August 02 2022, 10:06

A bit about pharmacies in the USA. Mainly prescription drugs, prescribed by a doctor. The doctor sends the prescription electronically to your pharmacy, and you just go there and redeem it by showing your ID. I periodically get a cough, almost every year, and I just in case do an X-ray and see a doctor. Thank God, it’s all clear, but the doctor prescribes some medications because something needs to be prescribed. Thankfully, not antibiotics – that’s generally the first thing they do. This time the doctor prescribed a mouth spray, which turned out to be expensive at the pharmacy ($320, Budesonide/Formoterol). It could have been ignored if it had been approved by insurance, but it wasn’t, and I was asked to pay out of pocket. When I asked the doctor to prescribe something else, he prescribed the same budesonide/formoterol, but in some other form, and although the spray again cost over $300, this time the insurance paid two-thirds. I paid $100, which was unpleasant to do, but I decided to try it.

You can also just buy in the pharmacy. In the cough section, there are about 40 different medicines. But they all are a combination of guaifenesin (secretolytic, expectorant; also known as a substance to keep horses under anesthesia 😉) and dextromethorphan, which is essentially not a medicine but a symptom-removing agent (effectively, a cough center blocker in the brain). And that’s it, nothing more. A huge stand only of bottles and pills, where one of these two active substances is presented alone or both in various proportions. Everything else — only through a doctor.

In Russia, for example, Bromhexine (expectorant) is popular. In the US, it has no FDA approval. The FDA also did not approve Ambroxol. I tried to find acetylcysteine, it’s a mucolytic too. And there’s a problem — it was only available as “vitamins,” but now even that is not available (


preview

NatMed Pro – N-Acetyl Cysteine: Supplement, Drug, or Both?
https://naturalmedicines.therapeuticresearch.com/news/news-items/2021/january/n-acetyl-cysteine-supplement,-drug,-or-both.aspx

July 27 2022, 02:01

Got access to DALL-E 2.

It’s astounding, really.

The first two images were generated from the phrase “mother with daughter holding an open red umbrella on a horse on a rainy night in the style of a painting”. Cool, huh!

Another two images were generated from the phrase “The painting depicts a scene with watches melting slowly on rocks and the branch of a tree, with the ocean as a backdrop”. I simply described a Salvador Dali painting, and DALL-E created its own version.

I also asked DALL-E to draw a picture based on the phrase “an impressionist oil painting of a female figure skater dancing with her male partner, with the mountains and lakes as a backdrop”, as Masha is currently skating in Lake Placid.

It’s rare that computer stuff surprises me, because I generally understand how they work. I kind of understand here too, but just barely, with great difficulty, and the result is much, much better than the wildest expectations.

I think that soon there will be specialists emerging who craft requirements for such devices in the form of text. They will manage to get higher quality images through their formulations

July 20 2022, 10:12

I wonder why all voice-to-text programs do not try to identify the topic and load terms specific to that topic? Here’s a conversation about, say, horses. Load a dictionary with terms, horse brands, typical names, names of racetracks, or whatever else, and run recognition again, using terms from this dictionary with more weight than terms from, say, IT or cooking. Understanding that it’s about horses from the text AI can do for a long time. But adapting, it cannot.

Or consider recognition in Teams. Microsoft, you have gigabytes of messages from chats and groups. It’s quite obvious that roughly the same words will be heard in the audio. Why not compile a dictionary of such words and load it into the voice recognition system to make transcriptions more accurate? I’m not even talking about the fact that the same person writes and speaks generally about the same topic. If we take his messages for the dictionary and apply this to the statements he makes, and secondarily to those who are in the call, it would be just perfect.

One could think about how to improve existing recognitions. Like, over a week accumulate knowledge of how, say, Medik8 sounds and is spelled (from chats), and then change all incorrectly recognized ‘medicate’ to Medik8 in past meetings (including updating the search functionality to reflect the changes). Understanding that it is wrongly recognized is nontrivial for a machine, but still possible, since the word medicate clearly does not fit grammatically, unlike Medik8.

A proper startup needs to emerge that integrates with messengers and meeting apps, doing all this smartly, while charging some money. If all internal meetings were transcribed (properly! with replicas, names, taking into account the topics) and there was a unified search considering access rights (you can only search meetings you are invited to), it would be a supertool.