afghan officials say the taliban have seized the south—west city of zaranj — the first provincial capital to fall to the militants for five years. during a un emergency meeting to discuss the worsening violence, its envoy to the region demanded the militants end their offensive. nearly half the regions in greece are on high alert — as the worst wildfires in decades rage across the country. the government has issued a warning about a second wave of blazes as strong winds whip up many of the fires that were being brought under control. and the penultimate day of action at the olympics is under way in tokyo. kenya's peresjepchirchir has won the women's marathon title, and later medals will also be up for grabs in volleyball, ——basketball, volleyball and golf.

now on bbc news, click looks at how artificial intelligence could be used to help health services around the world. this week: a special programme. an artificial intelligence make healthcare better for all of us, and save the nhs? here in the uk, the national health service has been in crisis for many years. looking after an increasingly ageing population, with complex needs. fighting for resources, it has been a breaking point. and then... the pandemic hit. there

will come _ then... the pandemic hit. there will come a _ then... the pandemic hit. there will come a moment _ then... the pandemic hit. there will come a moment when - then... the pandemic hit. there will come a moment when no i will come a moment when no health service in the world could possibly cope. find health service in the world could possibly cope. and indeed many countries _ could possibly cope. and indeed many countries people - could possibly cope. and indeed many countries people 's - could possibly cope. and indeed many countries people 's health i many countries people �*s health systems have struggled with the sheer number of covid patients. even those which didn't have people spilling out onto the streets had to put all other treatments on hold. the nhs is an old health system. in fact, it is several different systems that sometimes work together and sometimes don't. and now, it is trying to reinvent itself and embrace technology to beat the queues of covid—i9 —— that covid-i9 has the queues of covid—i9 —— that covid—i9 has created. this is a story i started filming just before the first uk lockdown, when we were not yet wise to masks or social distancing, but there was definitely a hint that something was coming. i wanted to find out how artificial intelligence could be used to help take up the strain that the nhs was already under. and on what

about to do something that i had never done before. i wasn't nervous until i asked my twitter followers what i could expect. very loud noise, claustrophobia. now i'm nervous. i'm going to have an mri scan on my liver. i have to lay absolutely still?— lay absolutely still? right. this magnetic _ lay absolutely still? right. this magnetic method --| this magnetic method —— magnetic residence imaging scanner, mri scanner will be able to see my tissues and fluid in great detail and produce an image like this which will be looked at by a radiologist to see if they can find anything unusual. the thing produces really strong magnetic field, which means nothing metal can come in stop that includes you. but this is no ordinary mri scan. instead of being read by a human, no ordinary mri scan. instead of being read bya human, my mri is going to be read by an artificial intelligence. this

is a computer that will look at the images and read them at least as well as a human radiologist would. breathe out and hold your _ radiologist would. breathe out and hold your breath. - radiologist would. breathe out and hold your breath. now- radiologist would. breathe out| and hold your breath. now this film is not _ and hold your breath. now this film is not about _ and hold your breath. now this film is not about computers . film is not about computers stealing jobs. this is a film about computers filling the gaps in an nhs that is short of money, short of nurses, and short of the highly skilled radiologists who can read and interpret images like this. but thatis interpret images like this. but that is a really difficultjob that is a really difficultjob that takes years of training. i mean, how could you teach a computer to do that? well, here is a classroom full of medical students, and this... is a computer. the way you teach eachis computer. the way you teach each is very different.- each is very different. what ou each is very different. what you have — each is very different. what you have in _ each is very different. what you have in front _ each is very different. what you have in front of- each is very different. what you have in front of you - each is very different. whatl you have in front of you here are ct scans of a human along

with cancerous tumours. i will be teaching you how to identify them and hopefully save lives. the human brain is brilliant at learning things. it can understand spoken words... it can understand diagrams. just through the teacher's descriptions, these students can have a good guess at finding tumours in these images. they get some right... they get some wrong. with repetition and practice, their brains make more and more connections. which strengthen with success. until eventually, we really understand the task in front of us. and we can do it well. but computers are dumb. they literally know

nothing. they can't look at a diagram and imagine it in real life, they can't understand spoken words. normal teaching methods won't work. so instead of trying to describe to them what we want them to learn, we teach them using trial and error. millions of trials and facts. —— in fact. the computer starts by circling completely random parts of the image. it just guesses, it doesn't even know what it's getting at. all it knows is when its gas is right and wrong. and mostly because it is getting, it is going to be wrong. and the great thing about computers is they can do this over and over again, really fast, and they remember everything. and every time it does accidentally get it right, it makes a connection

to its previous correct guesses. with every right answer, certain connections get stronger. and with every wrong answer, others get weaker. through this barrage of guesses and network grows, similar to the human brain, which starts to distinguish between right and wrong. a digital thought process that we call a neural network. until eventually, after a multitude of attempts, it has very few failures, and a lot of success.— lot of success. ok, let's try some more _ lot of success. ok, let's try some more difficult - lot of success. ok, let's try some more difficult scans l lot of success. ok, let's try - some more difficult scans now. so if we just take your gown offm — so if we 'ust take your gown off... �* , , off... and this is the technique _ off... and this is the technique that - off... and this is the technique that could j off... and this is the - technique that could help to relieve the pressure on the uk's rest screening service. ——

breast. almost 1000 women die from breast cancer in the uk every month. the aim is to catch the disease early when it is most treatable. that means that more than 2 million women have their breasts scanned for potential cancer every year. mi; potential cancer every year. my entire role _ potential cancer every year. ij�*i entire role is potential cancer every year. ij�*i: entire role is clinical diagnostics within breast. you know, it takes up a great deal of my life, it is really important, and we are kind of fighting a big fight. bernadette works at the lincolnshire breast screening service, and she is part of a strengthening —— shrinking workforce. the years the nhs has been unable to train or even recruit enough radiologist, and many services now face chronic staff shortages. this is further obligated by the fact that each mammogram needs to be read independently by at least two specialist clinicians. so

independently by at least two specialist clinicians.— specialist clinicians. so we have two _ specialist clinicians. so we have two people _ specialist clinicians. so we have two people reading l specialist clinicians. so we have two people reading aj have two people reading a mammogram, because it is actually a really difficult process, 0k? what looking for is tiny, tiny cancers. maybe two millimetres, maybe a little tiny smudge within a breast. and we are human, some of us miss them. so that second read is that second opportunity to pick up that tiny little smudge that may, you know, change a woman's life. 50 that may, you know, change a woman's life.— that may, you know, change a woman's life. university hospital has been trialling a new tool that may help. this is mere, and ai trained to spot breast cancer. —— mia. the aim of this project is for mia to be the second reader, speeding up the whole process. so mia has had a look at the programme —— the

mammogram as well, can we have a look at what mia thinks? so in this a look at what mia thinks? sr in this particular case, mia has also marked the area that i would be concerned about and it makes a callback decision itself and as you can see it says callback, malignancy. here in nottingham, _ says callback, malignancy. here in nottingham, jonathan - says callback, malignancy. here in nottingham, jonathan and his team have been testing mia on the hospital's historical data sets. these other women who have been to this clinic in the past and been diagnosed by a human. the aim is to see if mia would have made the same decisions. 50 would have made the same decisions.— would have made the same decisions. so this time, mia hasn't actually _ decisions. so this time, mia hasn't actually placed - decisions. so this time, mia hasn't actually placed any i decisions. so this time, mia i hasn't actually placed any mark on the image here, and its opinion is no recall is required. actually this lady did come back to some extra tests and this well—defined mass came back and this was assist, a perfectly harmless cyst. so the recall would be a false positive, this lady didn't actually need to come back. b. didn't actually need to come back. �* . , didn't actually need to come back. �* ., , ,

back. a false positive is when the reader — back. a false positive is when the reader thinks _ back. a false positive is when the reader thinks that - back. a false positive is when the reader thinks that there l back. a false positive is when | the reader thinks that there is cancer, but on further investigation it turns out that they were wrong. and over at they were wrong. and over at the cambridge breast unit, professor fiona gilbert has been conducting research to try and reduce those false positives. something that not only put a lot of strain on the system, but also a lot of strain on the patient. it is obviously _ strain on the patient. it is obviously causes - strain on the patient. it is obviously causes quite . strain on the patient. it is obviously causes quite a i strain on the patient. it 3 obviously causes quite a lot of distress to the woman being called back, and the majority of them turn out to be normal. it is a lot of work for us to be assessing all of these women for a relatively small number of cancers. for a relatively small number of cancero— of cancers. and when a programme _ of cancers. and when a programme screens . of cancers. and when a i programme screens over of cancers. and when a - programme screens over 2 million people every year, these little percentages do matter. every year around 70,000 women are given a false positive result, and some never attend a screening again. fiona and her team have been testing and her team have been testing and ai built by google health to see how the software compares to human radiologists. so in the retrospect of a study

that was done, the algorithm performed better than some of the individual radiologists. and worse in other radiologists. so when they took the average performance of all the average performance of all the radiologist, the algorithm would have called back fewer people. would have called back fewer --eole. �* , , .,. people. but this research ro'ect people. but this research project is _ people. but this research project is still _ people. but this research project is still in - people. but this research project is still in its - people. but this research project is still in its early | project is still in its early stages, and is set to move to the next phase of development this year. in the meantime, batting not —— back in nottingham, mia has finished its initial trial they will now be tested in 15 more nhs sites against ongoing cases to make sure that it works.— sure that it works. artificial intelligence _ sure that it works. artificial intelligence is _ sure that it works. artificial intelligence is the - sure that it works. artificial intelligence is the next - intelligence is the next amazing transformation, so that is what excites me. and it excites me that i don't know the full potential. we know that cancers were missing —— we are missing tend to be smaller, they tend to be more aggressive, and those other ones that we want to find. but

as the pandemic hit, all of this had to stop. and as breast clinics emptied out, elsewhere the nhs was about to be overrun. the coronavirus pushed intensive care units yonder their limit. but even before their limit. but even before the pandemic, these wards had been running at very near capacity. and it is here that one of the biggest killers in the uk, and indeed the world, lurks. it is called sepsis. so se sis lurks. it is called sepsis. so sepsis is — lurks. it is called sepsis. sr sepsis is when there is a severe infection, and when the body's response to that severe infection leads to organ failure. it infection leads to organ failure. .., infection leads to organ failure. , ., failure. it can set him without warning. _ failure. it can set him without warning. and _ failure. it can set him without warning, and in _ failure. it can set him without warning, and in fact - failure. it can set him without warning, and in fact anyone. l failure. it can set him without l warning, and in fact anyone. -- warning, and infactanyone. —— and effect anyone. warning, and in fact anyone. -- and effect anyone.— and effect anyone. this is a woman in _ and effect anyone. this is a woman in her— and effect anyone. this is a woman in her 50s, - and effect anyone. this is a woman in her 50s, she - and effect anyone. this is a woman in her 50s, she was and effect anyone. this is a - woman in her 50s, she was very unwell initially.— unwell initially. people are admitted _ unwell initially. people are admitted to _ unwell initially. people are admitted to m _ unwell initially. people are admitted to icu for- unwell initially. people are admitted to icu for many l

unwell initially. people are - admitted to icu for many reason but sepsis tends to strike when patients are at their weakest. we literally collect every heartbeat and every breath. when you first cut —— starting intensive care, when i started, it was difficult to see what was important because there was much data. but was important because there was much data-— much data. but remember, computers _ much data. but remember, computers love _ much data. but remember, computers love data. - much data. but remember, computers love data. and l much data. but remember, i computers love data. and they love spotting patterns in it. they don't get tired, and they never stop. so the doctors and computer scientists teamed up to create something called ai clinician, an algorithm that can detect and even predict early signs of sepsis, and then advise how to stop it.- early signs of sepsis, and then advise how to stop it. what we have used _ advise how to stop it. what we have used is — advise how to stop it. what we have used is a _ advise how to stop it. what we have used is a large _ advise how to stop it. what we have used is a large database. have used is a large database with 20,000 patients, and then we tested it in another 80,000 patients. that is more than any doctor could ever see in their lifetime. ~ , ., lifetime. were proving successful, _ lifetime. were proving successful, but - lifetime. were proving| successful, but neither lifetime. were proving i successful, but neither the ai successful, but neither the al or the humans were ready for a

new type of sepsis, one brought on by covid—19. new type of sepsis, one brought on by covid-19._ on by covid-19. there are few differences. _ on by covid-19. there are few differences. -- _ on by covid-19. there are few differences. -- a _ on by covid-19. there are few differences. -- a few- differences. —— a few differences. —— a few differences. 0ne differences. —— a few differences. one is the intense information they had in the long made the oxygen levels really low and so they needed high amounts of oxygen. if they didn't tire of their breathing so quickly as many patients with other pneumonia often struggle with their breathing, we also saw things like a lot of blood clots, which we occasionally see in other sepsis, but were far more frequent in this disease. nearly all the patients coming through to icu now have this new sepsis, and for the ai, this was an illness it had never seen before, nor had any data on. and remember, just because a computer has learned to be good at one task, doesn't mean it can do a different ones. the ai clinician was back to square one. but the humans were not. their idea was to

gather data from icus across the uk to create a database of treatment that worked so the doctors in the thick of it could learn from each other at speed and save more patients. for the first time we built a picture of how disease was treated in real—time on the database across the country. you can see the diversity of the different approaches the people are trying, what was working and what was not working. working and what was not working-— working and what was not workinu. ., , ., ., , working. now this data will be used to retrain _ working. now this data will be used to retrain ai _ working. now this data will be used to retrain ai clinician i used to retrain ai clinician and one day it may help to treat sepsis caused by covid. so it is about 18 months since we started making this programme and in that time we have all started to use technology in new ways and some of that has included gp consultations over video call, for example. so i think it probably means that many of us will be happy to accept medical

help from machines and computers in the future. but for those computers to be good at theirjobs, they for those computers to be good at their jobs, they will need for those computers to be good at theirjobs, they will need a lot of data. a lot of our personal medical data. many services like instagram, google and facebook do not charge us any money. instead they get to use our personal data to try and work out how better to target us with ads. for many of us that seems like a fair exchange. they make our lives easier and they connect us to the world. but now, those same companies that have been collecting and profiting from our personal data are moving into healthcare. amazon, microsoft and google are all trying to grab a seat at the table and dine on our medical data. and the nhs has something

unique to offer. a whole nation's information from birth to death. that's something that might make one feel uncomfortable. i might make one feel uncomfortable. ., �* ,, uncomfortable. i don't think many health _ uncomfortable. i don't think many health professionals l uncomfortable. i don't think. many health professionals sit down with their patients and say do you want your data to be sold to google. i don't think those conversations are being heard. ~ �* ., , ., ,, heard. we're not 'ust talking about truth heard. we're not 'ust talking about the big i heard. we're notjust talking about the big household i about the big household technology names either. many other companies that you never have heard of also want access. it is an interesting and controversial issue. there are some people who believe it is an asset to the nhs and we should sell it, others who believe that in some way or another that is not a good thing to do from a moral point of view. �* of view. but if we get it right and it is a — of view. but if we get it right and it is a big _ of view. but if we get it right and it is a big if, _ of view. but if we get it right and it is a big if, the - and it is a big if, the benefits could be huge. a unified joint app store of medical data could really speed up medical data could really speed up research and lead to new treatment. and ai would be used

to look for the patterns hidden within. and that is something that could soon be helping those who are suffering from long covid. those who are suffering from long covid— long covid. thanks to artificial _ long covid. thanks to artificial intelligence | long covid. thanks to i artificial intelligence we long covid. thanks to - artificial intelligence we were able to link patient records with imaging and produce an analysis of 41,000 patients, imaging and clinical associations in about two months. associations in about two months-— associations in about two months. �* ., ., , ., ., months. adriana is part of a team that — months. adriana is part of a team that has _ months. adriana is part of a team that has been - months. adriana is part of a team that has been looking| months. adriana is part of a i team that has been looking into why long covid only affects some people. why long covid only affects some people-— why long covid only affects some people. why long covid only affects some --eole. ~ ., ., .,, some people. what we found was uuite some people. what we found was quite important — some people. what we found was quite important and _ some people. what we found was quite important and shocking i quite important and shocking because it was not obesity per se was fatty liver disease linked to obesity that increased the risk up to five times of being severely affected by covid—19. times of being severely affected by covid-19. this is one of almost _ affected by covid-19. this is one of almost a _ affected by covid-19. this is one of almost a million i affected by covid-19. this is i one of almost a million people suffering from long covid in the uk. he is 40 and a doctor

and he moved to the uk from india to complete his specialist training. in the last week _ specialist training. in the last week of _ specialist training. in the last week of march i specialist training. in the last week of march on i specialist training. in the | last week of march on the weekend i was on weekend duties are not feeling well. i went back home immediately, i self isolated and then i tested positive. i was not thinking that this virus will affect me because i was a fit adult and i thought it would just be ok, like any other flu. there was a moment when i thought i could just die and i had a three—year—old child and i would not be able to see her. i never knew that this virus could leave me or others with so many problems and disabilities. 50 so many problems and disabilities.— so many problems and disabilities. ., , ., disabilities. so to understand his illness — disabilities. so to understand his illness more _ disabilities. so to understand his illness more he _ disabilities. so to understand his illness more he signed i disabilities. so to understand his illness more he signed up to a study aiming to help long covid patients. a trial that

uses the very same ai that lowered me into that big magnetic tube earlier. this ai had originally taught itself how to spot certain liver characteristics as well as spotting unusual masses it also learn to assess the overall health. and here is what made of mine. , w' health. and here is what made of mine. , a , health. and here is what made of mine. , , ., of mine. here we picked up a small area — of mine. here we picked up a small area which _ of mine. here we picked up a small area which is _ of mine. here we picked up a small area which isjust i of mine. here we picked up a small area which isjust a i small area which is just a cyst, a simple fluid—filled sack. cyst, a simple fluid-filled sack. ~ , w' cyst, a simple fluid-filled sack. ~ , ., , sack. in the mri pick that up as something _ sack. in the mri pick that up as something unusual, i sack. in the mri pick that up i as something unusual, something of interest?— of interest? yes. but there is no concerning _ of interest? yes. but there is no concerning feature. i no concerning feature. 0therwise no concerning feature. otherwise it is a reassuring scan with no increased evidence of inflammation or increased risk of fat. of inflammation or increased risk of fat-— risk of fat. but here is the thin. risk of fat. but here is the thing- it _ risk of fat. but here is the thing. it turns _ risk of fat. but here is the thing. it turns out - risk of fat. but here is the thing. it turns out that i thing. it turns out that because this ai can assess the overall health of the liver it can do the same for other organs as well.— can do the same for other organs as well. and so we expanded _ organs as well. and so we expanded first _ organs as well. and so we expanded first to - organs as well. and so we expanded first to organs l organs as well. and so we i expanded first to organs such as the kidneys and pancreas, spleen and heart. 0riginally with the aim of looking at

diabetes but many of those organs were being affected by covid as well. if it was a radiographer manually analysing data from six different organs that could take up to 24 hours of work. this allows us to basically automate some of the most labour—intensive parts of the process. we most labour-intensive parts of the process-— the process. we are trying to identify which _ the process. we are trying to identify which kind _ the process. we are trying to identify which kind of - the process. we are trying to identify which kind of fat i the process. we are trying to identify which kind of fat is i identify which kind of fat is the one _ identify which kind of fat is the one that puts you at risk. it the one that puts you at risk. it is — the one that puts you at risk. it is not — the one that puts you at risk. it is not obesity as we thought up it is not obesity as we thought up until— it is not obesity as we thought up until now what makes you at higher— up until now what makes you at higher risk— up until now what makes you at higher risk is the visceral fat. _ higher risk is the visceral fat. the _ higher risk is the visceral fat, the fat in the organs. whereas— fat, the fat in the organs. whereas we have found people with low — whereas we have found people with low bmi with high—fat in the liver~ _ with low bmi with high—fat in the liver. a with low bmi with high-fat in the liver. �* g , with low bmi with high-fat in the liver. �* g , ., the liver. a jays part of the clinical trial _ the liver. a jays part of the clinical trial that _ the liver. a jays part of the clinical trial that may i the liver. a jays part of the clinical trial that may show | clinical trial that may show that young fit people with a normal bmi can also get long covid, if they have fat in their organs.— covid, if they have fat in their oruans. �* ., , ., their organs. and not everyone is aware of _ their organs. and not everyone is aware of what _ their organs. and not everyone is aware of what long _ their organs. and not everyone is aware of what long covid i their organs. and not everyone

is aware of what long covid is. | is aware of what long covid is. i think eventually when the acuteness went off i think more and more people will come up with long covid.— and more people will come up with long covid. there are many trials of the _ with long covid. there are many trials of the country _ with long covid. there are many trials of the country trying i with long covid. there are many trials of the country trying to i trials of the country trying to understand long covid and artificial intelligence has become a big tool in the race to find a treatment. but for many, help cannot come quick enough. technologies that bring about big change also bring new problems. there is no point in denying that. but right now, ai is changing the world and opening up possibilities for huge medical breakthroughs. it helped us to develop highly effective vaccines in record time. so, i know the pandemic is farfrom over time. so, i know the pandemic is far from over but i still feel kind of lucky that it happened now rather than even just a decade ago. and maybe it will mean the nhs is a bit more ready for the next one.

hello there. the very unsettled august weather continues into this weekend. low pressure nearby will generate further showers, and again, like friday, we could see some thunderstorms which could lead to some localised flooding in places. but there will be some good spells of sunshine in between, particularly across more southern areas. so, here it is, this area of low pressure, which is going to stick around both saturday, sunday and indeed even into monday. lots of isobars on the charts, so it'll be quite breezy again, particularly across southern, south—western areas, and across the northern isles, gusts of 30—40 mph. we'll have showers pretty much from the word go anywhere, but most of them will be across scotland, northern ireland, western england and wales. they will drift their way further eastwards into the afternoon, and again there'll be some torrential downpours in places. but some good spells of sunshine, particularly

across the south—east. another breezy day, these are the mean wind speeds. temperatures will be a bit disappointing for august, particularly when the showers come along, it'll feel quite cool. but in the sunny spells, we could make 20 degrees or so across the south—east. 0therwise, generally the mid to high teens celsius. as we head through saturday night, we continue with the breeze, further showers. again, some of them will be quite heavy, particularly across central, northern and western areas. perhaps turning a little bit quieter across the south east quadrant of the country. and again nowhere particularly cold, with overnight lows 12—14 degrees. so, into sunday, ourarea of low pressure still with us, drifting a little bit further northwards and weakening a little bit. there's fewer isobars on the charts, but there's still enough energy in the low pressure system to generate further showers, which again could be quite heavy in places throughout sunday. mainly across central and northern parts of the country, because i think as we head on into the afternoon, there may be a greater chance of seeing some sunnier, drier weather for wales, central and southern england.

so, that mayjust bump up temperatures to 21 degrees, slightly lighter winds. again, for most, though, the high teens celsius. into next week, then, for monday, our area of low pressure's still with us, so it's going to be another day of sunshine and showers. but the winds will turn light, and the system continues to weaken. and as we head on into tuesday, we've got this bump of high pressure which will build in, and that should settle things down. but low pressure always close by to the north and the west of the uk. so, we'll have most of the showers through the new week across northern and western areas. greater chance of seeing some drier, sunnier and warmer weather in the south and east.

ones. the ai clinician was back to square one.

this is bbc news, i'm rich preston. our top stories: gunfire. the taliban seizes a provincial capital in southern afghanistan, as the un's envoy demands the militants end their offensive. we are extremely concerned about the safety and security of people in cities under taliban attacks, and what reality would await them. —— brutality. nearly half the regions in greece are on high alert, as the worst wildfires in decades rage across the country. i am sarah mulkerrins, live in tokyo, day 15 of the olympic games where kenya's peresjepchirchir has won gold in the women's marathon. and the remarkable spectacle

of south africa's sardine run and the predators it attracts.