Posts | Federica Baldi

Artificial Intelligence for Blood Cell Analysis

Thu, 22 Sep 2022 14:39:35 +0000

Millions of patients go for blood tests every day, whether for a routine checkup, to monitor a pre-existing condition, or come up with a diagnosis. The analysis and counting of the different blood cells, including red and white ones, are typically performed manually. This means that experienced laboratory technicians examine each blood sample individually. However, as in other industries, Artificial Intelligence (AI) is also about to disrupt diagnostics in haematology. In particular, Computer Vision technologies help specialists get more accurate results faster by automating part of the process. Thus, making it possible to provide patients with meticulous results in a short time.

What is blood cell analysis?

Before delving into a specific use case, let us explain what blood cell analysis is. As the name suggests, this procedure involves examining a blood sample to learn more about its characteristics. The sample is usually taken from an arm, finger, or earlobe vein; in some circumstances, bone marrow blood cells may also be analysed. This test allows doctors to detect, count, and classify the different types of blood cells. Moreover, experts can detect abnormal cells if present, opening the door for early blood disease diagnosis.

Blood cell analysis is one of the most frequent tests prescribed by physicians to monitor patients’ health status. Abnormal values can be a flag for conditions such as platelet deficiency, infections, different forms of leukaemia, anaemia or coagulation disorders. On the contrary, if the values are in the normal range, the bone marrow can produce blood cells correctly, and no infections are present in the body.

How can AI be used to detect leukaemia in blood cells?

By the term leukaemia, we collectively refer to blood cell cancers. This group of cancers typically starts in the bone marrow and results in a high number of abnormal blood cells. Based on the type of blood cells affected and the cancer progression speed, we distinguish various types of leukaemia. In any case, early detection and diagnosis are key to a good chance of a cure.

It is paramount to identify malignant cells with high accuracy to reach a correct diagnosis. However, distinguishing abnormal cells from normal ones under the microscope is a challenging task. In fact, the morphological images of the two cell types appear very similar. This is where Artificial Intelligence comes into play. Indeed, not only can Computer Vision systems help pathologists and oncologists make quicker and data-driven inferences, but they could also detect details that might otherwise escape the human eye.

What is the solution we applied in this case?

To assess the feasibility of employing an AI solution for blood cell analysis and leukaemia diagnosis, we studied the case of Acute Lymphocytic Leukaemia (ALL). This is a particular type of leukaemia characterised by the rapid development of a large number of lymphoblasts (i.e., immature white blood cells). The aim is to detect whether a white blood cell is abnormal or not so that cancer can be diagnosed. The solution we implemented is based on an artificial neural network. Specifically, it is a Convolutional Neural Network (CNN). This particular Deep Learning architecture – inspired by the biological processes within the animal visual cortex – can analyse the content of images. In practice, by processing the data contained in the pixels, the network learns its way of interpreting images and automatically extracts the distinctive features of the blood cell.

During our study, we first performed a segmentation of the microscopic images. In this way, we obtained multiple images, each containing a single blood cell. Then, we trained the network to distinguish normal from malignant cells in inference.

What are the final results of this study?

Our study resulted in an accuracy of 99%. This means that, out of 100 cells, the system misclassified only one normal cell as malignant or vice versa. This shows how, despite the task’s difficulty due to the visual similarity between normal and abnormal cells, the latest advances in Artificial Intelligence allow for excellent results. Moreover, the results are reliable, and the blood analysis AI system is scalable and can operate in a matter of seconds. This means that clinics will perform tests faster and at a lower price while maintaining the same quality.

Nonetheless, it is important to emphasise that no Machine Learning system is perfect. In a critical field such as medicine, an expert must always assess all outcomes. Thus, the ultimate goal will never be to replace physicians and specialists with AI technologies. Rather, Artificial Intelligence must become an indispensable tool in the healthcare sector, assisting clinicians in achieving their main objective, which is to give patients the best care possible.

5 Artificial Intelligence experts you should follow on YouTube

Tue, 23 Aug 2022 19:27:00 +0000

AI is undoubtedly one of the hottest topics of the moment. Everybody talks about it, but do they really know what AI is?

Artificial Intelligence is a wide-ranging field involving a variety of different aspects and subtopics. It can be quite overwhelming for a beginner to approach it or to keep up to date with the latest breakthroughs. So, what resources should you include in your learning roadmap?

We asked our team of AI experts for the best free online tools to learn about AI and keep up to date with the latest developments and they all agreed that Youtube should be one of your go-to channels.

So, without further ado, here are 5 AI Youtube channels you should definitely include in your roadmap!

What’s AI?

What’s AI? is the channel of Louis Bouchard, a young and talented Canadian graduate student in AI. The aim of this channel – whose name is self-explanatory – is to make AI accessible to (almost) everyone.

Bouchard publishes one video a week, offering interesting content while keeping things simple. Each video focuses on one of the latest AI innovations, explaining it in a comprehensive but concise manner. The content is well edited and of high quality. Plus, the narrative is clear and very easy to follow.

This channel is definitely a good starting point for beginners to grasp new complex technologies. Moreover, thanks to the well cited references, you can always delve deeper if you want to.

Two Minute Papers

You may have already heard about this channel, but it is worth mentioning in this list. Created in 2006, Two Minute Papers is perhaps one of the most popular science channels. Its strength lies in effectively “squeezing” an understanding of interesting research works into just a couple of minutes (often a little longer).

The channel is run by Dr. Károly Zsolnai-Fehér, an independent Hungarian researcher with a PhD in computer graphics and machine learning. In each video, Zsolnai-Fehér breaks down a new scientific paper, aiming to convey its fundamental ideas.

Thanks to the simple and captivating format, the content is also accessible to beginners. However, this does not mean that it is not suitable for a more experienced audience. In fact, the channel offers a useful view of the latest advances.

Sparkd AI

Sparkd AI is our very own YouTube channel and it is a very useful tool for those who want to learn about AI both for business and educational purposes. Our team of machine learning experts and PhD level engineers provides our audience with clear practical content, ranging from short demos, how-tos and detailed explanations of different technologies, our videos focus on machine learning techniques and computer vision – arguably the most in vogue topic at the moment. Sparkd AI’ s content offers a real view of what it means to work with AI and the advantages it can provide in many areas.

To AI beginners, we suggest also to include the channel 101 Question of AI in their learning roadmap. In this weekly series our CEO Dr. Luca Marchesotti answers questions about AI in bite-sized episodes of less than 15 minutes. Topics include AI’s creative ability, misconceptions about AI, and AI in Social Media, just to name a few.

WelcomeAIOverlords

The point of WelcomeAIOverlords is to help people level up their machine learning skills. The channel is run by Zac Jost, a Research Scientist at Amazon Web Services.

The interesting thing about this channel is that it offers video lectures on many aspects of AI. Yet Jost is committed to eliminating formalism and unnecessary complexity. This way, the lectures become straightforward and it is easier to understand the key points.

Moreover, Jost firmly believes that you learn best by doing. For this reason, he provides hands-on videos along with the theoretical lessons. This allows you to learn how to work in practice on technologies from scratch.

Yannic Kilcher

Yannic Kilcher is a PhD from ETH Zurich’s data analytics lab and is now the Chief Technology Officer of DeepJudge. Having experience in the field, he decided to open his own YouTube channel.

Kilcher usually covers machine learning research papers, programming, and issues within the AI community. In addition, he focuses on the broad impact of AI in society. His videos are always very informative and detailed. But most likely the reason for his popularity is the fact that he is also entertaining. In this way, learning complex concepts is made enjoyable and not at all tedious.

Final thoughts

As we have already said, AI is a very broad field, but you have to start somewhere. Even if you already have experience in the field, it is crucial not to fall behind. AI experts’ Youtube channels help you with this.

In this article, we have listed only 5 of the many YouTube channels dealing with Artificial Intelligence. Our final advice is to start by subscribing to these, then find the ones that suit you best and include them in your roadmap.