davycro

That would require a dataset of ultrasounds from people having active myocardial infarctions, which we don’t have, and would take at least a year of academic coordination to assemble.

The current datasets are just labeled anatomy at end systole and diastole.

Awesome, thank you for the reference. That's the essence of finding regional wall motion abornomalities. When looking at the ultrasound I look for parts of the myocardium that don't change in size relative to their neighbors during the cardiac cycle.

Fast.Ai is incredible. Love it. I've bumbled my way through it and was able to make a unet learner trained from this dataset -- https://www.creatis.insa-lyon.fr/Challenge/camus/ -- in order to segment the left ventricle from an apical four chamber view. It's a start, but I still have a long way to go.

Done. Appreciate it!

Yes, however much of that research is for formal ultrasound obtained by a professional sonographer with an expensive machine. I'm interested in bedside ultrasound performed by an emergency physician with a mediocre machine (eg butterfly).

It seems the primary way to detect regional wall motion abnormalities is with speckle tracking, which requires way too much post-processing for a clinician.

A system that segments the left ventricle and finds akinetic regions in realtime from a parasternal long axis view or an apical four chamber view would be pretty nifty.

If you know of a paper or system that does this now then please let me know. I would love for someone else to have solved this, haha.

My email is Davidm.Crockett [at] Utah.edu

I am an emergency physician without any formal software training and for the last three months I’ve been trying to build a program that segments the wall of the heart from an ultrasound video and then identifies regions that aren’t moving (an early sign of heart attack).

There are many similarities between this man’s project and mine. And if I had his knowledge I may have cracked my problem by now and would have new way to detect heart attacks early.

Adidas = Adi Dassler

Good question. Proned patients are too sick to survive at home. As soon as the roll onto their back or sit upright they start suffocating and must work for their breath. Because of this they need help caring for themselves. They also are at risk of becoming much worse and needing more oxygen.

We can put a patient on oxygen without intubating them. High flow oxygen can provide up to 60 liters per minute. We can prone patients, and provide them with nutrition to help support their immune system. We save many patients with supportive care.

We can’t however cure the virus. And if it progresses far enough then no amount of oxygen will save a patient. These cases are hard because we have to watch someone die without any means to save them. Doing cpr on a patient with covid can feel pointless. If the lungs do work then no amount of chest compressions, oxygen, and cardiac drugs will save that person.

Although this post was written for physicians, I think hacker news will still appreciate it

I thought all sushi had to be flash frozen to kill parasites. Is this a myth?

Now I remember why Flash websites were popular in the early 2000s. I am glad css caught up!

Medicine moves fast and it’s impossible to stay up to date on everything. In general I do not incorporate the latest discoveries into my daily practice unless the evidence is absurdly overwhelming, which rarely happens. Usually new therapies have marginal evidence to support them. I prefer to wait until multiple studies support a treatment before it becomes my standard of care, and by that time it’s likely going to be summarized in a Cochrane Review.

Should a patient come in with a preference for a treatment they read about on PubMed (which has never occurred in my practice) then I would likely accommodate their preference unless I thought it would put them at serious harm. A patient knowing more than I do about a particular disease happens more than you’d expect, especially if they have a rare chronic disorder. This does not diminish my ego or confidence in my knowledge base.

Every script I write is for a generic. I have never met a drug rep.

For many ailments there are five or six different drugs suited for the job. Goodrx helps me pick the most affordable.

I don’t know how goodRx makes money, but it’s a great tool for doctors. I use it every shift to pick affordable antibiotics, blood thinners, eye drops, etc for my patients. Yesterday I accidentally prescribed an ear drop that cost $320 for the bottle, when there was a $15 option that was just as effective. Only way for me to know that was GoodRx

The article extrapolates leadership traits from extreme and rare emergencies. In my experience these qualities work when error or poor team work will lead to death or harm of another person (say combat, trauma surgery, or aviation). They don’t apply as well to business or politics.

I encourage patients to read or listen to me dictate my note about them during our exam. Often they catch something that I had misinterpreted.

Notes intended for patients would have a different form than notes intended for doctors. Unfortunately medical records achieve neither. My notes are bloated with information needed to bill the maximum amount from the center of Medicare services, which ultimately hurts all parties.

The bloat obscures critical information from other doctors. Patients are unable to read a meaningful account of their care, and are charged more for worse services.

I enjoy easy page-turner sci/fi or fantasy because it helps me unwind after a stressful day at work. Some of my favorites this year:

- Vengeful and Vicious by VE Schwab - Collapsing Empire by Scalazi (late 2017, but close enough to 2018)

Not released in 2018, but still fun and new to me this year. - The Red Rising series by Pierce Brown

> we can’t afford downtime

Indeed. We joke in our emergency department that patient’s die during Epic downtime. We aren’t really joking.

Spot on. Epic isn’t designed for patients or clinicians. It’s designed for up-coding patient encounters, that is, to maximize the amount of money a hospital bills insurance or Medicare.

When a hospital switches to Epic, they do so because they know it will result in more money billed per patient. This justifies the outrageous price.

Those records are protected by HIPAA which requires they be encrypted, and epic logs anytime a person views patient information.

I went into medical school with the same goal. Intended to do an informatics fellowship after residency. After witnessing the Epic rollout at my hospital, I want no part in EMR development. The beuracracy is disheartening.

I’ve found other places in medicine to make software and have an impact.

That said, I truly hope you succeed. We need more hackers in medicine.

I needed medical records from a hospital across town for a patient on my shift last night. Unit clerk called the hospital to request records. They then printed about 60 pages from their electronic record. Those pages were faxed to our department. Our unit clerk scanned the printed pages from our fax machine into our electronic record.

Some patient notes are a scan of a fax of a scan of a fax.

Am doctor in an overwhelmed county hospital in the South. We are so overcapacity that most patients spend 12+ hours in a hallway waiting to be roomed. We have every motivation to discharge patients fast.

Hospitals are dangerous. You lose your sense of autonomy, are woken up throughout the night for blood draws or medication management, and are at risk for nomosocomial infection or delirium.

It’s an emotionally charged place and conflict between families and doctors happens often. Hard ethical questions happen. For instance a young man may come into the ED with massive polytrauma and require immediate blood transfusions to keep him alive. There’s a chance he would be the person to refuse a transfusion on religious grounds, and the family would be upset with our team for ‘violating his body’.

Because of situations like this, or the one described in the cnn article, ethic committees are an essential part of hospital care.

tl;dr Ultrasound video editor made with ReactJS for ER doctors. Download this sample ultrasound and try it out!

https://s3.us-east-2.amazonaws.com/sonogif/q0.mp4

Details:

I am an emergency physician. My peers and I love to share interesting ultrasounds with each other on twitter. We do this for medical education, and to see great cases.

Some examples -- https://twitter.com/buckeye_sanjay/status/963104781815484416 https://twitter.com/EMNSpeedofSound/status/96131521371385856...

One problem we face is that ultrasound videos contain protected patient information, such as their medical record number or date of birth. Because of this videos must be edited with desktop software prior to sharing.

In the past couple of months I started to teach myself ReactJS and thought it would be fun to attempt to create an in browser video editor that we docs could use to anonymize ultrasound clips. The end result is www.sonogif.com.

The app uses the canvas object overlayed on top of a video tag. To render, react creates a local .png file for each video frame. The png files are uploaded to an express server, which converts them into an mp4 file and gif file.

This keeps patient information local and prevents it from every being transmitted to my server.

Anyway, download this ultrasound file and try it out yourself!

https://s3.us-east-2.amazonaws.com/sonogif/q0.mp4

Different concept. Anesthesia for an operation leads to better outcome and more comfortable for the patient. Sucicide was common in people who needed a surgery prior to the advent of anesthesia.

1000/1000 for $60 in Chattanooga, TN. Love muni internet

Most of the municipalities listed in this table are located in the Southeast Appalachia area. Does anyone know why community owned fiber is biased towards this part of America?

https://cdn.arstechnica.net/wp-content/uploads/2018/01/muni-...

Correct. We do a non-contrast head ct and ct angiogram of the head/neck before giving tPA. In this case we may have found the carotid dissection on the ct angiogram.