#Buy CAD Online

[dandyshucks]

posting here bc theres less ppl following me on this blog compared to main FDJSKL but it is thundering (and therefore lightning is striking as well) currently and we've been in an intense heatwave and there's evac alerts and orders all around the area and i've just kind of been like 🧍‍♂️ for the past couple weeks trying to be chill about it bc this is the fourth or fifth year in a row of this so i am somewhat used to it but also aough :( also now i have an expensive drawing tablet that i would not be able to buy a new one of so i am trying to figure out if i should include it in my grab-n-go evac bag... its so big though and would take up so much room but RRRGHHH I DUNNO i hate these sorts of decisions 😭😭😭

[colicoffs]

Buy Cad 20 Bills Online - Coli Coffs

You can now easily buy CAD 20 Bills online. Visit the Coli Coffs website and choose from a variety of collectible notes. Best deals and reliable services. Buy today!

[buttercuparry]

For more than a year this has been the news; everytime I am to write a post for a family, I go online and this is all I get to see. Nothing has changed in all this time. The occupation forces still bomb schools without any repercussions- they murder hundreds of people-they lay down seige-they allow no aid to pass through and terrified Gazans remain at a loss as to what to do and how to survive.

So nothing changed and nothing got better, rather things are worse now than they were in October 2023. As it stands, the whole of Gaza strip has been bombed multiple times, causing people to be displaced over and over again and lose precious resources like clothes and shoes. I cannot stress how important these resources are when children have to live out in the open sand and when people have to travel long miles to either get a connection or food. My friend Siraj Abudayeh ( @siraj2024z ) has been displaced SEVEN times and 2 months ago his extended family was displaced when the occupation forces closed in on their camp. With the onslaught of winter Siraj has to provide all of his family members with winter clothes and shoes that they have gradually lost when they were forced to flee. He has to provide for TWENTY FOUR people. Currently your donations have helped Siraj to rent a shelter for 10 of his most vulnerable family members but this means that 14 others, including Siraj himself is still living in tents! He is so cold and so are his sons. The nights are terrible and the tent which went through the wear and tear of the weather for the better part of this year, is barely any protection at all.

So please, please donate to Siraj. He wants nothing expensive, he has said that he will buy the cheapest winter clothes that are available in the market. But please we must raise at least 3k for Siraj ASAP. He has received no donations for over 12 hours and is currently stuck at $93,426 CAD

Vetting #219

Once again, please remember that things are going to be much more difficult during this winter of 2024. The hospitals have been bombed all throughout the Gaza strip, making it difficult to seek treatment for even basic ailments. There are barely any shelters left standing where people might be able to retire to, to escape the winter cold and fuel to warm up is a luxury. Also the IOF has routinely attacked camps, after which people were forced to leave behind precious life saving items which were either destroyed or looted. So do not let Siraj down now, when there is an uncertain time ahead.

[globcoff65]

Buy CAD $5 Bills Online - globcoff.com

BUY CAD $5 BILLS online, which are best made from authentic counterfeit notes. We are a well-known company in the field that sells Canadian $5 bills Online. These fictitious Canadian dollars are mostly used to settle bills in a variety of establishments, including cafes, cinemas, and shopping centers.

Authentic counterfeit note currencies are a great choice if you want to buy counterfeit money online because they are quite dependable. They are even compatible with ATMs. But it’s better to be safe than sorry.  Buy CAD $5 bills online.

Features of CAD 50 bills

Intaglio printing

Watermarks

Security thread

See-thru register

Special foil

Iridescent stripe or transferring colors

Best faux one hundred greenback payments, Best online keep to shop for counterfeit one hundred US greenback. Contact us these days for exceptional deals.

We have discreet packaging, and we deliver all programs within forty-eight hours after the order and fee have been confirmed.

Kindly contact us through email or WhatsApp in case you no longer meet an agent online to speak to you. We do our best to provide customers with Grade A counterfeit cash if they want to through skip pen checks, UV, and different counterfeit checks. Buy fake one hundred Canadian Dollar payments.

WhatsApp: +1 (720) 334-7285 Email:info@globcoff.com

[globcoffs]

Buy Chinese Yuan Renminbi Online. We are one of the topmost dealers in supplying counterfeit money online across the world to over millions of customers that are satisfied with our CNY ¥100 for sale. We assure the best quality of fake money online at the cheapest rates.

USA : +1 (720) 334–7285

info@globcoffs.com

[chaoticamelay]

IT'S SO CUTE AHHH and it's even a white cat like alpine 🥹😭

steve rogers made bucky barnes a shirt with a drawing of alpine on it where she's wearing a kippah and it says HAPPY HANUKKAT btw!! i know because i was there

[ofmdrecaps]

11/23-26/2024 Daily OFMD Recap

TLDR; David Jenkins; Rhys Darby; Taika Waititi; Con O'Neill; Nathan Foad; Kristian Nairn; Samson Kayo; Brian Gattas & Connor Barrett; Damien Gerard; Articles; Fan Spotlight: TinyCrewBigDonations; NeverLeftPodcast; OFMD Advent Calendar; Love Notes

= David Jenkins =

David was out on Christmas Eve getting a tattoo in honor of his new baby Zosia and Kinga!

Source: David Jenkins Instagram

Also, in case you haven't seen yet, David has been writing some OFMD Fanfiction, or a Christmas Special since Christmas Eve on his Bluesky. Due to it spanning multiple days, and me not knowing when I'll get the next recap out, I'm putting in as much as I can now.

Source: David Jenkins Bluesky

= Rhys Darby =

Rhys spent the holidays in Aotearoa with the family!

Source: Rosie's IG

Another trailer for Badjelly!

View this post on Instagram

A post shared by Badjelly (@badjellytv)

Source: BadJellyTV

= Taika Waititi =

Taika also happened to be in Aotearoa for the holidays!

Source: VasJMorgan's Instagram / Mariana Vivanco's Instagram

= Con O'Neill =

Con had a beautiful message for everyone <3

Source: Con O'Neill's Instagram

= Nathan Foad =

Nathan's voiceover agents had fun doodling their clients! Can you find Nathan?

Source: Nathan's Instagram Stories

= Kristian Nairn =

Kristian is sending some updates on his new Spectrum playlists, Cameo's and orders from his website! Lots of love coming from our beloved Wee John <3

View this post on Instagram

A post shared by Kristian Nairn (@kristiannairn)

Source: Kristian's Instagram

= Samson Kayo =

Samson got into the holiday spirit and sent a Merry Christmas to all!

Source: Samson's Instagram Stories

= Brian Gattas & Connor Barrett =

Hornberry and Siegfried wish you a silly Happy Holidays!

Source: Brian Gattas' Instagram Stories

= Damien Gerard =

More updates on where Damien is online now!

Instagram / Bsky

And a sweet holiday message from him!

Source: Damien's Bsky / Damien's bsky

== Articles ==

Thank you to our dear friends over at @adoptourcrew for ore articles featuring OFMD!

Source: Adopt Our Crew Bsky

== Fan Spotlight ==

= OFMD Buys Boats =

Last on the list for the Tiny Crew Big Raffle Donations was: World Central Kitchen - David Jenkins - $US 1,779.07, £278.69, €611.91, CAD$82.81! Well done all! I ran out of images in this recap, but the next will feature all the awesome BTS Lindsey Cantrell shared with the @ofmd-buys-boats and the rest of the crew for everyone's generosity!

Source: OFMD Buys Boats Instagram

= Never Left Podcast =

Never Left Podcast's latest episode is about Fan Fic! Check it out on their linktr.ee!

Source: Never Left Instagram

= OFMD Advent Calendar =

The OFMD Advent Calendar has come to an end, and the kind @tillychmo gave everyone some lovely treats for the last two! Including Cameos from the crew! Thank you so much dear for highlighting so many amazing artists and writers this holiday season, and spreading joy and love from our cast! You're the best!

The 24th Door features @tillychmo's artwork! A very cute piece with the full revenge crew and more!

The 25th Door has cameos from some of our awesome Cast! Damien Gerard, Erroll Shand, Ruibo Qian! Wow! What a surprise! Check it out on Bluesky / AO3!

Source: OFMD Advent Calendar

== Love Notes ==

Well Lovelies, we're coming to the end of the year here. As I write this, it's New Year's Eve. I ended up being really sick for several days so I couldn't finish these on time (so sorry for the jump scare, I legit thought I'd be gone for a long while after two docs told me what they thought it was-- but luckily, no surgery for me). I'll try to be getting the last couple of the year out tomorrow if I can <3 I did want to mention, I know the David Jenkins fanfic/OFMD Christmas special is causing a bit of contention out there. Some folks are worried it's S3, and are either upset because it doesn't meet what they were expecting so far, or worried that because it's there, we're not getting an s3 at all. These are all valid concerns! For me, I like to think of what David's doing as a little gift to keep our hope alive. Whatever reason-- whatever comes of it, David cares so much about us that he's keeping us all engaged this late into the year. How many creators can we say do that for their fandom (let alone with a newborn)? Did you feel that same tingly feeling I did as we got more and more updates? It felt like watching an episode of OFMD again, for the first time, not knowing where it was going. It was a roller coaster ride of emotions, and as usual David is perfect with timing of specific revelations for maximum affect. We get to interact with him and all of our crewmates in the wee hours of 2024, and feel that same love and excitement that brought us together in the first place-- and I really do think that's such a wonderful gift. Seeing so many folks get excited, and brains firing off a mile a minute on new meta and new headcanons has been wonderful to watch. Whatever happens, whatever comes of it, we are getting to feel that unexpected excitement again. That little flutter of uncertainty-- that spark of joy and hope. So don't despair okay? Whatever happens, we still have our S1, and S2, and we still have each other-- and if we get an s3 out of all this too-- great! Whatever happens, we still have this insanely talented, and kind, and breathtakingly creative crew, and I am so very grateful for each and every one of you. OFMD changed my life for the better. You all changed my life for the better. You all changed SO MANY lives for the better, just by being you and sharing your love of this brilliant show. OFMD is the gift that keeps on giving, and I'm grateful David's trying to give us a bit more to play with for now, to do with as we will. Happy Holidays Lovelies and Happy New Year. May you get the rest and love you deserve, and all the things you desire this year.

View this post on Instagram

A post shared by @thelatestkate

Source: The Latest Kate's Instagram

[sheikfangirl]

Just wanted to ask, did you see the new Zelda Lego set? What do you think of the BOTW Zelda/Link minifigures?

Hey it's a nice coincidence you are asking @nintendofan4ever, I was actually on my way to make a post about the Lego set 😂

Something to know about me in real life: I am a Lego enthusiast, and I do have a decent collection. (That's a couple of my models, but my 8yo son's collection is bigger) 🤭

Yup! I saw the Deku Tree Lego set alright! I fact, I was happy to finally see pictures without watermarks because the set actually leaked online last year and Lego groups were talking a lot about it.

The mini figs? They are GLORIOUS!! Gotta love Lego Zelink my friend! Their hair, accessories....The little Sheikah slate, the Shields. My only nitpick is their hand color. I think they should be colored like their gloves. I might switch them if i actually buy the set. Yeah...i dont know if i'll get it. I expected the retail price to be expensive but 400$ CAD?! It's insane for the amount of pieces...the majority of them being leaves. I got enough Lego xp to know this set will be quick to build... Also, since the set leaked last year, i have kept ALL my Lego VIP reward points for THIS SET. I got freakin 10 000++ points and it is still overpriced. But i waaaant iiiiiiitttttt! It's really a fight between my heart and my reason, i don't know who will win yet.

...but one of the reason i want the Deku Tree Set is that its also a set from my upcoming 18 pages Zelink comic: "The Message", a title that is absolutely not spoiling the plot, right? 😂

I now have an official deadline to release it publicly. I have to beat the Lego set hahaha. I will want to see stop motion versions of my damn comic!

Here is another teaser, just for fun! It's ugly rough sketches time!

Cheers!

[eemoo1o-animoo]

10 Grelle-based paraphernalia that I found on eBay, in no particular order but my subjective deeming of relevance or importance

Possibly part one of a series? Perhaps?

If so, then I probably won’t contain it to only eBay, and I’m going off of what comes up for someone who’s based in the UK, so maybe there’s more or less stuff available if you’re in other countries due to shipping limitations, etc.? But eBay at the moment is my main source, and I’m only doing this — work that anyone can do, nothing exceptionally thorough or insightful — because I’m just a little bored, and I know Grelle is the most valued in the fandom and TSOGTR hasn’t been writing itself for the past year and a quarter, so why would it start now?

This is written at 3AM-6AM on 29/8/2024, so if this is found long after this date then there’s a chance my sources have already been bought out or deleted for whatever reason.

1.

Grell Sutcliff Funko Pop 18 Kuroshitsuji Black Butler POP! Vinyl Anime Figure

At the time of writing this, there are no bids for this and so the price remains at £21.99 (item and shipping). The current bid is due to end on ≈1/9/2024. Wait for the last second to bid one pence higher than the asking price and you’re golden.

I actually have one of these, but it was gifted to me, and so I don’t know how much mine was sold for, but seeing as the four Black Butler Funko Pops have been vaulted and are sold for way higher in other places online, I think this is a pretty good deal.

(Fun fact, the person who gifted her to me as well as tracked down other cheap listings for the other three called her Greta. Consistently. So. Greta. Greta Sutcliff.)

As we can see from the item’s pictures, the figure has been taken out of its box, at the very least just to be show-cased or inspected, but it does appear to come with the box. Here is a listing on Amazon (9.99 GBP, excluding shipping) for 10 Funko Pop display sleeves that should match the item’s measurements.

The current conversions for this item’s price (18.99 GBP) are as follow:

25.06 USD

36.91 AUD

33.77 CAD

22.52 Euro

2,103.26 INR

855,31 Turkish Lira

39.92 NZD

3,623.33 Yen

Obviously I’m not including every conversion in here, only the most obvious/prevalent. If your country’s currency isn’t here, then my apologies, and feel free to bring it to my attention (with currency named) so that I may add it to any future episodes to this possible series, or to edit this post back with the conversions of that time.

2.

Black Butler PVC Keychain Grell

The description of this listing says that this is officially licensed and from one reverse image search to find a bunch of listings across several websites for it, I’m inclined to believe that this is actually an official piece of merch and not a fan-made item that is being resold as I had originally thought it to be. (Wow, I’m such a knowledgable Black Butler fan. I know, I know.)

There’s not much to say on this other than the fact that it’s cute and I think a few pence has been shaved off from the original selling price, though I’m not too sure. I imagine it would have easily gone for 8.99 GBP, but again I’m not too sure; it’s 3AM after all and this isn’t a serious project for me.

The current conversions go as followed:

11.87 USD

17.47 AUD

15.99 CAD

10.66 Euro

996.15 INR

405,49 Turkish Lira

18.90 NZD

1,715.62 Yen

3.

Grelle Black Butler manga plushie Anime Soft Toy

It’s not the best price, I admit. If you’re looking for something cheaper then maybe go for the 5" plush, and there’s another listing on AnimeWare, but once tax and shipping get evolved it ends up being more or less the same price, anyway.

I think this is under the same line as Plushlois comes from, and to that I say: if someone buys her, will that come back? Will Grelle’s face be remade by @fancymeatcomputers, as @nullbutler’s Plushlois once had?

Anyway, she’s super cute. Her hair is a bit too saturated, but that’s not much to lament over, and she’s pricey, but is still far less expensive than any available Plushlois listing I could find (£52!! Usually!!).

Here are the current conversions:

39.61 USD

58.30 AUD

53.35 CAD

35.58 Euro

3,323.16 INR

1,351.58 Turkish Lira

63.01 NZD

5,727.47 Yen

4.

Black Butler Kuroshitsuji Grell Undertaker Anime two sided Pillow Case Cover 250

I’d originally listed another pillow for listing number four, which goes for the same price, but decided not to as Grelle is more prominent on this one.

Not much to say about this other than as far as I can tell it’s just the pillowcase and not a case and a pillow, but if you want to add a lil extra something to your Grelle shrine or limit your chances of finding a normie partner by having this as the centrepiece to your 20-something pillow statement on your romantic four-post bed.

Here are the current conversions:

7.91 USD

11.64 AUD

10.65 CAD

7.10 Euro

663.67 INR

269.94 Turkish Lira

12.58 NZD

1,143.54 Yen

5.

Black butler Grell Sutcliffe figure Kuroshitsuji SEGA PVC Scale Anime s01

Another shambolic price, I know. But out of all the non-Funko Pop Grelle figures listed, this is definitely the cheapest.

Out of all of the non-Funko Pop Black Butler figures, it’s far from it. I think the cheapest figures went to Ciel and Sebastian, but with a massive gap between the cheapest Ciel figure versus the cheapest Sebastian figure (there’s a lot of listings for this particular figure, all with free shipping and of similar prices). There’s also a listing that sells Sebastian and Grelle together with free postage, which would value each figure as far less than what this one figure is being sold for.

The current conversions (of 130.22 GBP) go as follows:

172.00 USD

252.73 AUD

231.50 CAD

154.48 Euro

14,430.37 INR

5,874.98 Turkish Lira

273.31 NZD

24,865.90 Yen

6.

Black Butler Grell Sutcliff Cosplay Costume Red Death Men Full Set Outfits New

I’m not much of cosplayer or a cosplay fan, but from the few full cosplays I found that range from £40-50, this seems like a good price.

There’s also a good few listings for Grelle-style shoes? The cheapest of which would be here. And both the shoes and the clothing come in men’s sizes, so it’s a match if you ask me.

The current conversions of the cosplay’s price are:

23.76 USD

34.92 AUD

31.98 CAD

21.34 Euro

1,993.77 INR

811.52 Turkish Lira

37.76 NZD

3,435.09 Yen

7.

Kuroshitsuji Black Butler Madam Red Angelina Dalles Anime Figurine - Play Arts

Look. I know this isn’t a Grelle figure, but I assume that most if not all Grelle fans are also Madame Red fans. And you have to admit that this is a damn good price for a figure of this type. In fact, I’d say this figure is the second cheapest, right behind the aforementioned Ciel figure.

Also the seller has the username “welsh lady”. Surely it’d basically be treason if I didn’t promote it.

The conversions of the item price (6.99 GBP) go as follows:

9.23 USD

13.57 AUD

12.43 CAD

8.29 Euro

774.85 INR

315.29 Turkish Lira

14.68 NZD

1,336.09 Yen

8.

USED Dolk Black Butler Book of Circus Grell Sutcliff Cast Doll 700mm Japan RARE

This is a fucking awful price. It hurts just to imagine buying this. However, it seems like a rare or custom-made doll, in a fancy box, and it was the only listing I could find of it, meaning it’s not a re-sell.

There’s not much to say — as you can see I ranked it pretty low on the relevancy scale, but I don’t know: maybe someone’s into this sort of thing, especially if it comes to Grelle.

The item prices are converted as follows:

2,805.52 USD

4,123.16 AUD

3,776.66 CAD

2,519.24 Euro

235,385.31 INR

95,779.75 Turkish Lira

4,458.47 NZD

405,826.75 Yen

9.

1 x playing card Black Butler Anime Sebastian Michaelis - Jack of Hearts ZT 85

For one playing card, to me this price seems ghastly. There’s also a Sebastian variant for the exact same price, but that’s all I can find in this style. The other listings for Black Butler playing cards aren’t in this style, coming wholly in colour with patterned frames. As far as I can see, on eBay there are only two listings for this type of playing card, and I’ve already linked them both to this post.

The current conversions of the item price (2.49 GBP) are:

3.29 USD

4.83 AUD

4.43 CAD

2.95 Euro

276.04 INR

112,34 Turkish Lira

5.23 NZD

476.23 Yen

10.

Black Butler Ceramic Mug Grell, Sebastian

There’s a more expensive mug with just Grelle on it that I’d contemplated listing here instead. (And, hey! From Germany, too. The TSOGTR imagery just won’t quit.) But this one is cheaper.

Besides, enough scrubbing with some iron wool will get rid of that ugly demon mug (no pun intended) in no time, surely!

The current conversions go as follows:

19.62 USD

28.84 AUD

26.41 CAD

17.62 Euro

1,646.22 INR

669,76 Turkish Lira

31.17 NZD

2,841.32 Yen

[choshasan]

sometimed it feels like japan, china and korea just... hate making money??

Like... why do you make it so fucking hard to buy stuff outside of your country?? I get it if it's like.. more niche merch like underground artists and stuff, they often won't really find an audiance outside of their homecountry, but like... when something is super popular globaly.. idk, maybe?? Try and sell some outside of the country???

This is absolutely NOOOOT a post directly directed at Hoyoverse and making it's merch fucking impossible to find online and me being really mad because I cannot for the life of me find how and where to buy the new HSR perfumes, specifically Aventurine's....... noooooo....... ahem....

But I often do face tgis issue with japanese and korean stuff, like.. trynna buy kpop albums is a hell, I always have to go through resellers on amazon and such,

If it's an older album, I just cannot get it at all, and lightsticks, other than BTS's they're impossible to find...

And like.. gods have I cried over not being able to find and order japanese merch..

I'm always glad when that worldshopping pop up comes up, cuz like, while it's gonna be 50$-75$ CAD more on my order, at least I'm getting my shit shipped...

But like fuck... espetially blue period.. it has such a big audiance outside of japan, yet, the collab merch is so hard to get.. I wasn't able to get shit from Burberry collab, converse collab nor the perfumes they released, and I wanted Hashida's so much.. :/

And figurines are the fucking worse..

It's always like do not order fakes!! They're made illegally!!

*proceeds to give zero ways to get them outside of japan* ... bruh.

Anyway.. I'm mad cuz after 7 years, I finished my Shawn Mendes perfume, and the two I've wanted (blue period & HSR) have both not had ways to be ordered outside of their homecountries... and if I wanted to rebuy my shawn mendes perfume, it was the 1st one when it released, so it'd be 150$ at least... so damn annoying... and me is frugal ✨️

The only way anyone is getting me to pay 150$+ for perfume is with fucking shipping and import fees...

[radioactiveradley]

Been a while, crocodiles. Let's talk about cad.

or, y'know...

Yep, we're doing a whistle-stop tour of AI in medical diagnosis!

Much like programming, AI can be conceived of, in very simple terms, as...

a way of moving from inputs to a desired output.

See, this very funky little diagram from skillcrush.com.

The input is what you put in. The output is what you get out.

This output will vary depending on the type of algorithm and the training that algorithm has undergone – you can put the same input into two different algorithms and get two entirely different sorts of answer.

Generative AI produces ‘new’ content, based on what it has learned from various inputs. We're talking AI Art, and Large Language Models like ChatGPT. This sort of AI is very useful in healthcare settings to, but that's a whole different post!

Analytical AI takes an input, such as a chest radiograph, subjects this input to a series of analyses, and deduces answers to specific questions about this input. For instance: is this chest radiograph normal or abnormal? And if abnormal, what is a likely pathology?

We'll be focusing on Analytical AI in this little lesson!

Other forms of Analytical AI that you might be familiar with are recommendation algorithms, which suggest items for you to buy based on your online activities, and facial recognition. In facial recognition, the input is an image of your face, and the output is the ability to tie that face to your identity. We’re not creating new content – we’re classifying and analysing the input we’ve been fed.

Many of these functions are obviously, um, problematique. But Computer-Aided Diagnosis is, potentially, a way to use this tool for good!

Right?

....Right?

Let's dig a bit deeper! AI is a massive umbrella term that contains many smaller umbrella terms, nested together like Russian dolls. So, we can use this model to envision how these different fields fit inside one another.

AI is the term for anything to do with creating and managing machines that perform tasks which would otherwise require human intelligence. This is what differentiates AI from regular computer programming.

Machine Learning is the development of statistical algorithms which are trained on data –but which can then extrapolate this training and generalise it to previously unseen data, typically for analytical purposes. The thing I want you to pay attention to here is the date of this reference. It’s very easy to think of AI as being a ‘new’ thing, but it has been around since the Fifties, and has been talked about for much longer. The massive boom in popularity that we’re seeing today is built on the backs of decades upon decades of research.

Artificial Neural Networks are loosely inspired by the structure of the human brain, where inputs are fed through one or more layers of ‘nodes’ which modify the original data until a desired output is achieved. More on this later!

Deep neural networks have two or more layers of nodes, increasing the complexity of what they can derive from an initial input. Convolutional neural networks are often also Deep. To become ‘convolutional’, a neural network must have strong connections between close nodes, influencing how the data is passed back and forth within the algorithm. We’ll dig more into this later, but basically, this makes CNNs very adapt at telling precisely where edges of a pattern are – they're far better at pattern recognition than our feeble fleshy eyes!

This is massively useful in Computer Aided Diagnosis, as it means CNNs can quickly and accurately trace bone cortices in musculoskeletal imaging, note abnormalities in lung markings in chest radiography, and isolate very early neoplastic changes in soft tissue for mammography and MRI.

Before I go on, I will point out that Neural Networks are NOT the only model used in Computer-Aided Diagnosis – but they ARE the most common, so we'll focus on them!

This diagram demonstrates the function of a simple Neural Network. An input is fed into one side. It is passed through a layer of ‘hidden’ modulating nodes, which in turn feed into the output. We describe the internal nodes in this algorithm as ‘hidden’ because we, outside of the algorithm, will only see the ‘input’ and the ‘output’ – which leads us onto a problem we’ll discuss later with regards to the transparency of AI in medicine.

But for now, let’s focus on how this basic model works, with regards to Computer Aided Diagnosis. We'll start with a game of...

Spot The Pathology.

yeah, that's right. There's a WHACKING GREAT RIGHT-SIDED PNEUMOTHORAX (as outlined in red - images courtesy of radiopaedia, but edits mine)

But my question to you is: how do we know that? What process are we going through to reach that conclusion?

Personally, I compared the lungs for symmetry, which led me to note a distinct line where the tissue in the right lung had collapsed on itself. I also noted the absence of normal lung markings beyond this line, where there should be tissue but there is instead air.

In simple terms.... the right lung is whiter in the midline, and black around the edges, with a clear distinction between these parts.

Let’s go back to our Neural Network. We’re at the training phase now.

So, we’re going to feed our algorithm! Homnomnom.

Let’s give it that image of a pneumothorax, alongside two normal chest radiographs (middle picture and bottom). The goal is to get the algorithm to accurately classify the chest radiographs we have inputted as either ‘normal’ or ‘abnormal’ depending on whether or not they demonstrate a pneumothorax.

There are two main ways we can teach this algorithm – supervised and unsupervised classification learning.

In supervised learning, we tell the neural network that the first picture is abnormal, and the second and third pictures are normal. Then we let it work out the difference, under our supervision, allowing us to steer it if it goes wrong.

Of course, if we only have three inputs, that isn’t enough for the algorithm to reach an accurate result.

You might be able to see – one of the normal chests has breasts, and another doesn't. If both ‘normal’ images had breasts, the algorithm could as easily determine that the lack of lung markings is what demonstrates a pneumothorax, as it could decide that actually, a pneumothorax is caused by not having breasts. Which, obviously, is untrue.

or is it?

....sadly I can personally confirm that having breasts does not prevent spontaneous pneumothorax, but that's another story lmao

This brings us to another big problem with AI in medicine –

If you are collecting your dataset from, say, a wealthy hospital in a suburban, majority white neighbourhood in America, then you will have those same demographics represented within that dataset. If we build a blind spot into the neural network, and it will discriminate based on that.

That’s an important thing to remember: the goal here is to create a generalisable tool for diagnosis. The algorithm will only ever be as generalisable as its dataset.

But there are plenty of huge free datasets online which have been specifically developed for training AI. What if we had hundreds of chest images, from a diverse population range, split between those which show pneumothoraxes, and those which don’t?

If we had a much larger dataset, the algorithm would be able to study the labelled ‘abnormal’ and ‘normal’ images, and come to far more accurate conclusions about what separates a pneumothorax from a normal chest in radiography. So, let’s pretend we’re the neural network, and pop in four characteristics that the algorithm might use to differentiate ‘normal’ from ‘abnormal’.

We can distinguish a pneumothorax by the appearance of a pleural edge where lung tissue has pulled away from the chest wall, and the radiolucent absence of peripheral lung markings around this area. So, let’s make those our first two nodes. Our last set of nodes are ‘do the lung markings extend to the chest wall?’ and ‘Are there no radiolucent areas?’

Now, red lines mean the answer is ‘no’ and green means the answer is ‘yes’. If the answer to the first two nodes is yes and the answer to the last two nodes is no, this is indicative of a pneumothorax – and vice versa.

Right. So, who can see the problem with this?

(image courtesy of radiopaedia)

This chest radiograph demonstrates alveolar patterns and air bronchograms within the right lung, indicative of a pneumonia. But if we fed it into our neural network...

The lung markings extend all the way to the chest wall. Therefore, this image might well be classified as ‘normal’ – a false negative.

Now we start to see why Neural Networks become deep and convolutional, and can get incredibly complex. In order to accurately differentiate a ‘normal’ from an ‘abnormal’ chest, you need a lot of nodes, and layers of nodes. This is also where unsupervised learning can come in.

Originally, Supervised Learning was used on Analytical AI, and Unsupervised Learning was used on Generative AI, allowing for more creativity in picture generation, for instance. However, more and more, Unsupervised learning is being incorporated into Analytical areas like Computer-Aided Diagnosis!

Unsupervised Learning involves feeding a neural network a large databank and giving it no information about which of the input images are ‘normal’ or ‘abnormal’. This saves massively on money and time, as no one has to go through and label the images first. It is also surprisingly very effective. The algorithm is told only to sort and classify the images into distinct categories, grouping images together and coming up with its own parameters about what separates one image from another. This sort of learning allows an algorithm to teach itself to find very small deviations from its discovered definition of ‘normal’. 

BUT this is not to say that CAD is without its issues.

Let's take a look at some of the ethical and practical considerations involved in implementing this technology within clinical practice!

(Image from Agrawal et al., 2020)

Training Data does what it says on the tin – these are the initial images you feed your algorithm. What is key here is volume, variety - with especial attention paid to minimising bias – and veracity. The training data has to be ‘real’ – you cannot mislabel images or supply non-diagnostic images that obscure pathology, or your algorithm is useless.

Validation data evaluates the algorithm and improves on it. This involves tweaking the nodes within a neural network by altering the ��weights’, or the intensity of the connection between various nodes. By altering these weights, a neural network can send an image that clearly fits our diagnostic criteria for a pneumothorax directly to the relevant output, whereas images that do not have these features must be put through another layer of nodes to rule out a different pathology.

Finally, testing data is the data that the finished algorithm will be tested on to prove its sensitivity and specificity, before any potential clinical use.

However, if algorithms require this much data to train, this introduces a lot of ethical questions.

Where does this data come from?

Is it ‘grey data’ (data of untraceable origin)? Is this good (protects anonymity) or bad (could have been acquired unethically)?

Could generative AI provide a workaround, in the form of producing synthetic radiographs? Or is it risky to train CAD algorithms on simulated data when the algorithms will then be used on real people?

If we are solely using CAD to make diagnoses, who holds legal responsibility for a misdiagnosis that costs lives? Is it the company that created the algorithm or the hospital employing it?

And finally – is it worth sinking so much time, money, and literal energy into AI – especially given concerns about the environment – when public opinion on AI in healthcare is mixed at best? This is a serious topic – we’re talking diagnoses making the difference between life and death. Do you trust a machine more than you trust a doctor? According to Rojahn et al., 2023, there is a strong public dislike of computer-aided diagnosis.

So, it's fair to ask...

why are we wasting so much time and money on something that our service users don't actually want?

Then we get to the other biggie.

There are also a variety of concerns to do with the sensitivity and specificity of Computer-Aided Diagnosis.

We’ve talked a little already about bias, and how training sets can inadvertently ‘poison’ the algorithm, so to speak, introducing dangerous elements that mimic biases and problems in society.

But do we even want completely accurate computer-aided diagnosis?

The name is computer-aided diagnosis, not computer-led diagnosis. As noted by Rajahn et al, the general public STRONGLY prefer diagnosis to be made by human professionals, and their desires should arguably be taken into account – as well as the fact that CAD algorithms tend to be incredibly expensive and highly specialised. For instance, you cannot put MRI images depicting CNS lesions through a chest reporting algorithm and expect coherent results – whereas a radiologist can be trained to diagnose across two or more specialties.

For this reason, there is an argument that rather than focusing on sensitivity and specificity, we should just focus on producing highly sensitive algorithms that will pick up on any abnormality, and output some false positives, but will produce NO false negatives.

(Sensitivity = a test's ability to identify sick people with a disease)

(Specificity = a test's ability to identify that healthy people do not have this disease)

This means we are working towards developing algorithms that OVERESTIMATE rather than UNDERESTIMATE disease prevalence. This makes CAD a useful tool for triage rather than providing its own diagnoses – if a CAD algorithm weighted towards high sensitivity and low specificity does not pick up on any abnormalities, it’s highly unlikely that there are any.

Finally, we have to question whether CAD is even all that accurate to begin with. 10 years ago, according to Lehmen et al., CAD in mammography demonstrated negligible improvements to accuracy. In 1989, Sutton noted that accuracy was under 60%. Nowadays, however, AI has been proven to exceed the abilities of radiologists when detecting cancers (that’s from Guetari et al., 2023). This suggests that there is a common upwards trajectory, and AI might become a suitable alternative to traditional radiology one day. But, due to the many potential problems with this field, that day is unlikely to be soon...

That's all, folks! Have some references~

[crabsnpersimmons]

Hello there!

I'd like to know your advice on drawing tablets?

As in, what do you use for your digital art? Is there any you suggest?

I'm planning on getting one, but there are so many options, and I'd like to know your opinion and preference.

Thank you ^^

hi anon! thanks for the question!

i draw with a samsung galaxy tab s7!

BUT if this is your first time drawing digitally, then it's a good idea to test out if you like it with cheaper options first (so you don't drop a whole lot of money on a hobby/medium of art you don't enjoy)

some suggestions:

if you have a smart phone, try drawing with your finger using free to use drawing apps like ibispaint X or HiPaint (i recommend HiPaint cuz the UI is a lot simpler and less overwhelming, but both have a good selection of tools. i wanna say ibispaint has better text tools tho and there's A LOT more tutorials and resources online) drawing digitally with your finger is different from drawing digitally with a pen AND drawing traditionally, but this is a good way to familiarize yourself with digital art tools. a very simple entry point if you already have a smart phone.

if you don't want to draw with your finger (understandable, i personally couldn't stand it) you can try simple drawing tablets, which are the tablets that you plug into your computer/laptop, come with a pen, but they don't have a screen you draw directly on. they're actually a lot more affordable nowadays (i saw some for under $50 CAD, which is WILD), so that’s a good starting option if you want to draw with a pen. with a pen, you have more control with line weights than you do with your finger, but it takes practice to get used to. it also takes some getting use to drawing on a tablet and looking at your monitor, but this is where i started haha. i still have my old bamboo wacom create drawing tablet. but nowadays, other brands like xp-pen and huion are very comparable to wacom, so definitely consider getting a cheaper one to try out first. of course, if you go with a drawing tablet, you'll need an art program on your computer. i've used photoshop and clip studio paint and they're very good, but probably overwhelming for people who are new to digital art (they are loaded with tools and you have to really familiarize yourself with the menus and everything). there are probably simpler options available... but i'm not as familiar with them. oh wait, i think there are some drawing tablets that can also connect with your smart phone too nowadays. man, that's so cool. that's probably a better option instead of purchasing photoshop or clip studio paint right off the bat.

if you REALLY want a screen display experience, try a cheaper general tablet. iPads are a popular option if you're comfortable with the Apple ecosystem (i'm not 😅) but they can be pricey PLUS you have to buy the Apple pen separately. for Android, I'm pretty sure even the Samsung Galaxy Tab S6 is a solid choice and it comes with a pen unlike an iPad. the SPen is built with Wacom tech, so you know it's quality. and while iPad has Procreate, HiPaint is available for Android and it's basically a Procreate clone (i explain more about HiPaint here) otherwise, ibisPaint X is also good and infinite painter is probably the most sophisticated free to use drawing app i've encountered (sophisticated as in it does a pretty good job at imitating traditional media in their brushes. it's cool, but the free version is very limited).

also if you like watching videos, Brad Colbow and Teoh on Tech on YouTube are great resources. they specifically review tech from an artist's perspective.

why i suggest testing it out is because if you're used to drawing traditionally, then switching to digital is a different experience. some artists can't get used to drawing digitally, and that's fine. that's why it's a good idea to test it out with a cheaper option first to see if you like it and to give yourself time to familiarize yourself with drawing apps/programs. i actually kinda hated drawing digitally at first. i didn't have the same level of control that i was used to with drawing traditionally. and overtime i made a lot of bad habits with my digital art that i had to unlearn by switching back to traditional art for a couple years (some bad habits like: relying on the Undo button to make "perfect" lines; trying to draw really fast because the pen glides over the screen unlike a pencil has friction against paper) so... give it a try and just let yourself have fun. don't worry about making anything perfect, just have fun!

[colicoffs]

Buy Cad 20 Bills Online | Coli Coffs

If you want to buy CAD 20 Bills online, Coli Coffs is a reliable website for this. Here you can find high quality Canadian notes at a reasonable price. Order now and add a new one to your collection.

[zomnommbie]

currently trying to figure out how i could digitalize the mask to be a 3d print/render so it makes cosplaying him a fuck ton easier lmao

at first i was gonna buy a mask and maybe do a repaint of it but the ones i found online, the absolute cheapest and closest to the movie mask was 60$ CAD.. and im a cheap bastard

there where also ones where they where 200$ CAD, and they where super accurate and cool i will admit but like, 200$ is insane, so out of spite i spent months on the one i have lmao

if i am able to digitalize it so anyone who has access to a 3d printer is able to have a cheaper alternative, would you be interested? i cant promise anything as im completely out of my depth here but im stubborn and dont want people to have to spend money on something that doesnt bring the satisfaction it should

sorry for the rant ♛

[kureffui]

"That dream... Make it come true!"

N Harmonia Acrylic Charm (2.5")

Will be available at Anime North 2023 (and potentially on my online shop at a later date)!

Update: You can now buy him (and others) on my etsy! Use code HALLOWEEN23 for 20% off any order of $75 CAD or more til October 31st! https://kureffuiart.etsy.com

[lambment]

What drawing tablet do you use, and would you recommend it? I'm in the market for a new one, but I have NO idea what actually makes a good one, so I'm asking around

The vast majority of my posts are made with my 1st gen 12” ipad (with a paper feel screen protector) + procreate!

I also use an older model of a 25” wacom cintiq I got back in 2016 when I went to anim school. I wouldn’t consider this cost efficient if you are new to digital art tho, and would recommend a bamboo or intuos. I’ve also heard huion has pretty good tablets now though, and cheaper.

If you’re looking to buy:

I snagged the ipad for 100 CAD (a HUGE steal at the time) a couple years back off of a local online marketplace. I’m sure if you look on marketplaces for a used old Gen now, they should be relatively cheap. Just keep in mind the price of an Apple pen + planned obsolescence. (It’s lasted me all through college and now tho).

If you were looking for a cintiq, I would recommend contacting sellers after they’ve been displayed at conventions, they sell at a big discounted price because of sampling.