Tools for Tomorrow
January 17, 2020
”How Facial Recognition Technology is Identifying Rare Genetic Disorders”
Around the world more than 350 million people suffer from rare genetic diseases, a number higher than the total population of the United States. Unfortunately many people who suffer from these conditions fail to receive a diagnosis as a doctor is unlikely to encounter many victims of a particular condition and hence may be unfamiliar or even unaware of a specific disease. Some of these conditions, such as Cornelia de Lange syndrome and Angelman syndrome exhibit distinct facial traits such as arched eyebrows and light skin tone and hence the same facial recognition technology used in airports and smart cities is now being used in hospitals to help identify patients who might have a rare condition.
US company FDNA has already demonstrated the power of facial recognition in rare genetic condition discovery in several experiments. By using algorithms which have been trained on facial images of patients with certain diseases, facial recognition software was tested against the performance of human experts with encouraging results. FDNA’s facial recognition software was able to distinguish a genetic condition 90 percent of the time compared with 70 percent for human experts. When asked to specifically recognize the rare condition Noonan syndrome compared to four other conditions, the program was 64 percent accurate, significantly higher than chance alone.
In the coming years facial recognition technology will become a standard part of a medical checkup, and will be integrated into several aspects of the healthcare industry. NEC is a global leader in facial recognition technology and offers many solutions for the healthcare industry including using its NeoFace Watch facial recognition solution to efficiently manage patient check-in, employee management and security services.
”How AI is Driving the Drug Discovery Business”
The drug discovery business is one of the most challenging business environments on the planet. Developing a new drug from inception to approval costs on average US$2 billion and can take decades to finally receive regulatory approval. As a result of these harsh market conditions, drug prices are high, piracy is rampant, competition is limited and many cures for many rare diseases are often not developed due to the lack of a business case.
For example, in the United States and Europe the National Center for Advancing Translational Sciences Genetic and Rare Disease Information Center (GARD) estimates that 50 million people suffer from a rare disease and has identified 7,000 such conditions. Due to the problems mentioned above, however, only 5 percent of these conditions have any type of approved therapy. Examples of rare conditions include Duchenne muscular dystrophy, ovarian cancer and cystic fibrosis.
Artificial Intelligence may very well hold the key to revolutionizing the drug discovery business by optimizing the efficiency and hence costs of the process. Currently 9 out of 10 newly-developed medicines fail to reach the clinical trial phase. A major reason for this is that there are just too many variables for a human being to monitor once a correlation between a chemical compound and an effect on a disease is found. This is where AI can play a major role as massive datasets can be analyzed quickly and demographic and genetic factors can be more readily isolated for further study. Once machine learning algorithms have been trained with large datasets the entire industry will experience a massive digital transformation.
NEC technology is working to improve the speed and efficiency of the drug discovery process, and the company is specifically working to help find new vaccines for several types of cancer. NEC is deploying its `NEC the Wise` AI platform to help analyze the results of clinical trials of head, neck and ovarian cancer around the world with the hope of developing a faster and cheaper cure.
”How Facial Recognition Technology is Optimizing Agriculture”
Agriculture is arguably the world's most important industry, and is under increasing pressure to feed and clothe the global population as it nears 8 billion people. But agricultural livestock is responsible for some of the largest environmental and health challenges facing the planet. For example, livestock is the leading cause of greenhouse gas emissions across the world at 24% of the global total, and livestock production alone generates more than 15% of the global total. In fact, the methane gas produced by livestock has a 20 times greater impact on greenhouse gases than do typical carbon dioxide emissions. Furthermore, animals consume 70% of the 10,000 tons of antibiotics used in the United States every year which contributes largely to bacterial resistance to medicine. The agriculture industry needs to pivot to a precision-agriculture business model and technologies such as facial recognition are starting to facilitate this transformation.
Facial and behavior recognition technologies are not just for people, as they are starting to be used to manage animals as well. Facial recognition has matured to the point that it is able to accurately able to identify an individual animal by its facial characteristics or the pattern on its hide. Current biometric services for agriculture are able to determine the overall health of an individual animal, such as determining if it is sick and how much it is eating and drinking. Animal tracking and monitoring have been utilized by the cattle industry but facial recognition technology offers some unique benefits to farmers. Camera-based biometric monitoring removes the need for expensive wearables, wireless connectivity, battery issues and data plans resulting in continuous and cost-efficient farm management. This technology reduces the amount of feed, pesticides, fertilizers and antibiotics used in cattle rearing and makes the business both more cost-effective and environmentally friendly.
NEC provides cutting-edge innovations and leads the world in facial recognition technology.
”How 5G will Revolutionize Enterprise Communications”
Despite global growth in Internet use across the world, many business users still have very limited options for connectivity and prices remain expensive for business use. For example, according to data from the Federal Communications Commission more than 100 million American households and businesses only have access to one ISP, which keeps prices high and speeds low. Furthermore, especially in developing countries Internet monopolies have kept prices high and many emerging markets in Asia, Africa and Latin America charge more than US$100 per Megabyte of Internet service.
5G mobile broadband service will change the way in which businesses communicate by offering a real wireless office experience, often at a lower price and by offering faster speeds. Tests of live commercial 5G service in markets such as the United States are already showing speeds of over 600 Mbps and connection speeds of over 1 Gbps are expected in the near future. As more communication service providers launch commercial 5G service, markets will become more competitive as businesses will have more options for enterprise connectivity.
5G solutions will also transform the way in which businesses communicate internally, delivering high speed voice and video communications in a wireless environment. For example, NEC has tested its 5G equipment in Japan among hospitals, mobile doctors and ambulances. As a result, all parties are able to view video in 4K definition, allowing first responders to prepare adequately before physically seeing a patient. This new method of communications replaces using voice over radio, and is only possible with the new features of 5G technology.
NEC has recently started a partnership with Samsung Electronics to collaborate on jointly developing their 5G portfolio by combining their mobile network and IT solutions. The partnership will see both companies cooperate in sales and marketing of 5G solutions to telecommunications operators starting in 2019 and in the future the companies with utilize technologies such as AI to create new 5G services.
”How High Performance Computing is Transforming Industries”
NEC has long been an industry player in the supercomputing space and manufactures a line of solutions designed to take Machine Learning, Artificial Intelligence and Big Data Analytics to the next level. NEC’s SX-Aurora Tsubasa solution is capable of between 2.15 and 2.45 trillion floating operations per second and has been deployed in a variety of industry use cases. NEC recently signed a deal with the German weather forecasting service DWD to use its HPC products to produce more accurate weather information, resulting in better information for consumers and businesses alike.
High Performance Computing (HPC) has traditionally only been used for intensive scientific research due to the high costs involved, but due to economies of scale HPC is increasingly being used by enterprises. According to estimated 2.5 quintillion bytes of data are created every day, and businesses are struggling to collect, secure, store and perhaps most importantly analyze this data. Today’s fastest supercomputers in the USA and China have already breached 122 quadrillion FLOPS and 93 quadrillion FLOPS calculation speeds respectively at the demonstration level and are hence increasingly seen as an important business tool to generate new insights from data.
For this reason, HPC is starting to be used in industries such as finance, healthcare and energy to analyze the massive amounts of data being generated by enterprises. Common use cases for such industries include advanced cybersecurity in banks, new drug discovery in healthcare and improving resource exploration in oil and gas fields.
For example, the German weather forecasting service the DWD recently deployed NEC`s HPC solution in order to provide high-accuracy weather predictions. By using HPC, DWD will analyze large amounts of weather data to generate predictions 12 hours in advance. These predictions will be used for a variety of use cases including aviation, marine shipping, energy and natural disaster warning services. Therefore HPC will be used by a variety of enterprises in Germany in order to facilitate safer and more efficient business operations.
HPC is also starting to emerge as an important tool for medical researchers, as shown by recent research initiatives from the U.S. Department of Energy's Argonne National Laboratory. The institute has recently partnered with several universities to use HPC to research to improve cancer immunotherapy treatments. The group is working to create a digital twin of cancer patients, each one unique to a patient's genome. By applying the processing power of HPC to a digital twin, millions if not billions of possible virtual drug combinations can be tested on a patient with the hope of improving cancer therapy which is only effective 10% to 20% of the time.
NEC has long been a player in the supercomputing space and manufactures a line of solutions designed to take Machine Learning, Artificial Intelligence and Big Data Analytics to the next level. NEC`s SX-Aurora Tsubasa solution is capable of between 2.15 and 2.45 trillion floating operations per second and has been deployed in a variety of industry use cases.
”How Technology is Helping Enterprises Overcome Staffing Challenges”
Many developed countries are facing a decline in population which in turn has created labor shortages, especially in labor-intensive sectors. This has in turn put pressure on companies to raise wages and therefore incur lower margins or in some cases to reduce operational hours, reducing revenue. Several countries in the world already have a negative population growth rate such as Japan, Spain, Italy and many Eastern European countries and many more developed markets are expected to join the group in the coming years. The issue is particularly pronounced in Japan as according to the Ministry of Health, Labor and Welfare there are over 8 job openings in the security industry for every applicant, 5 in the construction industry and 3 in the logistics industry creating major challenges for domestic businesses.
As such, many companies are turning to technology to address the labor crisis, as technologies such as service robots, the Internet of Things, Artificial Intelligence and Big Data Analytics are all being increasingly used to automate processes formerly done by human employees. Service robots, for example are already being used as security guards in airports and train stations while drones have started to commercially deliver goods. NEC has created Internet of Things solutions in several industries to overcome this challenge. For example, in retail NEC has created sensor networks which are able to monitor the volume of goods inside of a store in real time. This will optimize inventory tracking and re-stocking which can be done with less human labor. AI can even be used in conjunction to automate re-ordering of goods and demand forecasting which is currently done by people.
In a partnership with 7-Eleven, NEC has been carrying out trial operations of a self-checkout store in Tokyo’s Mita ward since 2018.
”5G Network Slicing Business Models in the United States”
5G mobile network deployments are well underway in the US market, as networks are now live in many major cities such as Dallas, Chicago and Atlanta and these networks are generating download speeds of anywhere between 250 Mbps – to more than 1Gbps. And while the much faster download speeds that 5G brings are currently generating the most attention in the media, 5G networks will eventually offer much more than simply a bigger pipe with which to consume data.
One of the other key new solutions that 5G networks will bring is network slicing. Network slicing is simply defined as splitting a single mobile network band into several virtual network slices, all of which can have different characteristics. For example, a factory might use network slices for security cameras, IoT sensors, drones and connected vehicles in the future but all of these different devices will have different bandwidth, security, storage and QoS requirements. Therefore, a network slice could be configured to manage all of these different services, which is made possible due to the higher bandwidth and lower latency offered by 5G.
Network slicing is possible to some extent but 5G will bring new capabilities and new business models. Because 5G will offer such high bandwidth anything from 4K streaming video to very low wireless sensor use will be possible with network slicing. And because 5G networks are increasingly cloud-based, there will be a variety of on-demand capabilities and network slices will be highly customizable. For example, a 5G network slice could only be made to cover a specific area like a sports stadium on the days that games are played. If a company added a new feature as part of an IoT deployment or changes the requirements of an existing service, the changes to the network slice could be made instantaneously.
For this reason, mobile operators in the United States are preparing to offer 5G network services, especially in the area of public safety. Network slicing for local governments will enable highly optimized networks for first responders, and these networks also very often have very strict service level agreements of up to 99.999% availability. For example, if the police are monitoring an incident via connected cameras in a particular area the network quality could be instantly improved. In the case of a natural disaster network quality could be boosted for certain services only for a limited period of time and for a specific area. This new business model will ultimately result in a better quality of service, on-demand access and with flexible pricing compared to current services in the market.
NEC offers a wide variety of 5G network solutions and services, including several deployments with major Tier 1 service providers.
”How are Industries Adopting Blockchain in the United States”
Blockchain technology quickly emerged as one of the biggest topics in the IT sector when the rise of cryptocurrency brought blockchain into the public's eye and it has interest levels have remained high ever since. The concept of creating a decentralized, distributed ledger like the one associated with Bitcoin seems like a technology which is well-suited for the financial industry and therefore most of the focus of blockchain technology has been focused on how to disrupt and transform the financial industry. More recently, however, there is an increasing focus on how blockchain can be used to disrupt other industries as the need for greater transparency and security in business transactions is widespread. In the United States a variety of new industries are now testing blockchain technology and have yielded some interesting results.
In the education sector, diploma fraud is a major problem as estimates place the number of fake diplomas in the job market as upwards of 20%. To address this issue the Massachusetts Institute of Technology recently became the first major university in the United States to issue course-completion certificates using blockchain technology. Students are now able to seamlessly share their ID-verified diploma with potential employers and other institutions thereby saving considerable time and resources.
Another area where blockchain is gaining ground in the US market is in food safety, as contaminated food products sometimes result in illness outbreaks and lead to expensive product recalls. Walmart has been very active in the area and recently started requiring suppliers of lettuce and spinach to use a blockchain database. By using blockchain technology, the source of contaminated products can be quickly identified and isolated.
Blockchain has also found a use case in the public sector in the United States and was used for the first time to verify voters in an election in 2018. In the state of West Virginia 150 voters in the 2018 primary election used an app in order to cast votes. This app uses blockchain technology in order to verify one's identity as a voter provides several forms of physical identification which is again verified using facial recognition technology. Voter fraud is a major concern in the United States and blockchain could play a large role in helping to eliminate this problem.
NEC has spent significant R&D efforts on blockchain technology and has developed a system with several important features such as enabling privacy, governance and scalability which will be key components of future IoT deployments. NEC is also actively developing blockchain solutions with industry partners and recently announced a partnership with NGO Bitcoin Argentina. The two companies will create a blockchain-based secure ID system which will enable better access to government services in underprivileged areas of Buenos Aires.
”How AI is Driving Further Change in Retail in the United States”
In the United States due to the massive uptake of e-commerce service there has been a massive closing of US retail locations which the media has dubbed the “retail apocalypse”. E-commerce now comprises 10.7% of all retail sales in the United States according to U.S. Census statistics in the second quarter of 2019 and this has led to a significant closing of retail locations in the country. More than 6,000 retail locations have closed in the U.S. so far which is more than the 2018 total and this trend is likely to continue as an estimated 75,000 more retail locations will close in the United States from now until 2026 as the market readjusts itself in the e-commerce era.
In order to deal with increasing pressure from e-commerce vendors and to provide a better user experience, U.S. retailers are increasingly turning to Artificial Intelligence technology in order to do so. For example, Kroger, which is the largest grocery store chain in the United States is increasing its use of autonomous delivery vehicles. Customers who wish to make an order can use an app to order goods which can be delivered via an autonomous vehicle for a flat fee of US$5.95 with no minimum purchase value in select U.S. locations.
Another important AI tool that US retailers are using is the chatbot, as functionality has increased substantially over the last few years. For example, supermarket chain Whole Foods is now using a chatbot which is intelligent enough to perform several functions. For example, this chatbot can help a consumer locate a store, find products within a store, help customers find specialized products for those which dietary restrictions and can even suggest recipes for customers who purchase specific ingredients.
Even AI-driven voice recognition technology is starting to be deployed in the retail sector in the United States as McDonald's recently acquired an AI-voice recognition company focused on food ordering named Apprente AI. This system will allow drive through ordering to be processed by AI and hence will allow for considerably faster operations. This recent move is the second major AI initiative undertaken by McDonald`s as the company purchased a big data analytics company Dynamic Yield in 2018 which provides real time insights on the purchasing behavior of the 68 million customers that it serves daily.
NEC has a variety of IoT solutions for the retail industry including face recognition for self-checkout, automatic equipment monitoring and AI-based ordering systems. NEC also has a global partnership with 7-11 inc. in order to drive digital transformation across the chain`s 36,000 stores worldwide.
”How is IoT Changing the Construction Industry in Singapore”
Singapore is home to a vibrant construction sector as the country as a constant demand for new housing and office space, tourist infrastructure and land reclamation projects, and the size of the country has increased by 25% in the past 200 years due to these projects. Singapore’s population has increased by 45% over the last 20 years and remains a key global business and tourist hub and hence the construction industry is especially vital to the future economic growth in the country.
Singapore’s construction industry has faced challenges in recent years despite the constant demand for new building projects. The country hosts nearly 300,000 foreign construction workers to keep up with demand, but new immigration restrictions which are under review could limit this number resulting increasing costs and Singapore does not have a large supply of low-cost labor. As the country also has limited natural resources, most construction materials must also be imported which also makes construction companies susceptible to price fluctuations in commodities. Singapore has also experienced a few construction injuries and fatalities, and this is an area where the government seeks to improve.
In the face of these challenges, Singapore’s construction industry is turning to new IoT technologies in order to create a more profitable and safer work environment. The country`s Building and Construction Authority is heavily encouraging the industry to adopt new digital technologies and they envision a future where construction sites will have little to no on-site human labor. According to them, advanced robots will soon be able to perform advanced construction work functions creating a safer and more efficient industry. Buildings of the future will all have a digital twin and use real-time IoT data to monitor individual building conditions and 3D printing will allow components to be made on-site as opposed to having to be imported.
Some of these new technologies are already appearing in the market, as the country has developed a robot named OutoBot which can wash and paint the exterior of the country’s housing estates. OutoBot has a robotic arm and a spray nozzle which can dispense both water and paint, and its creator claims that it can reduce washing time by 30% and painting time by 20%. The robot needs a two-person crew compared to a traditional crew of five people for the same job, and workers are no longer required to ride in a gondola at extreme heights which can be quite dangerous.
NEC has a full suite of IoT solutions designed for the construction industry and these services include providing temporary on-site wireless networks, big data solutions to predict weather conditions and power requirements, secure access to construction sites using biometrics and wireless monitoring and control of heavy machinery. In 2018 NEC incorporated 5G connectivity into its construction solution portfolio and demonstrated a better-performing remote control solution due to the high quality 4K images which 5G can provide. More recently the company demonstrated an autonomous backhoe solution which is able to load sediment with significantly less human labor.
”How Will Biometrics Revolutionize Retail Banking”
ATM and credit card fraud remain major challenges for financial institutions, as at the global level credit card fraud totaled a massive US$24bn in the last twelve months and there has been a rise in hacking taking place at ATM machines. For example, in 2016 a criminal enterprise was able to steal over US$12m from Japanese ATMs in less than two hours in a highly coordinated operation using cloned ATM cards. A new type of cyberattack called `jackpotting` where malware can be inserted in an ATM and cause it to eject all its money are also on the rise and 82 such attacks have been recorded in Germany alone over the past several years.
Financial institutions are also facing challenges in terms of operating numerous physical locations which is becoming increasingly burdensome as more and more banking in done online. This has been a particular problem in all over the world as for example in the United States 1,974 retail bank branches were closed in 2018 with most banks citing cost pressure as the key reason for doing so. And while it is inevitable that some branches will close due to the rise of online banking, this has made some processes, such as onboarding customers, much more time consuming and resulted in a diminished customer experience. For this reason many financial institutions around the world have turned to biometric services as a way to fight back.
In Japan, for example, Seven Bank and NEC recently announced that they have co-developed biometric ATMs which will allow customers to open new accounts. Customers will have a QR code, can scan ID documents and have a photo taken to complete the customer onboarding process in 20 to 30 seconds. Seven Bank plans on replacing all its 25,000 ATMs with this system by 2024. These ATMs also have advanced IoT functionalities as they are embedded with sensors which can predict component failures and even have an embedded AI system which can anticipate the demand for cash at specific machines.
And while ATMs will be transformed into customer service centers using biometrics soon, facial recognition technology will eventually also allow for financial transactions to be done using biometric authentication. For example, Taiwan’s E.Sun Bank recently demonstrated an ATM which can allow customers to withdraw and deposit funds using only a face as an ID. Going forward this could eliminate the need for physical ATM cards resulting in more efficient banking operations as there will be significantly less activity directed to call centers and other back office operations who deal with lost and stolen cards, forgotten PIN numbers and distributing and manufacturing physical cards.