Artificial Intelligence

As artificial intelligence’s (AI) sprawling influence reshapes industries from logistics and healthcare to automotive and manufacturing, Taiwan is poised to leverage its cutting-edge capabilities and rich history in semiconductor manufacturing to stake out a leadership position in AI. Taiwan’s semiconductor manufacturing industry accounts for a major share of the region’s GDP and, with its manufacturing prowess, the region is fertile ground for using AI to optimize and even revolutionize chip manufacturing. In an AI and Semiconductor Smart Manufacturing Forum recently hosted by SEMI Taiwan, experts from Micronix, Advantech, Nvidia and the Ministry of Science and Technology of Taiwan (MOST) shared their insights on how deep learning, data analytics and edge computing will shape the future of semiconductor manufacturing. Here are four key takeaways.1. Monitor, Forecast, and PreventToday, tier 1 foundries use AI tools to combine equipment know-how and manufacturing statistics in managing massive Fault Detection (FD) data, much in the way that a car’s tire-pressure monitoring system helps maintain safe inflation levels and prevent accidents. For example, AI enables the real-time collection and monitoring of massive amounts of processing data, then alerts system administrators of any hardware failures or other manufacturing abnormalities.AI also makes it possible to adopt Run-to-Run (R2R) control to automate manufacturing process adjustments and corrections by providing feedback that can drive higher processing efficiency. In addition, virtual metrology replaces manual sampling inspection for comprehensive quality control, enabling foundries to improve yields, reduce costs, and strengthen their competitive advantage.2. Beyond Automation: Edge Computing The evolution of IoT is giving rise to a paradigm shift in the industry as the recognition grows that smart factories must go beyond automation to focus also on intelligence. All information – from equipment status and manufacturing process statistics to on-site environmental data – needs to be collected through sensors. In highly time-critical scenarios, returning all sensor data to the cloud for processing is time-consuming and impracticable. This is where edge computing’s real-time features and lower cost than cloud computing come into play.How does edge computing work in a smart factory? First, a rich trove of data from various devices is collected and integrated via Manufacturing Execution Systems (MES). Software analysis then produces a real-time factory production status before production data is visualized through a combination of system platforms and human-machine interfaces. In the end, the data is analyzed realtime in the cloud so failures can be predicted and prevented to help increase capacity and reduce costs. The approach is even capable of Bill of Materials (BOM) predictions, allowing better collaboration between upstream and downstream suppliers.3. Deep Learning Accelerates AI Deep learning enables autonomous driving, intelligent voice assistance and many other AI breakthroughs. The heart of deep learning is its ability to automatically process and learn data in various formats such as images, video and text with no human domain knowledge. This increases predictive accuracy and efficiency in processing massive amounts of data. Deep learning also enhances the efficiency of human-machine collaboration.4. Taiwan’s Competitive Niche: Industry 3.5Industry 4.0 is not just about improving production management. It also focuses on integrating supply chains, even among competitive companies. For Industry 4.0 to thrive, rival companies must grow together. The first and third industrial revolutions centered on disruptive technologies like steam engines, transistors and digital, while the second and fourth revolutions homed in on competition among various business models, platforms and industry ecosystems.While Taiwan’s strengths include innovation, short time-to-market, low manufacturing costs, and high supply chain management efficiency, the region still lags advanced countries in basic industry and research capabilities. Squeezed by Chinese supply chains and high-end manufacturers in advanced countries, Taiwan should start by carving out an Industry 3.5 niche for the island’s manufacturers. SEMI will continue to facilitate cross-industry connection, collaboration and innovation to help manufacturers seeking higher production efficiency and lower costs incorporate AI as a core competitive advantage. At SEMICON Taiwan 2018, SEMI will unveil its Smart Manufacturing Journey, an exhibition that gathers leading AI companies such as ABB, Advantech, Nvidia, Sony and UPS to demonstrate a comprehensive roadmap for smart manufacturing technologies and applications. For more information, please visit the SEMICON Taiwan website.Emmy Yi is a marketing specialist at SEMI Taiwan.

Part 1 of this two-part piece explores best-practice perspectives on collecting and utilizing smart data in industries outside semiconductor manufacturing, one of the important takeaways from the Smart Manufacturing panel discussion at SEMI ASMC 2018. Part 2 examines the potential benefits to be realized by pairing human Subject Matter Experts with smart silicon assistants, and what these new arrangements mean for semiconductor device manufacturing. The spacecraft Discovery and its HAL 9000 computer system had a digital twin. Did you know? Stanley Kubrick’s seminal film “2001: A Space Odyssey” had its theatrical release 50 years ago this April. “2001” isn’t just a great science fiction film. Rather, it’s a great work of cinema overall, across any category. (The American Film Institute lists “2001” as #15 in the AFI Top 100; a bit below “Vertigo,” a bit above “It’s A Wonderful Life.”) It’s a film so distinguished and so prescient that its lessons can inform our thinking about smart manufacturing, Industry 4.0, and artificial intelligence (AI) today. Not to give too much away, but the earth-bound digital twin of Discovery / HAL identifies a diagnostic error the onboard, Jupiter-bound HAL 9000 has made, things go awry from there, and one of the mission pilots, astronaut Dave Bowman, is forced to intervene. At the recent SEMI Advanced Semiconductor Manufacturing Conference, ASMC 2018, on 02 May 2018 in Saratoga Springs, NY, five diverse panelists representing capital equipment, IDMs, academia, the semiconductor supply chain, and smart manufacturing best practices outside the semiconductor industry engaged in a lively discussion with the ASMC attendees. They explored where “smart” is in our industry today, where it’s headed, and what that’s going to mean for us -- the professionals who have brought semiconductor manufacturing to the current state of smart, and are looking to implement an ever-smarter tomorrow. Not to give too much away, but the panelists and audience agreed that there’s nothing artificial about pairing human intelligence with machine-based smart manufacturing. Implementing an ever-smarter tomorrow in semiconductor manufacturing requires smart people just as much as it requires smart machines. Moving towards “smart” means understanding how to derive useful information and actionable intelligence from the ever-increasing pool of big data created during semiconductor manufacturing. Modern manufacturing sites are extensively instrumented today, and create massive amounts of data to consume, decipher, base decisions upon, or discard. As we dig into this problem we realize that equipment and processes in our industry are both obviously complex, but, also, subtly complex. Semiconductor manufacturing tools easily contain 100s to 1000s of components working together to produce nanometer scale, angstrom scale, or even atomic scale features using complex chemical, physical, and plasma processes. There is a plethora of potential failure points and modes, and despite our best efforts to collect more data, many processes continue to be only poorly observable. On top of that, semiconductor fabrication processes are always drifting, and the operational context is continually changing as we change product mix, process maintenance swap-out kit components, and operating conditions and recipes. Sounds like … hospitals, and healthcare? When you see your doctor, she will collect and look at your instrumented data – blood work, blood pressure, weight, and other quantifiable factors. But, typically, your doctor won’t draw a conclusion based on that analysis alone. Rather, your doctor will sit with you, ask probing questions, and record what she asked, your responses, and what she saw, what she heard, and what she thought. Then she’ll build a hypothesis, combining the “anecdotal” data with the instrumented data, and derive from that data set both a likely diagnosis and an effective course of action. In this case, beyond the instrumented data, two humans, and their natural language input, are part of the equation: the patient, with his observations and thoughts, as well as the doctor, with hers. And it’s been a formula for success. Healthcare has made huge, step-function improvements across a spectrum of deadly diseases, as well as with less-deadly chronic afflictions, by harvesting this complex input, committing the proven disease presentation – disease diagnosis – and disease treatment models to medicine’s collective memory, and then teaching the next generation of healthcare providers both the general methods and the standard protocols essential to maintaining good health and successful outcomes. Maybe, in medicine, what seems a big data environment is really just clusters and clusters of loose small data connected by the collective neural network of highly trained doctors and their colleagues. Nancy Greco (IBM Watson), Dave Mayewski (Rockwell Automation), James Moyne (University of Michigan / Applied Materials), and Paul Werbaneth (Intevac, Inc.), along with Julie Jacob (Ernst Young), and Carson Henry (Micron Technology), were members of the SEMI ASMC 2018 panel discussing Industry 4.0 and the Future of Commercial Semiconductor Device Manufacturing. All opinions here are purely our own. Please contact Paul Werbaneth via email at [email protected]. The SEMICON West (July 9-11, 2018, in San Francisco) Smart Manufacturing Pavilion features working production equipment on the floor and three full days of speakers providing insights on building the infrastructure needed to enable AI. Equipment from Bosch Rexroth, Cimetrix, Rudolph Technologies, INFICON, Final Phase Systems, OMRON, DISCO and Edwards Vacuum will showcase cutting-edge smart manufacturing technologies. For information on the SEMI Smart Manufacturing initiative and how to get involved, please click here.

The fast-growing automotive semiconductor market means big change for the IC supply chain. Beyond the obvious demands for reliability and traceability, the sector is moving towards simpler and lower-cost solutions while facing the daunting challenge of automating driving in a complex world. The need for simpler and cheaper automotive intelligence will likely drive acquisitions to build complete platform solutions that are easier to integrate. This demand has already spawned a market for pre-configured test cars to save developers time and money, and is driving LiDAR (Light Detection And RADAR) towards lower-cost, solid state solutions. “The growth of the automotive electronics market provides a great opportunity for the IC supply chain to differentiate on specialty processes and quality for the high-volume automotive business with its long design cycles,” says Scott Jones, principal, strategy, at KPMG, who will speak in the automotive program at SEMICON West. “This differentiation is a chance to reduce chip suppliers’ dependence on scaling volume for the mobile phone world with its short-cycle volatility of winning and losing sockets.” He notes that increasing demand for automotive ICs is also reinvigorating the eight-inch supply chain and spurring opportunity for specialty products such as compound semiconductor devices for power efficiency. Supplying the automotive market also means addressing automotive reliability requirements, which can be 10 times more stringent than for consumer devices. At the same time, the industry must sustain fast-paced development cycles required for the volume and diversity of low-cost IoT devices, manage the segmented supply chain for both those markets, and still spread development costs. Another big challenge for the supply chain will be to automate testing and update vast amounts of embedded software in these automotive devices. “The more complete solution a company can put together, the more the automakers will gravitate to it. They want simplicity,” Jones suggests. Smaller players will need to differentiate with IP and acquire other IP provider to build a broader platform, or be acquired and folded into an all-in-one solution.AutonomouStuff helps accelerate and simplify development of autonomous driving solutionsAutonomouStuff is helping to speed development of these platforms. The company has grown from a sensor distributor into a supplier in the emerging niche of vehicles preconfigured with key interfaces for sensors and controls. These interfaces can then be customized by integrating different components for developers to test their applications. AutonomouStuff offers developers a lineup of vehicle models pre-configured with the interfaces needed to add desired chips, sensors and software to develop their autonomous vehicle systems. Source: AutonomouStuff.“Whether they’re major chipmakers or AI software startups, they don’t have a year to build their own vehicle platforms themselves for developing autonomous vehicle systems,” says Wolfgang Juchmann, VP sales and business development at AutonomouStuff. Juchmann, a SEMICON West speaker, will bring a demonstration vehicle to the show. “In four to six weeks we can prepare a custom test car with selected sensors, enabling users to start testing their computer platforms and software. It’s faster and more cost-effective for us to supply the car with the needed interfaces.” He notes that developers are using some 300 AutonomouStuff vehicles in the field. AutonomouStuff customers are starting to transition from testing on a single car or two to testing on mini-fleets with 50 to 100 vehicles. Beyond sensors and pre-configured vehicles, the next step will be to add more data intelligence services to help with capabilities like tagging the data for training, Juchmann says. AutonomouStuff already offers hardware to support Baidu’s Apollo open-source software stack and data set. The company was recently acquired by the Swedish holding company Hexagon to help support expansion.CMOS silicon LiDAR nears automotive qualificationInnovations in the hyper-competitive LiDAR market, where burgeoning demand is driving the race to develop various types of solid-state devices, may also help reduce the cost of autonomous vehicles. Among the roughly 40 LiDAR suppliers, at least one – Quanergy – is taking advantage of 45nm and 32nm foundry CMOS volume production. The company uses voltage through the semiconductor stack to change the refractive index, controlling the phases of optical beams and the resulting interference patterns of light exiting the chip to quickly steer the laser beam without the need for moving parts, much like the phased array radar its team developed earlier. Solid state LiDAR image with object recognition software. Source: QuanergySo far, most of the small LiDAR units have shipped to the security, industrial automation, drone, robots and 3D mapping markets. However, Quanergy CEO Louay Eldada, another SEMICON speaker, says the company is also winning automotive designs and expects automotive shipments to take off early next year, once automotive certification testing is completed. “We can get design wins because standard CMOS production at TSMC makes us a known entity,” says Eldada. To prevent component misalignment, the company produces its own specialized packaging to secure the laser, phase control ASIC, optical phased-array emitter, detector array, and receiver readout ASIC at its plant in Silicon Valley or the facility of its automotive partner Sensata. Through its software business, Quanergy offers an artificial intelligence (AI) perception program for object recognition and LiDAR tracking. The solution uses the people-tracker software the company acquired from Raytheon.SEMICON West this year expands to three full days of automotive electronics programming and features a Smart Transportation Pavilion. Other companies with experts who will speak as part of the program include XPT/NIO, Infineon, McKinsey, Voyage, GM Cruise, Bosch, Deepen AI, Airbus A3, Nvidia, Excelfore, Byton, Macronix, SK Hynix, SAP, Xilinx, Achronics, California Fuel Cell Partnership, Velodyne, Lam Research, KLA-Tencor, SCREEN, Rockwell, Versum Materials, TechSearch International, Entegris, ASE, Amazon, Continental and Wind River. www.semiconwest.orgPaul Doe, SEMI

Device manufacturers continue to invest. Spending in cloud data center (compute, networking and storage), automotive (content per car increases), industrial (on content, factory automation, and positive macro trends), and consumer (gaming) end-markets is particularly strong. We see capital expenditure growth in 2018 and early indications pointing to sustainable spending into 2019. We also expect 14 percent increase (YoY) for fab equipment spending in 2018, up from the February forecast of 9 percent, and expect 9 percent increase in 2019, adjusted from the February forecast of 5 percent. 92 future facilities/lines with various probabilities are scheduled to start production in 2018 or later. Fab investment is just one indicator of how growing demand in areas such as from Artificial Intelligence (AI), cloud/data storage, automotive and Internet of Things (IoT) is driving unprecedented spending in the semiconductor industry. Below are a few highlights* of recent SEMI FabView insights. Details of each project can be found in FabView online 24/7 or World Fab Forecast report (Excel format). Infineon’s new 300mm Fab in Austria - Infineon is planning a new 300mm thin wafer Fab for Power Devices in Villach, Austria. Rumors on Toshiba’s new Fab plans - More 3D NAND fabs in the future at Toshiba are feasible. The timing will depend on market conditions, and our forecast will adjust accordingly. Vanguard's possible 300mm foundry fab - Vanguard's management said it might buy or build a 300mm fab in the near future as all 200mm fabs are essentially full. Powerchip plans to build new memory fab in Taiwan - Powerchip is investing more in expansions since Memory pricing is holding up. Rohm announced to build a new SiC fab in Fukuoka Japan - Rohm announced its plans to build a new SiC fab. Micron is building a new fab in Singapore - Micron broke ground in a ceremony for a new fab in Singapore on April 4, 2018. Bosch had groundbreaking ceremony of their 300mm fab in Dresden end April 2018 - Investment of 1 billion Euro. This is the biggest single investment in Bosch’s 130-year history. SEMI FabView, a mobile-friendly, interactive version of SEMI’s popular World Fab Forecast, delivers on-demand fab information such as fab spending and capacity for over 1,100 facilities, including over 82 planned facilities worldwide, across a wide range of product segments including Power, GPU, Memory, Foundry, MEMS and Sensors fabs. Fab data include region, start of construction, operation, construction and equipment spending, capacity, wafer sizes, product types and geometries. SEMI FabView subscribers receive forecast model updates through SEMI’s World Fab Database. Click here for a trial to experience SEMI FabView first hand. *Actual updates provide more detail Christian G. Dieseldorff and Clark Tseng, Industry Research Statistics Group, SEMI.

The fast-maturing hardware and software that are enabling practical applications of equipment intelligence and machine learning mean disruptive change for microelectronics manufacturing. But first comes the basic work of building the basic infrastructure, figuring out IP separation, and learning to solve physical problems in the digital world. Just how much can the semiconductor industry leverage industrial IoT practices from other industries? Common wisdom may be that industrial software solutions aren’t well suited to the IC sector’s complex needs. But GE Digital enterprise account executive Luke Smaul, currently working with Intel, argues that semiconductor fabs and toolmakers are dealing with similar issues as GE did when it first started working with Delta Airlines to monitor the GE engines on Delta planes. Smaul will speak at SEMICON West about GE’s work with Intel over the past few years and, in particular, how its solution for cloud security and IP separation can work for ICs. “GE learned to provide IP security and separation in the aviation space with its suppliers, which moved us all up the value chain, providing a big engine for growth,” says Smaul, who started his career as an IC engineer. “GE Aviation saw a 25 percent increase in issue detection rates by leveraging the same common platform. We’ve shown that we can protect Intel intellectual property in its own cloud space and control who can access what.” A toolmaker can access only particular fab data as needed for analysis, and then can reveal only the output from the analysis and a subset of supporting data. “IP separation has to happen, and it will unlock huge added value,” Smaul says. GE’s Predix solution aims to supply an easy-to-use, plug-and-play system for analytics to enable a yield engineer without a deep data background to select a supported sensor, a gateway to connect automatically to the cloud, and an analytics application to test a hypothesis of how the collected data relates to yield. “This empowers the yield engineer to use and unlock information for a quick improvement, even for simple things such as looking at the impact of degradation of fan performance over time on yield,” says Smaul. “Though the scope may be small, the impact on yield in aggregate, and when scaled, is large.” “There needs to be much more collaboration across the industry to make this work, and to share best practices,” says Smaul. “Just as GE moved from selling gas turbines to selling power-as-a-service, vendors of other big, expensive assets like IC equipment will likely change their business model from selling tools towards selling yield-as-a-service. This will simplify life for the fab while bringing the toolmaker more opportunity to sell improved capabilities on existing tools.” More human intelligence makes AI smarter Applying AI neural network approaches such as deep learning to predict outcomes from digital models is enabling disruptive advances in speech and image recognition, but applying it to complex IC manufacturing problems such as predictive maintenance has been a challenge. These neural networks require massive amounts of data to train, and the IC sector doesn’t really have big data, just a lot of little data clusters due to the dynamics and context richness of processes. This data is difficult to combine for analysis. In addition, the neural network provides only an answer but can’t explain why, notes Michael Armacost, managing director of advanced service engineering at Applied Materials. “We’ve learned that it works better if we do not ignore what we know already, but rather incorporate expert knowledge in a structured way to help us focus on the key features and the key data,” says Armacost, who will also speak in the program. This includes choosing the most important steps to include in the model, identifying the limited data to collect and how to filter the data for outliers, and then selecting the final parameters and features, adjusting the limits, and making adjustments as results drift change over time. The less data needed, the better for the complicated issue of IP protection as well. The big gains from these new analysis approaches will likely require data from more than one company and supporting security for remote connectivity. “Some end users are attempting to do the AI all themselves, but in the long term there will need to be collaboration across companies,” says James Moyne, University of Michigan professor and consultant to Applied Materials, another speaker. Collaboration will need to balance the value of the solution against the risk of compromising IP. “The low-hanging fruit are applications such as predictive maintenance in areas that do not involve high-priority IP. Another approach will be to limit the amount of shared data needed – to first build the model on a wide range of data, but then to use only a very small amount of data to operate the models.” Ready-made models could speed the process Coventor’s semiconductor process models are finding initial applications in R D whereby companies use the simulation to understand the effect of process variation on their complex designs. Instead of running dozens of actual wafers to optimize semiconductor processes, users can instead quickly simulate the results of complex process interactions on their design. Going forward, the process models could find a wide range of applications, from accelerating stabilization of new processes in the fab to enabling real-time co-optimized control across previously independent unit steps to improve wafer uniformity. “This improved uniformity across wafers and equipment could potentially reduce the need for costly physical silicon validation,” suggests Joseph Ervin, Coventor director, semiconductor process and integration, another SEMICON speaker. “Making use of in-situ metrology for real-time control also demands a digital model to process and analyze the collected data for quick response. This area has tremendous potential for improving semiconductor process control.” SEMICON West features a Smart Manufacturing Pavilion with displays and three full days of speakers on building the infrastructure needed to enable disruptive artificial intelligence in the microelectronics sector. www.semiconwest.org The SEMICON West Smart Manufacturing Pavilion features interactive Touch Liquid Crystal Displays (TLCD) and working production equipment on the floor from Bosch Rexroth, Cimetrix, Rudolph Technologies, Inficon/Final Phase Systems, OMRON, DISCO and Edwards Vacuum. For information on the SEMI Smart Manufacturing Initiative and how to get involved, please click here. Paula Doe, SEMI

Do you email your doctor when you have a tickle in your throat? Wear a fitness tracker or use an app to monitor your sugar levels, exercise or nutrition? If so, congratulations! You are a part of the rapid growth of digital medicine.Since you’re on the leading edge of this trend to enhance the efficiency of healthcare delivery, you might enjoy learning more about how the medical industry is transforming healthcare in this collection of podcast episodes and video. These are my top picks and just the tip of the proverbial iceberg in how modern medicine is taking today’s technology and applying it to best practices for remote patient monitoring, medical diagnosis, rural healthcare and more.Enjoy! And let me know if you find any others worth the listen!1. Inside Angle: 3M Health Information Systems - Telemedicine: Enhancing Access to Improve OutcomesAccess to healthcare can be a matter of life or death. For some patients, this may mean taking a day off work and driving for hours because services are not available in their hometown.Inside Angle host Dr. Gordon Moore interviews Barb Johnston, co-founder and CEO of HealthLinkNow, about the implementation of telemedicine programs. They discuss how technology impacts telemedicine adoption and related regulations and benefits clinicians and patients in the telepsychiatry and mental health industry. 2. NPR’s The Salt: What’s On Your Plate – This Chef Lost 50 Pounds and Reversed Prediabetes With A Digital ProgramThis short audio clip and article dives into lifestyle and wellness apps designed to motivate users to eat healthier, exercise more and – in some cases – save them from a preventable disease, like diabetes.Wellness apps like Omada Health rely on smartphones, e-coaching, electronic nudging and other methods to encourage users to make and, more importantly, stick to changes that can improve their health. And it’s catching on as other organizations, such as The Centers for Disease Control and Prevention, recognize the difference lifestyle-change programs designed to prevent diabetes are making. 3. Red Hot Healthcare: Episode 57 - Bluer Skies for TelemedicineLast year alone, venture capitalists poured $7 billion into telemedicine. As an emerging concept, telemedicine has a long road ahead until it’s fully integrated and adopted into modern medical practices. Nathaniel Lacktman, partner and health care lawyer with Foley Lardner LLP, discusses legal issues and hurdles surrounding telemedicine today with Red Hot Healthcare host Dr. Steve Ambrose.Lacktman pointed to healthcare providers such as Mercy Virtual Care Center, Mayo Clinic and MDLive that are likely to lead the telemedicine race. This podcast episode is a great listen for entrepreneurs and medical professionals who are interested in learning about compliance and business considerations for the implementation of telemedicine. 4. Digital Health Today: Episode 56 – Eren Bali on Building the World’s Largest Connected Care NetworkEren Bali, CEO and co-founder of Carbon Health, is out to change our fragmented traditional healthcare system with his aim to build the world’s largest connected care network.For Bali, it all started with his sister’s experience consolidating his mother’s health records – scattered over 20 different systems. And from that challenge, his idea to create a universal network that aggregates patient medical records was born.Bali and Digital Health Today host Dan Kendall discuss Bali’s launch of Carbon Health, the new medical records system’s first implementation in a San Francisco primary care clinic, and how providers and patients can get on board with this model. 5. Digital Health Today: Episode 58 – Brennan Spiegel Gets Real About Virtual RealityThis episode of Digital Health Today centers on how Virtual Reality (VR) is being used to enhance patient care. Host Dan Kendall speaks with Dr. Brennan Spiegel, Director of Health Services Research at Cedars-Sinai Health System and Professor of Medicine and Public Health at UCLA, about how immersive technology like VR and Augmented Reality (AR) can improve patients’ experience as well as alleviate pain, anxiety, depression, addiction and more. In a particularly interesting segment, Dr. Spiegel calls the hospital a biopsychosocial jail cell and underscores its importance in not just treating physical ailments but, more wholistically, also addressing the psychological and social wellbeing of patients. 6. TED2017: Raj Panjabi – No one should die because they live too far from a doctor What do you do when access to a doctor means rowing a boat for hours? Millions of people around the world lack access to health care because they live in a remote town or village.In this TED talk, Raj Panjabi, physician in the Division of Global Health Equity at Harvard Medical School, Brigham and Women’s Hospital, and co-founder of Last Mile Health, offers a solution to the problem of healthcare access in the form of the Community Health Academy, a global platform to train and connect community health workers by leveraging devices like smartphones to bring preventative healthcare to even the most far-flung regions of the world. 7. GeekWire’s Health Tech Podcast: How AI is making humans the ‘fundamental thing in the internet of things’Can AI predict which patients are at risk for chronic diseases using their old medical records? This episode of Health Tech Podcast dives deep into the future of healthcare with a look at the potential for AI -- “augmented intelligence” or “assistive intelligence” – to improve patient outcomes, the focus of Ankur Teredesai, a data scientist at the University of Washington and co-founder and CTO of health AI startup KenSci.Clare McGrane, host of GeekWire’s Health Tech Podcast, speaks with numerous experts in this field. They compare precision health to a modern car constantly monitored by microprocessors. The minute something is wrong, you get a warning to take the car to a shop to resolve the issue. The same concept can be applied to humans and deliver big healthcare impacts as research in AI and machine learning continues to evolve. 8. NPR’s Shots: Can Home Health Visits Help Keep People Out Of The ER?In Washington D.C., the city with the highest per capita 911 call volume in all of the U.S., Mary’s Center is piloting a program to provide primary care telemedicine to patients who cannot, or in some cases, do not want to visit the clinic. NPR covers the story of Medicaid patient Dennis Lebron Dolman, who is currently receiving a mix of home visits and virtual treatment.Besides providing healthcare access to rural regions, telemedicine has the potential to reduce emergency room visits in cities as well as improve the health of patients in the long run. Learn more at SEMICON West’s Smart MedTech TechXPOT on Digital Medicine and Remote Patient Monitoring. Hosted by NBMC on July 12, 2018, from 10:30 AM to 12:30 PM, the event will feature healthcare industry leaders exploring state-of-the-art healthcare practices and the future of medical technology. Registration is now open!Amy Ly is a Technical Programs Marketing Coordinator at SEMI.

Artificial intelligence (AI) is making headlines everywhere, offering a range of capabilities, including location and motion awareness — determining whether a user is sitting, walking, running or sleeping. Behind the scenes, AI is capturing volumes of data. Makers of smartphones and fitness and sports trackers, along with application developers, are all clamoring for this data because it helps them analyze real-world user behavior in depth. Manufacturers gain a competitive edge by tapping this intelligence: Using it to improve user engagement, they increase the perceived value of their devices, potentially reducing customer churn. How can consumer-product manufacturers tap the built-in capabilities of MEMS inertial sensors — which are already ubiquitous in end-user devices — to make the most of AI? Machine learningProduct manufacturers can easily build an activity classification engine using commonly available smart sensors and open-source software. Activity trackers, for example, use raw data first collected via the MEMS inertial sensors that are already installed in smartphones, wearables and other consumer products.With the building blocks in place, consumer-product manufacturers can apply machine learning techniques to classify and analyze this data. There are several possible approaches, ranging from logistic regression to deep learning neural networks.One well-documented method used for classifying sequences in AI is Support Vector Machines (SVM). Physical activities, whether walking or playing sports, consist of specific sequential repetitive movements that MEMS sensors gather as data. MEMS sensors make good use of this collected data, which can be easily processed into well-structured models that are classifiable with SVMs.Consumer-product manufacturers have gravitated toward the SVM model since it is easy to use, scale and predict. Using an SVM to set up multiple simultaneous experiments for optimizing classification over diverse, complex real-life datasets is far simpler than with other approaches. An SVM also introduces a wide range of size and performance optimization opportunities for the underlying classifier.Cost impacts of processing, storage and transmissionIn practice, recognizing user activity hinges on accurate live classification of AI data. Therefore, the key to optimizing product cost is to balance transmission, storage and processing costs without compromising classification accuracy.This is not as simple as it sounds. Storing and processing AI data in the cloud would leave users with a substantial data bill. A WiFi, Bluetooth or 4G module would drive up device costs and require uninterrupted internet access, which is not always possible.Relegating all AI processing to the main processor would consume significant CPU resources, reducing available processing power. Likewise, storing all AI data on the device would push up storage costs.Resolving the issuesTo resolve these technology conflicts, we need to do four things to marry the capabilities of AI with MEMS sensors.First, decouple feature processing from the execution of the classification engine to a more powerful external processor. This minimizes the size of the feature processor size while eliminating the need for continuous live data transmission.Next, reduce storage and processing demands by deploying only the features required for accurate activity recognition. In one example created by UC Irvine Machine Learning Repository (UCI), when an AI model was trained using a dataset of activities with 561 features, it identified user activity with an accuracy of 91.84 percent. However, using just the 19 most determinative features, the model still achieved an impressive accuracy of 85.38 percent. Notably, pre-processing alone could not identify these determinative features. Only sensor fusion enabled the data reliability required for accurate classification. Third, install low-power MEMS sensors that can incorporate data from multiple sensors (sensor fusion) and enable pre-processing for always-on execution. A low-power or application-specific MEMS sensor hub can slash the number of CPU cycles that the classification engine needs. The onboard software can then directly generate fused sensor outputs at various sensor data rates to support efficient feature processing.Finally, retrain the model with system-supported data that can accurately identify the user’s activities.Additionally, cutting the data capture rate can reduce the computational and transmission resource requirements to a bare minimum. Typically, a 50 Hz sample rate is adequate for everyday human activities. This may soar, however, to 200 Hz for fast-moving sports. Reducing dynamic data rate selection and processing in this way lowers manufacturing costs while making the product lighter and/or more powerful for the consumer.High efficiency in processing AI data is key to fulfilling its potential, driving down costs and delivering the most value to consumers. MEMS sensors, in combination with sensor fusion and software partitioning, are critical to driving this efficiency. Operating at very low power, MEMS sensors simplify application development while accurately analyzing motion sensor data.Combining AI and MEMS sensors into a symbiotic system promises a new world of undreamt-of opportunities for designers and end users.This blog post is based on an original article that first ran in EDN. It appears here with the permission of the publisher.