Kia ora e hoa,

My brief absence has made me realise how much I miss landing in your inbox every Sunday morning with a topic that I am passionate about. This feeling makes me want to come back sooner, and so, here I am.

In the previous newsletter, before my declared absence, we talked about what is an algorithm. We ended by having the problem of grading diabetic retinal images — a task that is not as easy as sorting people's names!

Grading retinal images for diabetic retinopathy (DR) is an image classification problem. We have images, and we need to grade them into these possible categories: healthy, background DR referrable DR.

Why is this useful?

In Aotearoa, 250,000 people have diabetes and a quarter have DR. Fortunately, individuals with diabetes are screened with routine retinal imaging taken at least biennially. But, unfortunately, that's a lot of images to grade! On top of this, grading these images requires experts — of which there are too few.

Can AI plug this gap?

But first, what is deep learning? Deep learning is a subset of machine learning. Machine learning is where computers are programmed to learn from data. Machine learning includes methods from statistics like regression and optimisation from calculus.

Deep learning takes these mathematics principles to the next level. Taking inspiration from the human brain's architecture, deep learning involves virtual neurons that interact and interconnect. An example of what deep learning can do is classify images.

Images are translated into a pixel by pixel array. This is then fed to the input layer. The input layer connects to intermediate or hidden layers, which respond to certain elements of the image. The earlier layers respond to edges and diagonals, while the latter hidden layers respond to more complicated elements in the image. In the end, an output layer provides the result (e.g. cat or dog).

The deep learning model is able to respond in this way because it has been trained on known images. This means well-curated data is needed to train the model, so it responds in the way we want.

Let's create our own image classier. The task is to classify metal frames and acetate frames. We collect this data using Bing Image Search.

However, we can see that this isn't perfect. We get metal frames, but not the ones that we were quite thinking about.

Back to the drawing board... Gathering data is difficult. In the next newsletter, we will dive deeper into deep learning. Let's change the scope of what we are learning by detecting blue eyes and brown eyes.

Did you find this useful? If you did, please forward this on to family and friends so they can find it useful too.

Thanks for reading and all the best for the week ahead.

Ngā mihi nui,

Shivan :)

My Favourite Things

1. Article How many pets do you have? Derek Sivers talks about having too many passions at once. This resonates with me since I want to learn to program, write and work full time too.

2. Video Gibbons Lectures 2021: Tikanga AI: Enabling mātauranga Māori in the future of AI development This is an excellent lecture given by Potaua Biasiny-Tule detailing the effort to introduce technology to Tangata Whenua.

3. Website This Person Does Not Exist This website is created by Phil Wang and is based on StyleGAN developed by Nvidia. This is an artificial intelligence that can recreate people's faces.

Blog Posts

The Easy Way to Host Hobby Projects • Shivan Sivakumaran — shivansivakumaran.com Save money and time hosting your hobby projects using Ansible on a single virtual machine and domain.

Kindle Highlight of the Week

The excitement about the blockchain is based on its ability to serve as a tool for achieving and maintaining integrity in purely distributed peer-to-peer systems that have the potential to change whole industries due to disintermediation.

Daniel Drescher, Blockchain Basics