Podcast: AI For Better Battery Design & Training Microrobots

01 Mar, 2022

Artist’s rendition of a particle analyzed by a combination of machine learning, X-ray and electron microscopy. (Image credit: Ella Maru Studio)

In this episode, we talk about how AI is being leveraged for designing better batteries and training microswimmers.

artificial intelligence

In this episode, we talk about how AI is being leveraged for designing better batteries and training microswimmers. As always, you can find these and other interesting & impactful engineering articles on Wevolver.com.

This podcast is sponsored by Mouser Electronics.

EPISODE NOTES

(2:30) - AI & Imaging For Better Batteries

(12:30) - AI teaches microbots how to swim

Episode 59 was brought to you by Mouser Electronics, Farbod & Daniel’s favorite electronics distributor. Click here to read the article discussing the shortcomings of current battery technologies.

About the podcast:

Every day, some of the most innovative universities, companies, and individual technology developers share their knowledge on Wevolver. To ensure we can also provide this knowledge for the growing group of podcast listeners, we started a collaboration with two young engineers, Daniel Scott Mitchell & Farbod Moghaddam who discuss the most interesting content in this podcast series.

To learn more about this show, please visit the shows page. By following the page, you will get automatic updates by email when a new show is published.

Be sure to give us a follow and review on Apple podcasts, Spotify, and most of your favorite podcast platforms!

Bio-inspired deep reinforcement learning teaches these smart magnetic microbots to swim

Designed for use in the human body for everything from drug delivery to less-invasive biopsies, these tiny microrobots operate under the control of a deep-learning system trained with no modelling or prior environmental knowledge.

Gareth Halfacree

23 Feb, 2022

artificial intelligence

This article was discussed in our Next Byte podcast.

The full article will continue below.

Microrobots able to swim through fluid are of increasing interest in the medical field, capable of delivering precisely-targeted drugs, obtaining tissue samples, or even breaking up and removing blood clots. Actually controlling the devices once they’re inside the human body, though, remains a challenge — but one a pair of scientists from the University of Pittsburgh and Carnegie Mellon University may have found a way to solve.

Having developed simple magnetic microrobots, looking like tiny corkscrews, in a 3D-printed mold, the pair set about using deep reinforcement learning to create a system which could control their movement — without any manual training or simulation, and in fewer than 100,000 training steps.

The challenge

A tiny robot, easily fitting on the very tip of your finger, isn’t easy to control, and when you build it from complex composite materials — sometimes even integrating biological components — then set it loose in a complex environment like the human body, the challenge becomes near-intractable.

With too many variable, simulation is either entirely impossible or requires a range of sacrifices in the name of simplicity — potentially reducing the performance or capabilities of the finished control system. What Michael R. Behrens and Warren C. Ruder have proposed instead is to to do away with simulation altogether and create a control system based on deep reinforcement learning.

Inspired by natural learning, this system can drive swimming microrobots to a goal with no prior knowledge or modelling required.

Designed to mimic how real-world organisms approach learning, reinforcement learning aims to allow a deep learning system to adapt based on observations made about the sate of an environment. In this case, th

Image: Retinal Scans to Predict Cardiovascular Diseases [Image Credit: University of Leeds]

Can Retinal Scans Help in Predicting Myocardial Infarction?

The proposed AI system has an accuracy between 70% and 80% that can be employed as a second referral mechanism for cardiovascular examination.

Abhishek Jadhav

14 Feb, 2022

artificial intelligence

- healthcare

- machine learning

Medical examinations are primarily based on observing associations and building hypotheses to determine the flow of medical investigation and treatment. With the introduction of deep learning models, researchers have found a way to bring medical imaging to medical discoveries. However, several factors make it difficult to improve accuracy due to a wide range of parameters like patterns, values, colors and shapes associated with real data.

Retinal images have already been in practice by ophthalmologists to diagnose and monitor eye disease. The extension of medical imaging diagnosis for cardiovascular diseases has been a key strategy to reduce CVD prevalence in the popular. Cardiovascular imaging has always been secondary care due to the cost, making it difficult for developing and underdeveloped countries to adopt. But the significant association of retinal microvascular abnormalities with systemic and cardiovascular diseases has made retinal images a routine in ophthalmologic practice.

The research article published under Nature Biomedical Engineering, “Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning,” [1] shows that analysis of retinal images has enabled the prediction of cardiovascular risk factors like age, gender, smoking status and blood pressure. Extending the line of research led by the University of Leeds designed an artificial intelligence system that can process retinal images to predict the risk of myocardial infarction.

Deep Learning Leads the Way for Predicting Myocardial Infarction

In the paper, “Predicting myocardial infarction through retinal scans and minimal personal information,” [2] the researchers reported the use of retinal images in the early identification of heart diseases. The implementation of a deep learning model to analyze eye scans will result in identifying patients at high risk of a heart attack. Along with retinal scans, relevant patient metadata has been use

Podcast: AI For Better Battery Design & Training Microrobots

In this episode, we talk about how AI is being leveraged for designing better batteries and training microswimmers.

EPISODE NOTES

The challenge

Deep Learning Leads the Way for Predicting Myocardial Infarction

Predicting surgical outcomes with machine learning

In-home wireless device tracks disease progression in Parkinson's patients

Solving brain dynamics gives rise to flexible machine learning models

In this episode, we talk about how AI is being leveraged for designing better batteries and training microswimmers.

EPISODE NOTES

The Next Byte Newsletter