In this edition of Adventures of AI, we will take a look at how Netflix can help cure cancer. We will also try to understand how robots and humans can work together. But first, we have to talk about “Living Machines” and how they can help in averting the climate crisis.
A while back, we discussed Xenobots. They were tiny biological robots with self-healing capabilities. The good news is that the same team of scientists have made new organisms that are even better.
Meet the Xenobots 2.0, a tiny biological robot made from frog skin cells. Scientists who made the new form of biological robots were published in the esteemed Science Robotics. In addition to moving faster, the Xenobots 2.0 also have a longer lifespan.
The reason behind the capabilities of Xenobots 2.0 is their “bottom up” approach. Scientists took cells from the African frog Xenopus laevis and allowed them to grow and develop under suitable conditions. After some time, a spheroid was formed, and some of the cells differentiated to form cilia.
In the original Xenobots, muscle-contraction enabled their movement. Meanwhile, the Xenobots 2.0 can now easily move thanks to cilia. While all of this was happening, scientists at UVM were already designing computer models to enhance the capabilities of Xenobots 2.0.
As it turns out, the new generation of tiny robotic creatures is excellent at garbage collection. Xenobots would work as a swarm collecting microplastics from our ocean and river bodies. As an experiment, scientists tested this ability of Xenobots by making them collect Iron Oxide from a petri dish. As it turns out, these robots are pretty good at that.
Another handy feature in the Xenobots is memory recording. Tuft’s scientists have embedded a read/write protein in Xenobots 2.0 called the EsoFP. It normally grows green, but under the effect of light with 390nm wavelength, it appears red.
Scientists injected a messenger RNA to code for EsoFP protein in the frog embryo before stem cells extraction. This ability to record/sense information can be used in sensing diseases, pollutants, drug efficacy. Combined with locomotion, Xenobots 2.0 can also deliver drugs inside the body by a particular stimulus. Targeted cancer treatment can become more effective in the future.
On top of all these amazing properties, the Xenobots 2.0 also have the ability to heal themselves. They can heal a scar almost half the size of their body in just 5 minutes. Upon realizing the amazing future, administrators at Tufts University have established the Institute for Computer Designed Organisms. It’ll officially launch in the coming weeks and be responsible for pulling in necessary internal and external funding for the project.
Robots That “Eat” Their Way Out Of Trouble
Some people are simple; wherever they see food, they turn in that direction. This is why it was surprising to know that there also exists a robot of the same nature. Instead of food, it derives energy by oxidising metal surfaces around it and feeding off the electrons that come out. This robot is, in fact, so hungry that it only always moves in the direction where there is metal.
The robot in discussion is developed by Penn state engineer James Pikul. And it functions with the help of an Environmentally Controlled Voltage Source or ECVS. This system guides the robot wherever there is metal.
Up until now, robots had to rely on a Central Processing Unit or CPU and big batteries for locomotion and energy, respectively. ECVS accounts for both of those things. This seemingly small trait can also help scientists devise new and better ways to make a robot perform various functions.
The ECVS works like an OR gate. If there is metal nearby, the robot moves towards that location. If there isn’t, then it doesn’t. As the ECVS evolves, it will be used in programs that require automation. These robots will probably be used to venture into dangerous locations and either extract or plant important entities.
Robot Co-Workers Become Smart
While robots controlled by artificial intelligence work in various assembly lines around the world, they can pose a hazard to their human co-workers in some shape or form. Robots are used for cutting, drilling and pushing heavy stuff in many factories, and with so much power, any mishap by a robot towards a human can be pretty dangerous.
To circumvent this, robots have to follow the ISO standard and technical information to work with humans safely. Under this protocol, whenever a human approaches a robot, they have to slow down. When a human is right next to a robot, it completely shuts off. This method decreases production efficiency and often causes a delay in the overall process.
Thankfully Hongyi Liu from KTH, Royal Institute of Technology, has a solution. He’s training lifeless robots to understand the context and not just the process. Liu co-authored the paper alongside Professor Lihui Wang from the same institute. Their paper was recently published in the latest issue of Robotics and Computer-Integrated Manufacturing.
With Liu’s context-based invention. Robots will be able to judge a person’s next move accurately. It’ll learn from the habits of different people, and it’ll identify them based on their skeletal structure.
Liu compares such a robot with a self-driving car that applies brakes and accelerator and predicts the traffic light, gently coming to a halt saving the brake pad from wear and tear in the process.
“This is safety not just from the technical point of view in avoiding collisions, but being able to recognize the context of the assembly line,” says Liu.
How Netflix Can Beat Cancer
Netflix is an amazing service. Part of what makes it great is its ability to judge and recommend new TV shows and movies tailored to our taste. Netflix achieves this by combining Artificial Intelligence and Machine learning, wherein it monitors our viewing habits and cleverly suggests movies that we’d like.
Scientists believe that such an awesome machine learning tech can predict deadly diseases like Cancer and Alzheimer’s that have plagued humanity since the dawn of time.
Treating such diseases is a challenge because they evolve erratically. Their randomised nature of growth within the body oftentimes makes the treatment of cancer risky.
Researchers want to minimize this risk by using the models used by Netflix and Amazon to study the language of proteins in our body. There is an unknown number of protein molecules inside us, and they all interact uniquely. Implementing a wide-scale AI model to study them can help scientists better understand how cancer reacts to our bodies.
According to the researcher Dr Kadi Liss Saar, “Any defects connected with these protein droplets can lead to diseases such as cancer. This is why bringing natural language processing technology into research into the molecular origins of protein malfunction is vital if we want to be able to correct the grammatical mistakes inside cells that cause disease.”
Dr Saar explained that machine learning could be free from limitations that scientists have. It means we will see new evidence in areas where we have never expected before.
Almost everyone knows someone who has lost a precious soul to cancer. We wish Godspeed to these scientists as we move forward into a brave new world.
So make sure to stay up to date on it in our next issue of Adventures of AI.