multiple transducers. The probe was at least 40 cm in diameter. They did trials on uterine fibroids, among other diseases. One of the difficulties was while it looks good in theory, but the path ultrasound travels in the body is more complicated than, say x-ray or gamma ray. They expected a fine focal zone, but sometimes the focal zone was much larger than expected. This new wave of ultrasound equipment may have discovered better ways to control the sound beam.
How's progress on individualized cancer remedies based on mRNA?
Don't know about mRNA but individualized remedies based on CAR-T technology have been making significant strides in this area, with major commercialisation expected in the next 1-2 years