I was challenged to make a lens for a friend without buying any new optical elements. I have a pile of glass that I've salvaged from old camera lenses that I've modified. I wanted to make a cooke triplet but didn't have any negative elements on hand where I precisely knew their characteristics. I then realized that my eye glasses are relatively well characterized negative lenses, so I had an optometrist cut them into a disc shape and I built the lens around that. It performed very well in my simulations, but not so well in real life. But it was a fun project and my friend ended up with a unique lens for her camera.
I swear the only thing that the Bolte Bridge attracts more of than seagulls is photographers.
Also the desktop I have now for ML work is powerful enough to do a full 3d simulation of a lens using the full Maxwell equations from first principles. I remember doing the back of the envelope calculations that I'd need the Bluegene/L back in my undergrad days to do it for real, well: https://bnnbreaking.com/arts/video-gaming/nvidia-geforce-rtx...
> I swear the only thing that the Bolte Bridge attracts more of than seagulls is photographers.
Thanks for encouraging me to search about it, and quickly realize Bolte Bridge is a natural target due to relative level of architectural sophistication as well as being easily accessible.
Yes, OP was doing linear ray tracing to see how the lens would perform on the video. I remember doing the same in a computational physics class way back in the day and talking to the lecturer about what you'd need to simulate real optics without approximations in the equations.
The answer was the worlds fastest super computer at the time. I was a bit shocked that he claimed that was the state of the art for lens manufacture in industry too - no idea how true that was. But figured that if you needed that much computation it made sense.
Well I have 2x that under my desk now and I use it to do local development before I push to cloud machines with 100x the power where the actual work happens.
All it takes these days is a piece of aluminum foil and a needle. Tape the foil in front of your sensor (without light leaks), poke a small hole into the center of the resulting spot with the needle.
Takes high ISO settings to get something out of that, but it works and is dead simple.
My dad is giving photography lessons, and one of the constants are people realizing during session one that they, in fact, do need glasses when the autofocus does not agree with their eyes.
When I borrow my dads camera for the odd shot at times, I totally relly on AF, if I went by the view finder I am almost blind... Same thing the other way round!
As someone who has just moved back to Australia after ~20 years in the USA, I was surprised to notice how many people reverse park here. It seems to be an ego thing , particularly when I notice how many people do it at the supermarket, where it has to be super inconvenient.
Muscle movements, think talking, chewing, or blinking, generate voltages orders of magnitude higher on the scalp than sources from neural activity inside the skull.
I'm the CTO at Seer Medical, we make a wearable EEG device that's designed to be worn for ~10 days, including during sleep.
Totally agree that the electronics is not the difficult part (battery management, and power consumption aside). Our most loved features are the industrial design and materials science around our electrodes and ways of attaching them to the scalp.
We've designed and built our own electrodes too. Though we don't have to do multi-day, and we're looking for a specific signal during sleep, so probably not as challenging as what you guys are dealing with.
I still think though, that in a common setup the +-3mm is impossible to reach. People don’t sit perfectly still, may flex their muscles during the measurement, etc
I'm the CTO at a relatively well established medical device company (EEG/ECG). Before I arrived at my current company I spent a year of my own time tinkering in the BP space. The article, while relatively, light on details gives a pretty good sense of the problem with trying to replicate the traditional BP measurements using the optical methods used by smartwatches.
Smartwatches and similar devices use photoplethysmography (PPG) to measure blood oxygen saturation. The heme units in blood have different reflectance/absorbance spectra depending on their level of oxygen saturation. The only difference between the fingertip pulse oxygen sensors used in clinical settings and those used in smartphones is that the clinical ones use transmission, whereas the watches use reflection. While the methods are different, they arrive at very similar numbers (+/- skin color & racial biases).
Unlike pulse oximeters of yore, smart watches can take many thousands of readings per second. And if you offset your photodiodes by a known distance, you can calculate pulse transit time, and a variety of other measures beyond just your spo2.
The challenge has been in translating these measures back into systolic and diastolic blood pressure. To be accepted by the FDA you need to be able to show that your device is +/- 3mmHg. That's just not doable right now for the reasons mentioned in the article.
That said, the measures that you can get from high-speed PPG are probably not useless. But these measures aren't backed up ~200 years of medical practice. It is a very long road to introduce new measures into the medical system. Not only do you need to get FDA clearance to sell your device, but you need to navigate the reimbursement system to find someone who will pay for your device. Doing this in a 5 year time span would be unheard of. While cardiologists would like a better system than cuff based BP measurement, the upside of a better system is limited by the current medical system. And there's a significant cost and downside risk to anyone pushing for something different.
One company that I spoke to and is still working on the problem comes with an impressive biomedical pedigree, and they're starting in the non-diagnostic "wellness" space.
Another company is using mm-wave radar to measure pulse waves in a different way, but it appears that they've pivoted from being directly challenging the BP status quo into now selling their hardware sensors to more well established medical device companies.
Long story short, I'd be impressed to see blood pressure cuffs retired in my lifetime.
