For circumstances like this I create a VM with a version of MacOS that the app works with, and keep it on life support that way (doing this at the moment for Finale, music notation software that works well with Piano Marvel and thus gamifies piano practice for my kids…)
A counterpoint on focused ultrasound for essential tremor - just because you don’t penetrate the brain with a physical object doesn’t mean you don’t have side effects - as usual, benefits come with risks
> Tremor significantly improved in all patients. Seven experienced mild adverse effects, including 2 with transient gait impairment and a fall, 1 with dysarthria and dysphagia, and 1 with mild dysphagia persisting at 3 months.
I own the 1st gen of the Butterfly - in my opinion, it wasn't great image-wise compared with the contemporary conventional crystal based probes (thinking of cart-based machines with less flexible, more expensive probes etc so perhaps an unfair comparison). Would be cool if the newest ones mentioned in the article are becoming comparable with the crystal based probes - I can't comment. But I can say image quality is absolutely key. There are lots of cool AI based applications coming out all the time (I know much more about echocardiography AI than the foetal ultrasound AI mentioned in the article, but this is a similar paper where some ultrasound novices had AI guidance and were able to obtain useful echo images https://www.ahajournals.org/doi/10.1161/CIRCIMAGING.123.0155...). I get to use various machine vision based tools on echo images at work to automate various measurements - but at the moment, I find they fail badly if the imaging is anything but great quality, whereas humans can interpret them. Maybe future training sets will include more "technically difficult studies" (code for poor imaging) and AI tools will do better than they do now? Or there will be more augmentation of data sets with realistically degraded versions of images to add robustness? AI that worked on suboptimal images would be awesome, particularly in my setting (ICU).
Medical imaging is not about getting the expected result as is typical in most bodies, rather, it is about getting the actual result as is found in this body.
The applications I've seen are around analysis/alerting. Submit your raw data to a service and it screens for hundreds of rare conditions your doctor's never heard of.
Similar to blood work maybe nothing comes back conclusive but it might be interesting enough to trigger follow on investigation.
But the general theme is around augmenting the human decision maker rather than "preprocessing" the data in a way that might obscure or hallucinate important details.
False positives on tests are a well understood source of illness and iatrogenic harm, as well as being a well known money printer for those administering the tests. Trying to screen for a whack of rare diseases with AI still sounds dubious to me unless you have good reason to suspect one beyond high levels of medical anxiety.
I’m more interested in AIs ability to do things like read x-rays and identify illnesses with higher accuracy than a tired doctor on a 12 hour shift.
Sure, there are many applications and more coming. In your example of an actual result as is found in this body, the image segmentation [0] and recognition task is applicable.
One of the bullet points in the above is "Intra-surgery navigation." Let's say there is a known position of a tumor within an organ based on previous imaging. The human body is squishy and interior organs may shift in position or shape when a procedure is performed. Deformable image registration [1] guides the surgeon (human or robot) to the tumor location with greater precision than what was possible before.
Read this on mobile and the identifier longWindedNameThatYouCantReallyReadBTWProgrammersDoThatALotToo overflowed into the margins - I regard this not as a bug but a feature which helped make the author’s point :-)
We use this technique as a guide in our company. If someone (knowledgable) would ask "What does this method call do?" and the method name does not answer that, your PR doesn't go in the master.
E.g. getString(path) for loadConnectionStringFromDisk(configFilePath), tryConnect(30) for testSqlConnection(timeoutInSec), even the reader now knows what happens here and what input is expected.
Not quite the same but when we were rebuilding during COVID we knew we wanted two good WFH spaces but really only had room for one (tiny) office - as our second we built a shelf in our bedroom wardrobe that lifts up to reveal a work surface and a 32" monitor - inspired by this guy's work on ikeahackers:
Works really well and given the cost of space for a second desk it was a very cost effective solution - only minimally affects the utility of the wardrobe as a wardrobe. (We didn't build the laptop holder this guy describes, the laptop just gets sat on top of the clothes in the drawer below that shelf with attention to make sure we're not blocking a fan intake. We had a powerpoint and a ethernet port put into the back of the wardrobe but actually we just end up using Wi-Fi mostly.)
Nice! I think we’re going to turn the closet office back into a closet, add built in shelves, but make sure one of the shelves is at desk level and that others are removable. That way we still have a backup desk.
I'm a doctor (ICU) who also does some IT stuff for work (nothing complex, mainly writing small web apps). I really enjoy writing stuff where you know the requirements, make all the decisions, write the code and maintain it. However, doing this has allowed me to get some exposure to what bigger projects with lots of moving parts, data sources, stakeholders, regulatory requirements etc look like - and it's seriously hard work. Nothing like "coding, the hobby".
I guess this commonly occurs in many fields at a certain level of seniority - the "managing a large system involving many people" aspect can dominate the domain-specific part, be it software engineering, accounting, manufacturing etc. As such I'm really glad I chose medicine rather than SWE (even though I've been writing and loving code for >35 years, and it was a real toss-up when I went to uni) because:
1. You can still stay very hands on, even as a senior clinician, especially procedurally.
2. If you so choose, there's a lot of variety in what you find yourself doing as a doctor (my mix looks like making clinical decisions / talking to patients / families / doing procedures / performing and interpreting ultrasound / going to other hospitals to retrieve super sick patients and bringing them back in ambulances / mentoring / teaching / coding / managing a clinical service / etc - but there are lots of other options too). I'm not sure if this kind of variety is as easy to arrange as a SWE? (though I suspect I'm about to be corrected, thanks in advance.) Variety is quite important if you're easily bored, which is a common problem for bright people.
3. Although AI is coming to all fields, I do think the impact will look more like "better tools", rather than "job replacement", or "vast reduction in number of people needed", for longer in medicine (at least in my area). As a breadwinner this is a not inconsequential consideration.
Hope you find the career you love, and that it leverages the work and study you've already done in some way.
I can relate; I chose Wall Street (the finance side, not the IT side) and now work for myself. While I used and use my tech skills every day, I have never wanted to write code for money. <https://news.ycombinator.com/item?id=36027171>
A 27 or 30G insulin needle with some lidocaine (still often called lignocaine here in Australia ) is a nice addition to this procedure. I rarely omit it unless the patient is unconscious. If you inject the local anaesthetic under real time ultrasound, it can also serve as a seeker so you see if you’re off target with the 30G needle and adjust based on that information, which improves your success with the larger (commonly 22G) needle used to take the gas. Not unreasonable to request local anaesthetic, really - it is known to be a painful procedure.
For really small kids we tend to use a transilluminator rather than ultrasound for peripheral IVs (though it doesn’t tell you the depth like ultrasound does). For CVCs or bigger people, ultrasound is often very useful. Like all things, experience makes a big difference! In my opinion, for known tricky veins (in adults at least) it’s best to use local anaesthetic from the outset - it does make it slightly harder but if (when?) you miss it’s easier to keep the patient on side for the next attempt.
Edit: YouTube link to a transilluminator being used - beware the video does have an unhappy child in it
