In short
This episode of The Naked Scientist covers three science stories: a “supercharged” El Niño forming in the Pacific, an at-home urine test for bladder cancer, and new navigation tech for blind people. El Niño: meteorologist Aidan McGiven (Met Office) explains that weakened trade winds redistribute warm water and thunderstorms eastward, with major global knock-on effects. Key claims include Nino 3.4 sea-surface temperatures already 1.39°C above average (2.6°C weekly anomaly), possible ~3°C peak, and links to drought/wildfires in Indonesia/northern Australia, wetter Peru/Chile, stormier UK, and fewer Atlantic hurricanes.
Notable examples
Jakarta’s drought and Kalimantan wildfires. Bladder test: Doug Ward (University of Birmingham) says the Gallius urine test sequences DNA mutations across 23 genes, detecting >9/10 cancers in ~7 working days; positives require urgent cystoscopy, negatives less urgent. Mobilio app: Harvard PhD student Raymond Liu (Harvard) describes an app using GPS, phone motion sensors, and camera AI to provide spatialized audio turn-by-turn guidance and obstacle alerts; it’s not a replacement for cane/guide dog and aims for public release in 1.5–2 years.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOUnderstanding El Nino
0:06 to 0:40
Exploration of the El Nino climate pattern and its global impacts.
“You know that one friend who always finds the best travel deals, picks the right cards, and somehow ends up in first class for the price of coach?”
Understanding El Nino
1:03 to 1:39
Exploration of the El Nino climate pattern and its global impacts.
“Coming up, a supercharged El Nino brews in the Pacific, but what do we need to know?”
El Nino Mechanism Explained
1:39 to 4:04
Detailed explanation of how El Nino affects weather patterns worldwide.
“It can raise global temperatures and cause extreme weather, such as droughts in Australia to floods in Peru.”
Current El Nino Conditions
4:04 to 5:30
Discussion on the current strong El Nino and its expected peak conditions.
“So what is happening with El Nino at the moment?”
Global Weather Impact of El Nino
5:30 to 7:20
Impact of El Nino on global weather, including droughts and storms.
“But other organisations around the world are predicting even warmer than that, four degrees or more of warming.”
El Nino and Climate Change Research
7:20 to 9:39
Research on the relationship between El Nino and climate change.
“We're not bordering the Pacific, so we are subject to other influences from around the world.”
Unprecedented Climate Conditions Ahead
9:39 to 10:45
Discussion on future implications of an unprecedented strong El Nino.
“few centuries or even out to a thousand years.”
At-Home Bladder Cancer Test
10:45 to 11:28
Introduction to a new urine test for detecting bladder cancer.
“A new study has found that at-home urine tests can detect more than 9 in 10 cases of bladder cancer.”
How the Bladder Test Works
11:28 to 13:42
Explanation of the process and advantages of the bladder cancer test.
“Well the urine's in contact with the bladder cancer cells growing in the bladder and some of those cells release DNA into the urine.”
Future of Cancer Testing
13:42 to 14:01
Discussion on the future improvements and potential for similar tests.
“if you have a positive galleus bladder test result, you must urgently have the camera inspection.”
Show all 16 chapters
Urine Test for Bladder Cancer
14:01 to 16:10
Learn about a new urine test for bladder cancer and its potential future developments.
“I mean nine out of ten is still an incredible success rate.”
Urine Test for Bladder Cancer
16:22 to 17:15
Learn about a new urine test for bladder cancer and its potential future developments.
“have to sort through 12 conflicting opinions online.”
NASA's Nancy Grace Roman Telescope
17:15 to 24:20
Discover the capabilities and science goals of NASA's new telescope.
“Still to come, the Harvard scientist who has developed an app to help the visually impaired get around.”
Mobilio: Navigation App for the Visually Impaired
24:20 to 28:00
Learn how the Mobilio app uses AI to assist the visually impaired in navigation.
“Here to tell me more is Raymond Liu, first author of the study and a PhD student at Harvard University.”
Navigating Safely: The Role of Technology for the Blind
28:00 to 31:57
Learn how a new navigation app for the blind enhances safety and independence.
“sounds in the environment, which is very important for a blind user to be able to hear.”
Navigating Safely: The Role of Technology for the Blind
33:06 to 33:31
Learn how a new navigation app for the blind enhances safety and independence.
“On the show, you get research-backed insights and clear pros and cons, whether you're planning a big purchase or you just want to grow your wealth.”
Transcript
Automatic transcript. May contain errors.0:00Hey, this is Farnoosh from the So Money Podcast, and this episode is sponsored by Smart Travel, a new podcast from NerdWallet. You know that one friend who always finds the best travel deals, picks the right cards, and somehow ends up in first class for the price of coach? Smart travel is like that friend, but in podcast form. They cover all things travel finance, like which travel cards actually pay you back and which just make your wallet heavier for no reason, how to book flights without turning it into a part-time job, which lounges are worth it, and which are just an airport gate with better snacks and worse lighting.
0:37Travel smarter and spend less with help from NerdWallet. Follow Smart Travel wherever you get your podcasts.
0:47All engine running. Absolute genius. Get this. Welcome. Welcome. This is the show where we bring you science. What that essentially means is discovery. Advances. Research. Technology. Unbelievable. Without further ado, this is The Naked Scientist. Hello and welcome to the Naked Scientist podcast, the show that brings you the biggest breakthroughs and talks to the major movers and shakers in the worlds of science, technology and medicine. With me, Rachel Ralph. Coming up, a supercharged El Nino brews in the Pacific, but what do we need to know? Also ahead, we hear from the team behind an at-home bladder cancer test and a navigation app to help those who are visually impaired.
1:38El Nino is a climate pattern that affects weather worldwide. It can raise global temperatures and cause extreme weather, such as droughts in Australia to floods in Peru. However, the El Nino, which is now forming over the Pacific Ocean, could be the strongest in living memory. A recent study in the journal Science found that El Nino is getting stronger because of global warming. By studying ancient corals in the Galapagos, researchers discovered that temperature changes linked to El Niño have increased by more than a third in recent decades, which is much more than in the past. Here's Aidan McGiven, a meteorologist and senior weather presenter at the Met Office.
2:21Well, El Niño is a naturally occurring climate phenomenon, usually involving warm water across parts of the tropical Pacific Ocean. In a typical year, Easterly trade winds across the tropical Pacific push warm waters all the way towards Indonesia, where those warm waters pile up and cause a lot of thunderstorms and heavy rain through Indonesia and northern Australia. In an El Nino year, those Easterly trade winds weaken. So much of that warm water sloshes back towards central and eastern parts of the Pacific and spread out. This is a vast amount of seawater. It's a huge ocean covering a third of the planet.
3:05So it's an incredible redistribution of heat and thunderstorms from western parts of the Pacific towards central and eastern parts of the Pacific. And that massive redistribution of heat causes big consequences, especially across the Pacific area and countries bordering the Pacific. but elsewhere across the world as well we see weather impacts from El Nino and ocean impacts as well. El Nino was discovered centuries ago by Peruvian fishermen because they noticed that the usual upwelling of cold waters off the coast of South America occasionally weakened, the waters warmed and the fishing deteriorated in certain years.
3:48It typically peaked around Christmas which is why it was named El Nino de Navidad, the Christ child. So that's why it was named El Nino, because of its timing and its impact on fishermen off the coast of South America. But we're seeing something quite different to normal right now, aren't we? So what is happening with El Nino at the moment? The way that scientists monitor El Nino is through a region called the Nino 3.4 region, which is basically a rectangular area of the tropical and central Pacific, about twice the area of India. And when sea surface temperatures there are half a degree above average for the time of year, and that heat is sustained for several consecutive months, El Nino is declared.
4:35That happened earlier this summer. And those sea surface temperatures are now well above average. We're talking about the latest three-month average, May, June, July, being 1.39 Celsius above average. And the most recent weekly reading gave a 2.6 Celsius anomaly. Now, to put that into context, a strong El Nino would be considered one to two degrees above average and the strongest on record 2015-2016 El Nino dubbed Super El Nino that peaked at 2.6 Celsius above average and this El Nino hasn't peaked yet it's likely to peak later this year late autumn early winter so there's more warming to come and what the Met Office are predicting is that it will peak at three degrees above average that would make it the strongest in living memory.
5:30But other organisations around the world are predicting even warmer than that, four degrees or more of warming. So this is being dubbed the strongest Elm year ever recorded and perhaps even further back than that for centuries. Wow, I mean, those numbers are insane. We're seeing a lot in the news, like you mentioned Indonesia, there's the wildfires across Kalimantan, to the Indonesian part of Borneo. And there's toxic haze smoke now in Malaysia and the Philippines because of this. And I think I read this morning that in Jakarta, the capital of Indonesia, that they haven't experienced any rain in a significant period of time to the point that they're interested in doing cloud seeding, but the clouds aren't in the right sort of state to do that.
6:14But what impact is this super Anino around other parts of the world? Essentially, what this does is it creates opposite weather and climate impacts compared with normal. So Indonesia, typically lots of rain turns much drier. Likewise, for northern Australia, we see hotter, drier weather going into the Australian summer, especially for northern Australia, compared with typical, and that can cause a greater risk of wildfires. On the other side of the Pacific, it's normally dry for Peru and for Chile. And those parts of the world get wetter because of the redistribution of heat and thunderstorms across the Pacific.
6:54California can get very stormy weather through the winter because of an El Nino, especially Southern California. It can strengthen and redirect the storm track across the Pacific. And that can then have an impact on the UK's weather. So what we are expecting later this autumn, early winter is an increased chance of storminess for the UK because of a more active jet stream. It's only one of many factors that affect the UK's weather. We're not bordering the Pacific, so we are subject to other influences from around the world. But it certainly loads the dice for the UK in favour of more storms, more rainfall, second half of autumn into early winter.
7:35Meanwhile, other parts of the world where we do get a lot of rain, like India, the Indian monsoon, they typically turn drier and that can have an impact on food production and it can push up food prices around the world and another key thing and this is a positive actually with El Nino is that it reduces the number of hurricanes in the Atlantic we haven't seen a single named hurricane yet in the Atlantic and that's really unusual for this time of year hurricane season starts the start of June if that goes past the 11th of September that will be the latest first named hurricane in the Atlantic on record.
8:12In the Pacific, it has the opposite effect. We are at the moment seeing three named hurricanes in the Pacific. And that's unusual for the number of hurricanes. The last time we had three active hurricanes at the same time in the Pacific was during the last El Nino. So this is a real significant redistribution of energy and weather patterns and heat around the world. There's been some suggestions that the El Nino is getting stronger because of global warming. This is a really difficult question for scientists to answer because the instrumental record for El Nino only goes back to 1950 and this is a climate phenomenon that oscillates so dramatically decade to decade.
8:59It's actually a relatively short sample size when we about the instrumental record and how it changes with a warming climate. We do have sea surface temperatures that go back to the 1870s, so there's a slightly longer record for that. And there's this recent study in science that looks at a much longer time period, back a thousand years, and studies coral records than Galapagos. And that's a really important piece of research because that enables us to look at how sea temperatures have changed during a much longer period. That suggests that we have seen much stronger El Ninos recently compared with the past few centuries or even out to a thousand years.
9:42But it's one piece of the puzzle. It's an active area of research. And there are other pieces of research that look at computer modelling and how El Nino is likely to change in the future. And there are conflicting results from that. Some say that global warming will cause stronger El Ninos and more frequent El Ninos. Other pieces of research based on modelling suggest the opposite. So it's not a settled part of science just yet. What we do know is that sea temperatures around the world and the atmosphere are warmer than they've ever been because of climate change. And that will only amplify the effects of the upcoming El Nino.
10:21So in the next 12 months, we're facing an unprecedented strong El Nino. We're also facing a global climate that is warmer than at any time in the instrumental record. These are uncharted times. It's difficult to predict exactly how this will play out because we've never seen it before, these natural and human-caused impacts on the climate system. The Met Office's Aidan McGiven. A new study has found that at-home urine tests can detect more than 9 in 10 cases of bladder cancer. The research, which has been published in European Neurology Oncology, was led by the University of Birmingham and a diagnostics company, Nonicus.
11:03It's involved nearly 1 ,000 patients at seven hospitals in England and Scotland. And to tell me more is Doug Ward, the biomarker laboratory lead at the Bladder Cancer Research Centre at the University of Birmingham. Well, it's basically a test whereby a patient can simply pee in a pot, the urine gets sent to the lab and analysed and then the doctor will get an answer back from the lab saying whether that patient has bladder cancer or not. And how exactly does this test work? Well the urine's in contact with the bladder cancer cells growing in the bladder and some of those cells release DNA into the urine.
11:41So when the urine gets to the lab the DNA is extracted and then sequenced to look for the mutations which actually cause the cancer. So the test looks for mutations across 23 different genes which are all involved in the mechanism of how bladder cancer develops. And how long does it take from receiving the urine sample from the patient to finding out whether the patient has bladder cancer or not? So the turnaround time for the test I believe is seven working days and then the report is sent electronically to the doctor. And what are the advantages of testing like this? Well the main advantage is that the existing procedure, something called flexible cystoscopy, which is having a camera inserted into the bladder to look for tumours and this is very unpleasant for the patient.
12:35It's also quite resource intensive for the NHS because it needs a doctor and a nurse and equipment and a room. And also there can be quite long waiting times for these flexible cystoscopies at some hospitals. So the test is easier and maybe faster as well. With most things, there's advantages and some disadvantages. So what about the pitfalls of this test? Well, the main pitfall is, although we've got strong evidence now that it works very well and that it detects at least nine out of ten bladder cancers. It does miss some. Those that it misses are typically the very early and less aggressive cancers, but it's still not a perfect test.
13:17That said, very few clinical tests are perfect, and even the flexible cystoscopy, the camera into the bladder, gets it wrong quite often. It can both miss tumours, or it can misdiagnose things that aren't tumours as tumours. So I would say the main disadvantage is we don't yet have enough evidence that we can completely do away with the camera inspections of the bladder. We're being quite conservative and what we're saying is if you have a positive galleus bladder test result, you must urgently have the camera inspection. So the patients who really need the camera inspections get them faster and we're saying that if you have a negative galleus bladder result you should still have the camera inspection but it's less urgent so patients could be managed more effectively.
14:07I mean nine out of ten is still an incredible success rate. Do you think this test can be improved though to hopefully catch all of those cases? We're very happy with the test as it is but of course we have to evolve and develop and we're still doing research into how we can make the test even better and there'll probably be a version two coming along some years in the future. It's quite difficult because the very small tumours that we typically miss aren't shedding much DNA into the urine but we're definitely doing research into how we can make the accuracy even better than it is at the moment.
14:42And can similar tests be developed for other types of cancer? I know quite recently the NHS rolled out the cervical cancer home test kit for women who may have missed their cervical cancer tests at the doctor's surgery but I know it's slightly different because this is a swab test so how does this compare or can similar tests be developed for other types of cancer? All cancers will release some DNA either into the blood or into the urine and we're very interested to look further into urine to see whether it might be able to detect other urological cancers and gynecological cancers. The test will need adapting because the mutations which we find in different cancers are different.
15:24So we'll have to change the test slightly if we were to say to try and detect cervical or endometrial cancer. But we do think that that may be possible in the future. How would someone get their hands on this at-home urine test or bladder cancer test? So it currently has to be requested by a doctor. If it's private healthcare, Bupa and several other healthcare providers will reimburse for the test. And there are six or seven NHS trusts across England and Wales that are now using the gallius bladder test, but the majority of the NHS isn't. But we think it benefits patients and the healthcare system, and we would very much like to see it more widely implemented across the UK.
16:10That was Doug Ward at the University of Birmingham. Hey, this is Farnoosh from the So Money Podcast. Today's episode is sponsored by NerdWallet's Smart Money Podcast, where a team of trusted journalists does the research so you don't have to sort through 12 conflicting opinions online. On the show, you get research-backed insights and clear pros and cons, whether you're planning a big purchase or you just want to grow your wealth. They explain the why behind everything from investing and home buying to smaller habits like sinking funds, the strategy to make surprise expenses stop feeling like surprises.
16:47Make your next financial move with confidence. Follow NerdWallet's Smart Money Podcast on your favorite podcast app. The Naked Scientist podcast is produced in association with Spitfire, cost-effective voice, internet, and IP engineering services for UK businesses. Find out how Spitfire can empower your company at spitfire.co.uk.
17:14This is the Naked Scientist podcast with me, Rachel Ralph. Still to come, the Harvard scientist who has developed an app to help the visually impaired get around. But first, NASA has launched a next-generation telescope from Florida in a bid to explore dark energy, black holes and other mysteries of the universe. The telescope is called the Nancy Grace Roman, named after the agency's first chief of astronomy. It cost$4.3 billion to build and has a field of view 100 times larger than both the Hubble and the James Webb Space Telescopes. I asked Megan Argo, an astronomer and podcaster at the University of Lancashire, to tell us about the project and what we might learn.
17:58It's been described as the success of the Hubble but that's not really the case. It's a different instrument. It observes the sky in a different way and it gives us different capabilities to what Hubble has. So while it is going to do fantastic things, it's not a replacement for Hubble. They will work together and together they will give us more information and neither one of them can on their own. So it's an optical and infrared telescope, so it works in more than one part of the electromagnetic spectrum. And it's going to do some particular large surveys of the sky and it will have user time available as well so astronomers can propose observations that are done with it.
18:37There are two main instruments on board. One of them is an imaging and spectroscopy unit, and the other one is a coronagraph. So they're going to work in different ways, and their aims are different. So the imaging and spectroscopy device on board is going to do wide area images of patches of sky. And to give you an idea of just how wide field, the Hubble Space Telescope gives us very, very good resolution, so very, very small pixels on the sky, effectively. but over a very small field of view. So it gives us exquisite pictures, but of tiny patches of sky. So if you want to image something like the Andromeda galaxy, which is a beautiful nearby galaxy that covers about four degrees on the sky, you would need thousands and thousands of Hubble images to cover the whole thing.
19:21As if that wasn't exciting enough, there's also this coronagraph on board. And the coronagraph is a really interesting instrument. It's the most sophisticated coronagraph that we've ever launched into space. and the idea of a coronagraph is that it blocks out the light from a star which is quite bright relative to anything that's in orbit around that star and the idea is if you block out the light from the star it enables you to see the planets in orbit now to detect exoplanets planets around other stars we can do that quite easily if it's a large planet orbiting close to its parent star those are relatively straightforward to detect via other means but to detect planets that are much further away from their stars, which are quite interesting.
20:00And we need the full sample, we need a good range of planetary sizes and distance from their host stars to get the full picture. But to detect them is much, much more difficult via these other methods. But using a coronagraph to block out the light of the star, that gives us access to those wide orbit exoplanets much, much more easily. So there's a lot of science that this telescope is going to be able to do over the five years of its main mission. And a lot of astronomers are very, very excited to see what data it sends back. You've compared the Nancy Grace Roman telescope to the Hubble but how does it compare to other famous telescopes for example the James Webb telescope?
20:35With the James Webb again it's a smaller field of view so you get good resolution on a small patch of sky compared to Roman. James Webb will tell us about how structure in the universe first formed after the Big Bang which is one of the big questions that we don't entirely know exactly how that happened and then the Roman telescope will be able to give us more information on how the structure of the universe has evolved since that time by looking at large area surveys of galaxies in the more nearby universe. So they work together, they're complementary, they give you different views on the universe, and by combining the data you get a much more complete view of the evolution of the universe.
21:10You've described some things the scientists are planning to investigate, but is there a specific list of things per se that scientists are really keen to look at with the Nancy Grace Roman Telescope? Yeah, so there's the three main surveys that will be undertaken and they're in different patches of the sky and they have different aims. So there is a high latitude wide area survey, so high latitude in terms of the structure of the galaxy. So this is looking out of the galaxy towards one of the polar regions where there's less galaxy to get in the way, if you like the fewer foreground stars, so you can more easily see the distant universe.
21:43The high latitude wide area survey is the main one that's going to tell us about the evolution of structure in the universe. there's this technique called gravitational lensing and there's this idea called weak gravitational lensing where basically you look at the shapes of galaxies on a statistical basis over large areas of the sky and statistically you know what the average shape of a galaxy should be and by looking over wide areas and comparing what you actually see to what you expect from the models you can tell something about the distribution of matter in the universe and how the universe has evolved so that's one of the the key science goals for the telescope there's also a high latitude time domain survey.
22:19So this is a smaller area survey on the sky. And the idea is to take repeat observations of the same patch of sky over the five years of the mission. So I think the idea is to come back every five days or so and image the same patch of sky. And that will catch anything that changes on those timescales. So we'll be able to see things like supernova exploding stars, we'll be able to see neutron star mergers, we'll be able to see things that fall into black holes and the subsequent flare from the black holes as well. So again, lots and lots of interesting physics that we can do in there that requires time domain surveys.
22:52And then there's another time domain survey that, again, is a quite small patch of sky, but looking towards the galactic bulge to the centre of the Milky Way galaxy. And that will be a very quick survey. So they'll come back to the same patch of sky on a rapid cadence. So they'll be able to see things changing on short timescales. And that will look at hundreds of millions of stars, be able to tell us something about the properties of those stars and whether they are changing over time, but will also enable us to look for exoplanets via the transit technique, where planets pass between us and those stars and the light slightly dims.
23:26It will also help us with the coronagraph to look for wide separation planets, so planets that are orbiting a long way from their host stars as well, which are quite difficult to detect by any other means. It will also help us look, again, using this gravitational lensing technique, but this time gravitational microlensing. So when an object passes between us and a background star, you see a brief increase in the light from that background star, even though you can't see the foreground object, it's this gravitational pull that's doing the lensing. That will let us detect rogue planets, so planets that are not associated with a host star that have been kicked out of their planetary system somehow, and possibly isolated black holes as well.
24:06So lots and lots of really exciting science. And those are just the surveys that have been put together. there'll be lots of other stuff that astronomers will do with the data that isn't necessarily one of the key science goals for these surveys so lots and lots of science for for people to do over the next few decades that was the astronomer and podcaster Megan Argo now finally researchers at Harvard University have developed an app which will allow people who are blind and have low vision to move around independently it's called Mobilio and it combines machine learning algorithms and personalized audio cues to help the visually impaired to avoid obstacles, navigate paths and turn freely.
24:46Here to tell me more is Raymond Liu, first author of the study and a PhD student at Harvard University. So specifically relating to outdoor navigation, we surveyed over 100 people who were blind or visually impaired and we identified three main navigation challenges that any sort of assistive device would have to help with. And those are navigating to a destination with turn-by-turn directions, kind of like Google Maps. There's also keeping someone on the correct route, so preventing them from taking the wrong turn or walking the wrong way or walking off the sidewalk. And then there's being able to detect and alert the user of obstacles so they are able to avoid running into those.
25:31So what information is being inputted into this app you've developed to help guide someone who is visually impaired? So the app relies on whatever is in the smartphone. So the smartphone has a GPS. It has an inertial measurement unit that measures how the phone is moving. And then it has a camera, of course. And so we're able to use the GPS for these turn-by-turn directions, kind of like what Google Maps does. We use the camera with AI, specifically computer vision. So we have a computer vision model that takes in the camera image and outputs wherever it thinks there are sidewalks, crosswalks, roads.
26:12And so that helps us keep the user on the sidewalk without walking onto the road, for instance. And then we can also use either another computer vision model for obstacle avoidance or detecting obstacles, or we can use if the phone has a light detection and ranging sensor. You know, some iPhones have this sensor and basically it tells you the distance to the environment. And that is another way you can sort of tell whether there's something in front of the user. So this app uses sensors within the phone. It uses machine learning or AI as well, in combination with this and other models to bring all this information together for the user.
Read the full transcript
26:54Correct. So then what does this app do with the information and what does the user experience? Yeah, so with this information, so if we know roughly where the user is using the GPS, we know specifically where the sidewalks and crosswalks are from the AI computer vision model. And then we also know where are there obstacles in the environment around the user. We can say this direction is a sidewalk. we play spatialized audio feedback to the user, which consists of beeping sounds that play to the right ear, to the left ear, if the user should go to the right or to the left, for example. And these direct the user to go along the correct route, along the sidewalk without going out into the road.
27:40And then if there's an obstacle, we can direct the user to avoid that obstacle. So in regards to the beeping noises in the left ear and the right ear, I presume that the user will have to wear headphones? Yeah. And the safest option is to use open ear headphones. So there are these headphones that go over the ear, so they don't actually prevent the user from hearing other sounds in the environment, which is very important for a blind user to be able to hear. And so if a user uses the smartphone with open ear headphones, they can receive this spatialized audio feedback, which can direct them in very specific directions while also allowing the user to still hear sounds in the environment.
28:23Yeah, I was going to say that if the user is using headphones, they might not be able to hear any oncoming traffic or anything like that. But are there any other concerns for safety? I mean, for example, can the app or the phone detect a fast moving car quickly enough or other people walking in the way? Yeah, so this is definitely a concern for safety. You know, if there's someone walking your way, the app can detect that because the person is just a walking person or a moving person walking in front of you. A fast moving vehicle would be harder to detect. And so, you know, as a person is using the app, they should still use their acquired skills of orientation and mobility, still use the white cane for physical safety.
29:05You know, we don't want our app to be like a replacement for a white cane or a guide dog, but more as a supplement to assist with independent navigation. And so a person should still always be exercising safe navigation and not totally relying on the app for everything to do everything. And how effective is this app compared to using a cane or a guide dog alone? Compared to using, let's say, a cane, if you only had a cane, you wouldn't necessarily be able to travel to a new destination. If you had a cane or a guide dog, you could potentially travel a route that you've traveled before because you and your guide dog kind of know the route.
29:43If you have a cane and a guide dog, it can also help with avoiding obstacles, but sometimes there might be like a hanging tree branch that you could run into that maybe your dog doesn't see that our system could potentially help with. So we actually ran studies comparing someone using a white cane and our system compared to, for example, a white cane in Google Maps and our system was able to help people navigate much quicker and much more independently because people needed fewer corrections. They got confused less. And compared to just a cane, our system was also able to help people avoid obstacles much more effectively.
30:22Apps are constantly being updated and so forth. So are there any optimizations that need to be done or any plans to expand the capabilities of the app? Yeah, for sure. One big challenge I would say is being able to run this computer vision model on a smartphone and have it run fast enough for it to be usable in real time by the user, but also be accurate. That's one big challenge. And then another one is this app is kind of analogous to like self-driving cars in that it's, you know, you rely a lot on detecting where the roads and sidewalks are, detecting like obstacles and stuff. But the big challenge is that most of the research into doing computer vision in outdoor street environments is focused on self-driving cars.
31:07And so if you take a model, a computer vision model that was trained to be used for a self-driving car and you slap it on the phone, then as a person's walking around with this model, it's going to say that the sidewalk is actually a road. And basically there's this difference in the domains that's hard to overcome. And since there's so much focus on self-driving cars, vehicle perspective images instead of pedestrian images, that poses a really big challenge because there's not that much high quality data from pedestrian perspectives and so that makes it hard to train a really robust model for pedestrian use.
31:42Raymond Liu at Harvard. Whilst Mobilio is not yet available for public use, Raymond hopes in about one and a half to two years time, following further tests and improvements to the software, the app will be readily available for those who need it. Well that's it for today. Join us next Tuesday when we'll be navigating by taking driverless vehicles for a spin. But what's all the fuss with an autonomous bus? And what's the science that drives one? You can leave a review on whatever podcasting platform you use to get the show. These rankings really help our visibility and that matters, so thank you.
32:19And if you like what we make, you can support the team at thenakedscientist.com forward slash donate. It really means a lot and it helps keep the show on the road. I'm Rachel Ralph, and from the rest of the team here at The Naked Scientist, thanks for listening, and until next time, goodbye!
32:56Hey, this is Farnoosh from the So Money Podcast. Today's episode is sponsored by NerdWallet's Smart Money Podcast, where a team of trusted journalists does the research so you don't have to sort through 12 conflicting opinions online. On the show, you get research-backed insights and clear pros and cons, whether you're planning a big purchase or you just want to grow your wealth. They explain the why behind everything from investing and home buying to smaller habits like sinking funds, the strategy to make surprise expenses stop feeling like surprises. Make your next financial move with confidence.
33:30Follow NerdWallet's Smart Money Podcast on your favorite podcast app.




