Good Thinking

X-ray vision, hidden meanings and disease trackers: IBM's 5 in 5 for 2017

View 6 Images
The human brain may work in mysterious ways, but AI might be able to detect signs of mental health issues through the language we use
IBM Research
Experimenting with millimeter-wave radar imaging
Carl De Torres, StoryTK for IBM Research
With this silicon wafer, scientists can sort particles found in bodily fluids and detect diseases at an early stage
IBM Research
Networks of sensors like this miniature silicon chip trace-gas spectrometer may help us detect environmental pollutants
IBM Research
With AI, our words will reveal our state of mind
IBM Research
The human brain may work in mysterious ways, but AI might be able to detect signs of mental health issues through the language we use
IBM Research
Macroscopes will help us understand the earth's complexities in greater detail
View gallery - 6 images

While its late chairman Thomas Watson might have famously downplayed demand for computers back in the 1940s (to be fair, the machines he was referring to were clunky beasts that took up an entire room), that hasn't stopped the tech giant from making predictions about future trends. Since 2005, IBM has been releasing an annual list of five technological advancements that it thinks will take off in the next five years. In the past, they've focused on urbanization as well as machine learning and big data (with mixed results – perhaps five years is too optimistic a time frame for many of these technologies to take root). Unsurprisingly, given the emergence of AI and the Internet of Things, this year's predictions are focused on these two areas.

AI reveals a hidden side to what we say and write

With AI, our words will reveal our state of mind
IBM Research

Despite the fact that mental illness affects more than 450 million people in the world, a large number of cases go undiagnosed and unaddressed. What if machine analytics could help doctors predict and monitor conditions such as psychosis, schizophrenia, mania and depression? Our speech and writing patterns reveal a lot about our state of mind, and IBM believes that doctors could make use of AI to catch tell-tale signs that they might otherwise miss.

"Today, analysis of language is done in a labor-intensive process of manually interviewing and recording multiple lengthy sessions with a patient. There is no way to quantify or codify these sessions, resulting in a massive 'big data' problem," says neuroscientist and IBM researcher Guillermo Cecchi. "A tool that could, in near real-time, analyze and codify a sample of the patient's speech and provide an analysis would drastically shorten the time it took for doctors and caregivers to predict and diagnose patients. For all conditions, the earlier a diagnosis is made, the higher likelihood of a successful treatment and management of the disease."

A study conducted in collaboration with Columbia University psychiatrists showed that computers installed with a speech analysis program were able to predict - with 100 percent accuracy – who among a population of at-risk adolescents would develop their first episode of psychosis within two years.

Superhero vision is no longer just for superheroes

Experimenting with millimeter-wave radar imaging
Carl De Torres, StoryTK for IBM Research

X-ray vision might seem like something you read about only in comic books, but IBM believes that we too could one day see things that currently elude the naked eye.

At present, we can only see visible light waves, which make up less than 0.1 percent of the electromagnetic spectrum. IBM is working on a compact hyperimaging platform that makes use of millimeter waves for imaging. These waves have longer wavelengths than light and X-rays, and their high frequency – they occupy a radio spectrum between 30 GHz and 300 GHz – makes them ideal for transmitting large amounts of data. With millimeter wave imaging, users wouldn't be restricted to a specific portion of the electromagnetic spectrum, but would be able to see across a wide swathe.

The game-changing potential here is enormous: not only could it be used in road safety applications and help drivers identify dangerous conditions such as black ice, but the technology could also be integrated in our phones and used to check the nutritional value of food or the authenticity of a drug or bank check. On the flip side, this technology also raises privacy concerns – how do you stop people from seeing what you don't want them to see?

Connecting the dots generated by Big Data

Macroscopes will help us understand the earth's complexities in greater detail

Thanks to the Internet of Things, we now have more data than we can handle. However, under all that clutter lies a wealth of information that could offer solutions to many of the world's problems.

This is where macroscope technology comes in. IBM is currently building what it claims to be the world's first platform for collecting, curating and searching global data by space and time. In the case of agricultural communities, this technology could reveal new insights on climate, soil conditions and water resources, thereby helping farmers to make better crop-growing decisions.

Case in point: California-based Gallo Winery, which teamed up with IBM to develop a customized irrigation system using analyzed satellite data. What Gallo wanted to do was identify which sections of the vineyard needed water, and deliver it to those specific areas or plants. In an area prone to drought, doing this would also help reduce its water consumption. After one season testing the new technology, it was able to reduce water consumption by 25 percent while increasing crop yield by 26 percent.

Needless to say, IBM hopes to scale this concept and sell the technology to other farmers. However, in order for farming communities – especially those that need it most – to benefit from the technology, it has to be priced reasonably. Otherwise, the data will only be able to serve those who can afford it.

Catching disease before it wreaks havoc on your body

With this silicon wafer, scientists can sort particles found in bodily fluids and detect diseases at an early stage
IBM Research

"The challenge to finding a disease early is that many of us don't seek treatment until we have symptoms, which means the disease has already progressed," says computational biologist Gustavo Stolovitzky. It also means that many patients end up going into debt trying to stay alive. According to the American Society of Clinical Oncology, patients can expect to pay an average of US$10,000 a month for the most recently approved cancer medicine, while other therapies can cost upwards of $30,000 per month.

Lab-on-a-chip nano-biotechnology techniques would allow doctors to screen bodily fluids, such as blood and urine samples, for signs of disease before the development of full-blown symptoms. We've previously covered studies using such technology to detect cancerous tumor cells and the HIV virus. IBM scientists are going one up with lab-on-a-chip nanotechnology that can separate and isolate bioparticles down to 20 nanometers in diameter, a scale that will give scientists access to DNA, viruses and exosomes.

Apart from being more comfortable and convenient than traditional tissue biopsies or screening techniques, the portable nature of this technology means it could also be integrated in a handheld device for quick testing, the data of which can then be combined with that of other Internet of Things devices, such as fitness and sleep trackers, to generate an in-depth report of one's health.

Detecting pollution as it happens, with smart sensors

Networks of sensors like this miniature silicon chip trace-gas spectrometer may help us detect environmental pollutants
IBM Research

IBM believes that silicon photonics, an evolving technology that transfers data by light, will enable the development of affordable sensing technologies that can be situated on or near sites such as natural gas extraction wells, storage facilities and distribution pipelines.

This would enable oil and gas companies to identify and pinpoint invisible leaks in real time, as well as identify a wide variety of pollutants in the environment. Using AI, the accumulated data from these sites as well as other sources could then be used to build environmental models to detect pollutants as they occur. On another note, the technology could also be used to detect contaminants in the air we exhale, the results of which could then be used to improve the diagnosis of respiratory diseases.

While this could help reduce pollution and its impact on climate change, the onus is also on governments and corporations to develop concomitant measures to address the factors causing the pollution. It's one thing to be able to identify invisible leaks in real time, but if nothing further is done about the situation, then the technology will be for nought.

Source: IBM

View gallery - 6 images
  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
0 comments
There are no comments. Be the first!