Hacker News new | past | comments | ask | show | jobs | submit login
A woman’s brain stimulator stopped working after her home struck by lightning (nytimes.com)
75 points by dnetesn on May 6, 2018 | hide | past | favorite | 39 comments



I used to design these devices (deep brain stimulators).

As part of the design process for every FDA-regulated product, you are obliged to perform a risk analysis to judge the severity and probability of risks to a patient using your product.

One of the risks was “Exposed to a strong current that could destroy/disable the onboard electronics”. The design mitigation was a fail safe shutdown.


Sounds like it worked well:

"Fortunately, when he and his colleagues checked the woman’s implant, they found that it had not been damaged. A safety function designed by the manufacturer had taken its cue from the sudden current running through the house and caused the device to switch itself off, pre-empting any interference from the lightning. The programming was intact; the battery still had plenty of charge.

When they fired up the device, it worked perfectly, none the worse for the storm."


Seems like bad UX and/or user instruction, really; they should presumably have the device warn of this happening, and tell end users how to turn it back on. For a critical health device like this, having to visit the doctor to be told “have you tried turning it on and off again” doesn’t seem ideal. What if it had been a pacemaker or something?


It's very hard to design a device that can reliably tell that it has not been damaged in a way that hurts the user.

We have this with regular socket power too; if live and ground are shorted, connecting the device will shut off power to prevent the user from any harm due to a charged casing. There is no feedback other than "everything stopped working" because the GFI cannot tell if the situation is cleared without putting up power and potentially harming the user.

Think of the other scenario, the stimulator gets damaged, switches back on and kills the user. Pacemaker could do the exact same, if it's damaged it could kill the user very easily.


> Think of the other scenario, the stimulator gets damaged, switches back on and kills the user. Pacemaker could do the exact same, if it's damaged it could kill the user very easily.

I was talking to a guy who's implanted defibrillator failed a while back.

It shorted out and started shocking him as fast as the capacitor(?) could recharge. It was shocking him, his nephew, the EMTs and ER staff until they finally got the thing shut off. His doctor was saying he's really, really lucky to be alive.

So, yeah, fail safe is good.


"If an ICD hasn't been turned off and starts to fire, holding a magnet right over the device will temporarily disable it and keep it from delivering multiple shocks. Some hospice programs give magnets to families who are taking care of a hospice patient with an ICD at home."

http://articles.chicagotribune.com/2012-03-28/health/sc-heal...


Once again the auto industry is ahead of someone. No time to explain the safety built in to an airbag system. It's mind numbing.


That's because of the number of death per day in the auto industry. They have a lot of free guinea pigs to experiment with.


Well, assuming you don't have the defective Takata airbags.


The ethical solution to human life critical systems is to make them 100% open. The user should have complete access, even to things we suspect are not necessary, like masks themselves. Our courts can handle this. Demanding it once you need it wont work.


First it was not a critical life saving device so shut down was a completely viable option. Second, ideally you don't want people to have any interaction with these devices. If they can restart it they can probably mess with it in other less useful ways.

Really you want something like this to be as simple as possible without harming the patient.


Instructions for Use indicate that in such an event you should immediately go see your physician.

A pacer is different than a brain stimulator. And pacers generally try to restart themselves if they can. Brain stimulators, because of the sensitivity of where you’re stimulating (a few cm, literally, away from regions that control breathing, heart rate, etc), the least risky option is a shut down.


> they should presumably have the device warn of this happening

How? Shock your brain? It's an implant.

> What if it had been a pacemaker or something?

The same question stands. How do you make the device provide reasonable feedback? Make your heart skip a few beats? How exactly is a pacemaker going to warn a user about an error condition?

Given the circumstances, and the rarity of the error case, it sounds like the device switching itself off is the best course of action.


I can speak to this a bit. Some pacers can be programmed to beep in certain situations to indicate things like, low battery. Alternatively you can set some models up so that when a magnet is applied, different programmable actions occur such as, changing the heart beat rythm to indicate what the expected remaining battery life is. There are also "at home" base stations which upload stats to the manufacturer who can call the patient (through the base station) when there are unexpected problems.


>changing the heart beat rythm to indicate what the expected remaining battery life is.

Wow, now that is an interesting user interface.


There’s an external charger, per the article; it could presumably feed back to that when charging.


Connecting your brain implant to the charger after it unexpectedly shut off due to lightning seems like a good way to earn a Darwin award.


If the implant is shut off for safety reasons related to electrical issues, the last thing I'd want interacting with it in any way is the charger. Hell, we can't even get Macbooks that don't fry USB devices.


> Hell, we can't even get Macbooks that don't fry USB devices.

Not sure why you'd assume the mainboard and peripheral circuit design on MacBooks would be particularly good, they seem to make the same mistakes for years on end.


If it was obvious that the device stopped working, they made the right choice. The device can not tell with certainty that it wasn't damaged, so going into a specialist is the correct action.


Yep, that’s fair enough, I suppose. What I’d wonder about would be, would it always be obvious?


The title is incorrect it seems. The stimulator detected a fault condition, and turned off. This is what it was designed to do, it was working just fine. it really ought to be something more like:

> A woman's brain stimulator turned off after her home was struck by lightning.


can the title be changed to implant?

I first read it as brain simulator, and was expecting a eulogy for her dead ai


When I read the title, I was expecting it to be a tongue-in-cheek futurist reference to a television.


I read it as "woman's brain simulator". I thought this was some crass misogynist onion article.

Either "a simulator of a woman's mind stopped working after lightning struck and blah blah shoes or something."

Or a "woman who had a simulator for a brain was struck by lightning and now something something shoes something."

The title really should be changed to implant. Even without misreading the word the title isn't great.



Same here. My fault for reading too quickly, but I was seriously intrigued.


> A safety function designed by the manufacturer had taken its cue from the sudden current running through the house and caused the device to switch itself off

The way this is worded seems as if the device knows what current runs through the house wiring, through sensors in the house communicating with the device. Is that the case? Or do they simply mean the electromagnetic field of the lightning caused current in the implant?


The Medtronic device is designed to shut off if it thinks it is delivering enough current to go past a certain charge density in brain tissue (calculated/assumed)


Usually this happens when someone gets too close to a refrigerator; once in a Prius.

That the device survived a lightning strike is pretty awesome. I would MRI her head to make sure that there was no unintended ablation from the extra current through the implanted electrodes, however.

*edit - I guess she was not struck by lightning, but her house was.


An MRI with any metal effected by strong magnets in your body can be deadly. The least of your worries would be tearing the implant out through your skin. My lung doctor saved me from that when I was brought in unconscious in an emergency. My wife tells me he went down the hall at a full run... I have an ICD implanted.


There's specs for current Medtronic DBS systems (machine configured below certain RF power, 1.5 tesla magnet, specific head coils), as long as you stay within them, it's ok. Boston Sci/St Jude ones are "probably" compatible" but has not had FDA approval for MRIs yet.


I always assumed they run MRI patients through a metal detector first, seems like a sensible precaution to me.


They do, at least the one time I was in one (which was fMRI for a research study.)


There are MRI-safe ICDs now. So they can be designed to avoid heating and mechanical displacement in MRIs. For ones not designed for this, such as yours apparently, it can be deadly though.


The nightmare scenario version of this is Michael Chrichton’s The Terminal Man (1972)


If I remember the book correctly, the device didn't malfunction, but was positively reinforcing the brain to cause more seizures by stimulating pleasure producing areas while disrupting the seizure.


Yes — a different and more elaborate scenario, really. I suppose I was just thinking of ways machines embedded in your brain can go wrong.


TFA is not about Hitachi Magic Wand personal massager :(




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: