Jia Li: Machine learning and artificial intelligence could transform health care and education



hi everyone thanks for being here it's my honor to be here too my name is jolly I'm the head of R&D at Google Cloud a I I've also been a researcher throughout my career these grows have allowed me to participate in every step that AI needs to go from vision to reality I have been fortunate enough to participate in algorithm research building datasets and shipping products in reality AI is an incredibly interesting and exciting technology and today I'm here to share some of the avenues that it could change life for millions of people to start with I like to talk about two very traditional industries education and healthcare as we know healthcare is a very complex field with a lot of challenges AI could have the potential to change the outcome from individual patients to the entire hospitals health care typically start with the patient's lifestyle AI could help to provide accurate guidance to their lifestyle diet etc based on their past disease history genetics and prescriptions etc etc it can also provide automated monitoring and early assessment of critical conditions associating subtle precursors of signals that could correlate to emergent critical conditions that human would not be able to detect when a hospital visit is sorry hey I can play additional rows to help provide deep insights during and before and after the patient and the doctors one-one session it could also help to ease the workflow for doctors by automatically transcribing the session and filling out paperwork III could even provide deeper assisting diagnose diagnosis so that our doctors could be able to provide sophisticated diagnosis once the diagnosis recommendation is made artificial intelligence can also help to provide further treatment strategies including change of lifestyle prescription search surgery etc all of them above when a long-term stay is necessary whether a surgical patient or a senior patient in senior care intelligence system can provide further help to reduce the burden of nurses and doctors making rounds it can help to predict abnormal signals such as falling and agitated movement etc imitation machine learning can also help the entire hospital to run much more efficiently patient triage will take multiple patients medical records and help to ensure care is carefully designed and distributed in some cases medical conversation agents will help the patients to understand their symptoms without leaving their house in the first place here I'd like to talk more about some of the new research that I've participated in specifically the thoracic disease identification and the localization research as some of you probably know diagnosis skill is a very delicate skill some of the even very tiny mistake could cause very severe consequences in fact 10% of patient deaths is related to diagnosis arrows and according to professor Kurt lungless at Stanford four percent of all radiology Co interpretations contain clinically significant errors this number is especially significant if we consider that over 400 million such medical interpretations are carried out each year in the United States alone so let's look at the chest x-ray this is a identification problem even further chest x-ray remains a significant radiology challenge really really artists have to invest significant effort to understand and go through every single radiology image in order to make diagnosis recommendation if we can have some AI assisted the tool for them to get more insights about the radiology image for example we could predict the normal area of some potential disease in a radiological image that will help them to ease the process and make the entire process much more efficient however we're facing a chicken-and-egg problem here first we know our radiologists are facing a lot of challenges and efforts in order to get through all the medical images in order to give their interpretations we want to invent an AI assisted a tool in order to make that entire process much more efficient but in order to do so we need to get additional data and ask her radiologists to label a lot of data to train our method to build models this goes back to the exact problem that we want to help the radiologists to ease so in order to solve this problem we try to turn to the open source NIH chest x-ray dataset this is a fairly large data set with over a hundred thousand radiology images each of the image is associated with about up to 14 label disease labels mind automatically from the report which is relatively easier to get and as you can see here less than 1,000 of the of the images has bounding boxes associated to them which would each of them would require a board-certified radiologist to label the bounding box and that will require a lot of effort so typically this kind of data set is not well suited for traditional supervised learning which would require a lot of detail the labeling data so towards this problem we come up with a novel approach by combining the holistic global infirm about the disease as well as the local detailed annotation and we're able to predict both the disease type for the based on the global information as well as the local local protected area and highlight where the abnormal areas disease types could be and the overall disease prediction and suspicious region highlights works much better than state-of-the-art machine learning approach we're just at the beginning of this direction and we're not alone there are many partners and customers who are leveraging Google cloud for example zebra medico are using Google cloud to analyze new scans and deliver insights to hospitals to inform clinical decisions at scale but there are still much more remains to be explored and innovated in this space hopefully in the future ours specialists can spend less time on repetitive and error-prone tasks by working together with AI assisted the tools another area that AI could help is education as we know education is another very traditional field that is facing a lot of challenges it needs to balance the need of students and teachers with the complexity of schools and resources AI could unlock a lot of unique potential solutions here to start with AI could help to ensure our students have was a very safe environment to study and prevent them from dangerous actions such as in fighting or any other dangerous activities so that our education educators can focus on teaching and artificial intelligence systems can help taken care of the rest so more potential would exist in the education experience itself artificial intelligence algorithms could help to customize courses that is personalized to each of the student based on their past experience strengths weakness and personal preferences etc it can also turn abstracting examples to be very vivid real-world applications and examples and it could help our teachers to scale up the efforts by doing automated homework and exams assessments so this kind of experience can repeat through the course over the course of a semester a year and even the entire education experience so that we can provide highly personalized experience to each of the individual students and best of all such technologies can be both applied to stem as well as the arts for example we can easily extend some of the technology to a student stance and violin performance so I've talked about how I could help potentially change healthcare and the education in the future what about the countless other businesses beyond healthcare and education the real power of AI can be felt when its power can be leveraged by every possible businesses but that's a very challenging problem as we know that machine learning development is a very complex and resource consuming process it will require investment and expertise in every single step of the machine learning developments collect the data design model term model into two model parameters and evaluate deploy it and finally update and the iterates of the entire process it will be challenging to for most of the businesses because of the over twenty twenty-one million developers only 1 million of them have data science backgrounds and even fewer like thousands have deep learning background how do we solve this problem we make we made some attempt towards the solution of this by introducing the oto mo technology oh we need to bring it to auto ml is the data that we want to label and predict and auto ml will handle everything from there it gives the opportunity for any business or organization who wants to create customized models with very limited machine learning expertise earlier this year we've introduced the auto ml vision product basically the idea is the customers can upload and bring their labeled images and Auto ml technology will generate a customized the visual recognition model based on the data that they want to predict here is an example let's see how we do we could do weather prediction weather image classification here there are ten over ten different kinds of clouds each of them indicates a different weather pattern if we use the generically trained visual models here is what we are going to get will be easy to predict there is sky and cloud but we won't be able to know what kind of weather or what kind of cloud there is now if we try to upload all these domain-specific training images to auto ml vision here is what we can get total ml vision can learn what specific clouds or whether it means and give the prediction here for example Sara's Cirrus here and Ottawa Amell is product that based on multiple advanced technologies including learning to learn neural architecture search transfer learning hyper parameter tuning and more now let's take a look at how about how our customers are using Auto MO Zoological Society of London is a very good example it is a nonprofit organization that use camera-trap to track the wildlife of UT population over the world but that will generate meanings of unlabeled images or them to manually label each of the image as one of the wild animal type so the logical Society of London has been closely cooperating with our team to shape the auto MO product and now they are able to automatically label different wild animal types by using auto mo and we're very excited the potential of AI should bring to the put the way we protect our wild animal another example is Disney Disney is an early adopter of machine learning and the cloud platform that changed the way they interact with their customers and they extend their ability of visual recognition to recognize product images using auto mo now they're able to automatically detect characters and brand animal elements such as logo and color schemes and by leveraging this ability they are now able to provide more relevant search results and product recommendations another example is tactile graphics for those who are not familiar with tactile graphics it is a special type of images designed for blind to understand content it is very challenging to design such graphics because it needs to be tuned without perspective needs to be very simple and clear so that the blind readers can understand the content without being distracted by other unnecessary details because of the challenge of designing it it's very different countries all over the world they are electing the these tactile graphics into repositories for reuse purpose however these repositories they're not connected so a group of researchers try to use auto ml to differentiate what is a good tactile graphics what is not and then we can search online and find cute tactile graphics candidates now content publishers for the blind can are able to find kids tactile can candidates for their readers to understand so auto limo is part of the effort towards the trend of democratizing AI the real meaning of it is not just about how powerful technology is it is about also about how accessible it is Otomo vision is just one of the feature that we've democratized and we've seen so powerful examples from Disney from zoological and of London from the tactile graphics search engine its impact we've seen that in Disney that we are able to enhance the retail experience of one of the world's largest retailers and we've been able to empower wildlife conservation in a scale that we've never been possible to do in the past and we've also helped to improve interaction with the blind auto ml vision it's just the beginning we're going to also extend this to more features such as speech natural language processing and translation and more to bring more of these features to other fields automail vision as a single feature can already do so much we're very excited to see what the next wave can unlock technologies like Auto ml point to an exciting future in which AI is available in to everyone in a format that is easy to use regardless of what kind of problems they want to solve but solving a concrete problem isn't enough yet it's important it's equally important to understand what kind of problems we need to solve and to understand what people need in business in academia healthcare entertainment and countless other fields that are driving our society today ai is an incredibly exciting direction and the most exciting about it it's its potential to make life better for all of us I hope that everyone of us can contribute to this effort to make AI even walk up impactful thank you great thank you very much – and so we have plenty of time for questions so please you know the drill raise your hand if you have a question and the mic will come through here yeah thank you hi um I had a question regarding so as the models are abstracted and even combined and this becomes more accessible what tools do you have for introspection on why and how a prediction was made so for example say a retailer wants to identify potential shoplifters very good question so basically in order to understand what kind of like technology we can offer to different users we are also trying to understand what kind of problems they want to solve right so in the case of retailer wants to identify shoplifter they will help us to define what is a shoplifter and we can help them to to come up with just the technology to help that hi my name is Samantha and I work at USA as a software engineer my question to you is I mean it's really obvious everyone in this room that you know the need for machine learning and and artificial intelligence are in our community is prominent but what are you doing or how do you build a product that recognize the complexity around these type of techniques and you know you're going for education and scalability with these projects up our product so that everyone can utilize this technique but what are you doing to mitigate the risk of the misuse of these techniques and the misuse of these products right because I mean we've heard it today from Latonya and I think from Daniela like there's a huge risk and using this these techniques and the need for basic statistics etc is obviously prominent so what are you doing to kind of mitigate that when you're building products for widespread use that's a very good question I think as technologies and researchers this is very the question for us to explore how and make sure how high technology can be used only for good purpose in fact at Google we have an internal team who are especially focusing on this kind of problem how to understand bias how to understand how to make sure there is no misuse of Technology I'll have to say where all of us are at the early stage this is some serious topic that we should all contribute and explore down the road hi great by the way thank you so a I in education and the arts for our children actually scares me especially when you're talking about horses tasks and even learning music tailored customized for each child's preferences and maybe even their biology as humans we get the challenge each other to think outside the box to dream to learn what we thought we can learn to become wiser what is Google's vision and promise around AI in education thank you wow that's a very big question so here I'm listening out some of the potential ai ai research that we could we could make education software more powerful to assist our teachers the goal is to help that with more intelligent system and the intelligent algorithms our teachers can focus on creative and less repetitive work and hope to maximize everybody's interest every student's interest and capability during this education experience oh just a second we have a mic there yeah mister hi I had a question about some data you showed early on in the slide where you're looking at chest x-ray images and you need label data I have a knife question but I've always wondered if you can just look across time to eventually when a patient did show symptoms of some disease you were trying to diagnose and then go back and say yes this patient did have this disease and use that as the label do you know if that's possible or if that's too ambitious that's definitely possible and it's a very good question we have been working closely with radiologists to understand what's their real need because in the field people are focusing on given a radiology image and trying to come up with a disease label and after we talked to many specialists they are telling us this is not what we want because we have so much other information that we can get either from the patient disease history or from other signals from different reports etc and it's more helpful to give us the indicator or some proposal a normal area that's eventually how we come up with the idea to give assisted the recommendation and trying to give some of the recommendation about an abnormal area in our research hopefully by leveraging the useful information come up from the assistant to the specialists combined with many other information with information source we have to come up with the best solution or decision in the radiology analysis so I must do in Business Analytics currently and since you're an expert in artificial intelligence and machine learning I was wondering what your experience has been in artificial intelligence potentially creating a feedback loop so in the example of a potential shoplifter for example if we're identifying what a shoplifter is that can create a feedback loop about shoplifters and in future state that could change so are these models dynamic what are some of the challenges that you've experienced with feedback loops in the different types of artificial intelligence intelligence studies that you've done exactly I think it's totally possible to keep create the feedback loop and feedback loop who would make it make any a AI system to be more powerful and effective some of the simple example has you some of you probably know the recommendation system so based on how many how many clicks you've links that you've clicked proposed by the previous out which AI system we can learn a better AI algorithm based on that and that's one simple example in a more mature direction but there are many other fields that we are still experimenting and trying to learn how much we can improve hi one last question yes thank you the lucky one so when you talked about a I assisted diagnostic in the healthcare industry there are other players in this industry particular IBM Watson who's had a lot of coverage in that space as a leader in the AI space in the machine learning and LP space can you you know tell us the different approach that Google has taken versus the other vendors what's the name area that you play compared to the rest of the players in the industry mmm a very good question I have less access to other companies solution but at Google we really focus on collaborating closely with our customers for example hospitals and specialists trying to understand what's the real need and try to bridge the gap between the technology and the real solution and you mentioned there are many players in this field I want to say in healthcare that's the field we want as many players as possible we want everybody to contribute to this space to help of our life to be better that is a very nice in political answer thanks very much again G I believe for talking us today

Leave a Reply

(*) Required, Your email will not be published