Daily Archives: February 6, 2021

Wed all like to be able to look into a crystal ball and tell the future. With the use of AI and machine intelligence, we are now able to more accurately produce trend forecasts that accurately mimic reality. NWO.ai takes this one step further. The platformcombs through unstructured data to track 20M+ signals, adds a time dimension, eliminates noise, and prioritizes sources to identify microtrends and global cultural shifts. For example, the platform has recently considered and tracked possible geopolitical events and themes like will COVID-19 lead to war, will Bitcoin continue to rise, and the impact of the pandemic on Big Tobacco. The platform is already being used by several Fortune 500 companies across various industries to navigate these uncertain times.

AlleyWatch caught up with Cofounders Sourav Goswami and Pulkit Jaiswal to learn more about the companys recent launch out of stealth, future plans, and recent round of funding.

Who were your investors and how much did you raise?

We raised $3.5M in a seed round co-led by Hyperplane, Wavemaker, and Colle Capital with participation from Adit Ventures and SuperAngel.

Tell us about the product or service that NWO.ai offers.

Weve built the largest AI-enabled civilian database capturing the voice of the consumer. Using proprietary analysis of unstructured external data, were genetically sequencing the lifecycle of trends as theyre created and evolve.

What inspired the start of NWO.ai?

We started NWO originally to use disparate data sources to try to uncover geopolitical trends. As we curated and processed more data and began to incorporate proprietary methods of time-shifting data sources, we began to realize we were able to anticipate consumer and cultural shifts.

How is NWO.ai different?

Our platform goes beyond providing just descriptive statistics on the number of mentions of a keyword on a variety of data sources, but instead incorporates a time dimension and source prioritization component to determine not only the past performance of trends but also help predict future potential.

We have developed a novel cross-correlation, time-shifting algorithm that automatically weighs a number of data sources across various input streams and figures out the leading and lagging indicators, time-shifts the lagging ones and generates a composite signal.

What market does NWO.ai target and how big is it?

We are targeting the Total artificial intelligence market as a whole which is worth ~$40B according to Grand View Research, and expected to eclipse $700B by 2027. As a starting point, our first sights are set on the artificial intelligence application for supply chain and marketing use cases which is estimated at ~$8B, per Meticulous Research, and expected to cross $60B by 2027. Both these market estimates are growing fast as the focus on AI-powered decision making has become a key priority for businesses across almost all industries.

Whats your business model?

NWO.ai is a SaaS business that offers a paid monthly access to its web-interface platform. Clients pay a per-user monthly fee which is packaged into annual contracts.

How has COVID-19 impacted the business?

In todays fast-paced business environment, everything is evolving in a chain reaction of events. COVID-19 has radically accelerated a number of existing trends. There is no way to actively measure and anticipate cultural shifts.

Using the latest in machine learning and by tracking digital conversations on the internet, we enable monitoring key microtrends and issues that define our culture today.

In todays fast-paced business environment, everything is evolving in a chain reaction of events. COVID-19 has radically accelerated a number of existing trends. There is no way to actively measure and anticipate cultural shifts.

Using the latest in machine learning and by tracking digital conversations on the internet, we enable monitoring key microtrends and issues that define our culture today.

What was the funding process like?

Fundraising in the midst of a pandemic was definitely unique and nuanced. There were ironically more meetings, albeit via Zoom; and, since everyone was working from home effectively stretching available work hours, the product level diligence was very thorough with several follow-up Zooms. The reference checks and background checks took on additional importance, but it was also a great opportunity to assess how nimble and proactive our investors could be in a difficult environment. We are extremely proud and excited by the partners we made in this funding round.

What are the biggest challenges that you faced while raising capital?

Time. The pandemic stretched the process beyond what we had expected, and so we were in a race against the clock to ensure we concluded fundraising before we needed to make major capital commitments to business requirements.

What factors about your business led your investors to write the check?

I believe investors back entrepreneurs first, and the company thereafter. We were able to demonstrate our commitment and conviction to the investors over a period of 3 months. Moreover, we were able to engage in a partnership with SAP.io right before our funding, which I would imagine provided a great deal of comfort around our go-to-market strategy. Our biggest focus is to launch several POCs with the aim at converting them into annual contracts. We are targeting to showcase the value of our products within the larger consumer sector in businesses spanning from beauty brands to automotive manufacturers to insurance companies.

What advice can you offer companies in New York that do not have a fresh injection of capital in the bank?

Proof of concept with a beta customer who can provide feedback and credibility is of paramount importance. Until then, believe in your product and your team, and keep pushing ahead. Most importantly, when you do raise capital, make sure it is with great collaborators and partners.

Proof of concept with a beta customer who can provide feedback and credibility is of paramount importance. Until then, believe in your product and your team, and keep pushing ahead. Most importantly, when you do raise capital, make sure it is with great collaborators and partners.

Where do you see the company going now over the near term?

We are currently expanding our development team to increase the velocity of our product iteration process. We plan to continuously track and analyze our clients usage of the platform to define our short-term product roadmap, while also building toward our longer-term goals and company vision.

Sign up today

More here:

NWO.ai Raises $3.5M to Predict Cultural Shifts Before They Happen Using AI - AlleyWatch

Amazon.com has revealed plans to install AI-powered video cameras in its branded delivery vans, in a move that the worlds largest e-commerce firm says would improve safety of both drivers and the communities in which they deliver.

The company recently started rolling out camera-based safety technology across its delivery fleet, it said in an emailed statement on Wednesday.

This technology will provide drivers real-time alerts to help them stay safe when they are on the road, the statement added.

The companys plans were earlier disclosed in an instructional video about the cameras, reported earlier in the day by technology publication the Information. (https://bit.ly/2MPF68U)

Amazon said in the video that the cameras, developed by transportation technology company Netradyne, use artificial intelligence (AI) to provide warnings about speeding and distracted driving among other things.

They have been shown to reduce collisions and improve driver behavior, Amazons Karolina Haraldsdottir, a senior manager for last-mile safety, said in the video.

Amazon has come under some scrutiny in the past for accidents involving delivery drivers.

Our intention with this technology is to set up drivers for success and provide them with support for being safer on road and handling incidents if and when they happen, Haraldsdottir said in the video.

The video explains that while the cameras will constantly record video, they only upload footage if triggered by actions like hard braking, driver drowsiness, following vehicles too closely.

(Reporting by Vishwadha Chander in Bengaluru; Editing by Rashmi Aich)

Topics Insurtech

Was this article valuable?

Yes No

Here are more articles you may enjoy.

Get automatic alerts for this topic.

More:

Amazon Plans Cameras with AI in Its Delivery Vans to Improve Driver Safety - Insurance Journal

via Shutterstock

Public infrastructure operators are joining a growing list of AI technology adopters as they seek to streamline operations and better manage assets.

Among the latest is the El Paso, Texas, water utility, which this week announced plans to work with AI services vendor KloudGin Inc.

The field service and asset management provider, based in Sunnyvale, Calif., also works with other state and local water and gas utilities along with telecom providers. El Paso Water will use KloudGins cloud-based platform for mobile field service and work management tasks related to the Texas utilitys regulatory compliance efforts.

The partners said the water utility would deploy KloudGin AI-based management platform to integrate maintenance, inventory and customer information systems that, for example, can calculate excess water usage during droughts and apply penalties to water bills.

In an email exchange, KloudGin CEO Vikram Takru said the companys AI platform is geared toward managing field services ranging from dispatching a utilities mobile work force and contacting contractors to call before you dig alerts.

This is a toolset made for the front-line field workers who may not be up to speed on all the latest technology, Takru said. To workers in these industries and fields that are traditionally low-tech or no-tech, AI and machine learning sound complex, but KloudGins AI-powered system provides a straightforward means to monitor routes, location of equipment and directions.

The asset management service also enables predictive and automated maintenance, including AI-based instructions that can be used to train field workers on-the-fly. This helps increase worker productivity, safety and knowledge, and gets the right assets and workers to the right place at the right time, Takru added.

The shift to AI-based asset management is accelerating as utilities hustle to upgrade aging infrastructure. Thats especially true for parched southwestern U.S. border cities like El Paso. According to recent reports, El Paso Water has been forced to raise water and wastewater rates. Local media also report the water utility has been disconnecting service to customers who fail to pay their water bills.

Cloud-based AI field services also address the limitations of legacy field service and asset management systems, which KloudGins Takru describes as clunky, old, siloed [and] manually integrated. Those limitations hinder field operations and modernization efforts.

KloudGin promotes its AI-based approach as a step toward improving operational efficiency via a platform that manages both assets and field workers. Were seeing utilities and enterprisesexpand rapidly with solar farms, wind farms, charging stations and energy storage, Takru said.

Theyre going to need a platform to manage that, especially with product and manufacturing companies becoming service companies, he added. Everything is scheduling, route optimization,appointments and smart assets.

Related

About the author: George Leopold

George Leopold has written about science and technology for more than 30 years, focusing on electronics and aerospace technology. He previously served as executive editor of Electronic Engineering Times. Leopold is the author of "Calculated Risk: The Supersonic Life and Times of Gus Grissom" (Purdue University Press, 2016).

Read more:

Utilities Turn to AI to Manage Assets, Field Operations - EnterpriseAI

The COVID-19 pandemic has affected people's behavior everywhere. Fear, apprehensiveness, sadness, anxiety, and other troublesome feelings have become part of the daily lives of many families since the first cases of the disease were officially recorded early last year.

These turbulent feelings are often expressed in dreams reflecting a heavier burden of mental suffering, fear of contamination, stress caused by social distancing, and lack of physical contact with others. In addition, dream narratives in the period include a larger proportion of terms relating to cleanliness and contamination, as well as anger and sadness.

All this is reported in astudy publishedinPLOS ONE. The principal investigator was Natlia Bezerra Mota, a neuroscientist and postdoctoral fellow at the Brain Institute of the Federal University of Rio Grande do Norte (UFRN), in Brazil.

The study was part of Mota's postdoctoral research and was supervised by Sidarta Ribeiro at UFRN and Mauro Copelli at the Federal University of Pernambuco (UFPE), both of whom are affiliated with the Neuromathematics Research, Innovation and Dissemination Center (NeuroMat).

Neuromat is hosted by the University of So Paulo (USP) and is one of many Research, Innovation and Dissemination Centers (RIDCs) supported by So Paulo Research Foundation - FAPESP.

The results are consistent with the hypothesis that dreams reflect the challenges of waking-life experience during the pandemic, and that the prevalence of negative emotions such as anger and sadness during the period reflects a higher emotional load to be processed, the authors write.

According to Mota, the findings are corroborated by those of other studies published later by researchers in the United States, Germany, and Finland.

The Brazilian study was initially reported in May in apreprint posted to medRxiv, and not yet peer-reviewed at that time (read more at:agencia.fapesp.br/33664). "It's the first study on the subject to look empirically at these signs of mental suffering and their association with the peculiarities of dreams during the pandemic," Mota told.

For Ribeiro, the authors of the study managed to document the continuity between what happens in the dream world and people's mental lives, especially psychological distress. "This is interesting from the standpoint of dream theory," he said. "Another point worth highlighting is that they did so quantitatively, using mathematics to extract semantics."

The group deployed natural language processing tools to analyze 239 dream reports by 67 subjects produced in March and April 2020, shortly after the World Health Organization (WHO) declared a pandemic.

According to Mota, researchers at USP, UFRN, and the Federal Universities of Minas Gerais (UFMG), Rio Grande do Sul (UFRGS) and Rio de Janeiro (UFRJ) are conducting a multicentric study involving the analysis of data collected during a longer period (from the start of the pandemic through July) to see how dreams are affected by the deaths of family members, loved ones, friends and co-workers. "The plan is to publish the findings as soon as they're ready so that mental health strategies can be based on this knowledge," she said.

Dream accounts recorded by the volunteers using a smartphone app were transcribed and analyzed using three software tools. The first focused on discourse structure, word count, and connectedness.

The other two focused on content. One ranged words in certain emotional categories against a list associated with positive and negative emotions. The other used a neural network to detect semantic similarity to specified keywords, such as contamination, cleanliness, sickness, health, death and life.

In theirPLOS ONEpublication, the researchers say "the significant similarity to 'cleanness' in dream reports points towards new social strategies (e.g. use of masks, avoidance of physical contact) and new hygiene practices (e.g. use of hand sanitizer and other cleaning products) that have become central to new social rules and behavior. Taken together, these findings seem to show that dream contents reflect the different sources of fear and frustration arising out of the current scenario".

Mota noted that more suffering was expressed in the dream reports submitted by female volunteers, although this was detected indirectly. "There are studies on gender difference in the literature. Women report more negative dreams and nightmares. I think this has to do with women's history and daily lives, with working a double or triple shift, and the heavier mental burden entailed by concerning themselves with a job plus the home and children. The pandemic has made this worse," she said.

Reference: Mota NB, Weissheimer J, Ribeiro M, et al. Dreaming during the Covid-19 pandemic: Computational assessment of dream reports reveals mental suffering related to fear of contagion. PLOS ONE. 2020;15(11):e0242903. doi:10.1371/journal.pone.0242903

This article has been republished from the following materials. Note: material may have been edited for length and content. For further information, please contact the cited source.

The rest is here:

AI Techniques Analyze Dream Content During the Pandemic - Technology Networks

TOKYO, Feb. 5, 2021 /PRNewswire/ -- Edgecortix Inc. (Headquarters: Tokyo, Japan, CEO: Sakyasingha Dasgupta, hereinafter called "EdgeCortix") has signed a partnership with PALTEK Corporation (Headquarters: Yokohama, Japan, President: Naohide Yabuki, Securities Code: 7587, hereinafter called "PALTEK"), which sells and designs semiconductors and related products. Since its founding in 1982,PALTEK has used FPGAs to provide product development support and technical support to Japanese electronics manufacturers. The company's strength, cultivated through its semiconductor sales, lies in its ability to support customers from proposal through to mass production, and along with integrated design services.

Through this partnership, EdgeCortix will provide PALTEK, with their low-latency, energy-efficient hardware IP (intellectual property) Dynamic Neural Accelerator and MERA compilers, to be implemented on the Xilinx Accelerator Card "Alveo U50 Data Center Accelerator Card" *1. This enables high-throughput, real-time and low power edge AI applications in areas such as ADAS (advanced driver assistance systems), autonomous driving, robots, smart cities, drones, industry 4.0, etc.

Background and outline of this collaboration

In recent years, deep learning, which is part of a broader family of machine learning methods based on artificial neural networks, has realized dramatic improvements in accuracy in many fields such as machine vision and voice processing. However, majority of the current practical applications are based on cloud-infrastructure using GPUs. In general for edge computing, there are many use cases that cannot be processed efficiently in the cloud due to challenges arising from communication volume, load, and latency during inference, with the performance of inference processing on edge devices using streaming data becoming a critical issue.

EdgeCortix is a fabless semiconductor design company, co-designing the hardware and software for low latency artificial intelligence inference processors on embedded and telco edge devices. Through this distributor agreement, EdgeCortix's unique AI hardware IP, the Dynamic Neural Accelerator*2, can be implemented on the Xilinx Alveo U50 FPGA accelerator cards provided by PALTEK. This enables ultra-high speed and reduced power consumption (as compared to CPUs and GPUs), while processing of complicated arithmetic operations along with large amounts of data, typical in deep learning.

Against this background, by combining PALTEK's long standing experience in design and sales of electronic equipment and technical support from proposal to mass production, along with EdgeCortix's edge AI acceleration focused technology, we can rapidly bring high performance AI inference capabilities to edge devices. Additionally, this partnership will help provide easily deployable edge AI solutions to companies and system integrators without the need for deep know how on FPGA hardware or software.

Main Initiatives / Collaboration Measures

Establishing this distributor agreement, we will combine the strengths of both companies, accelerate the development and mass production of edge AI solutions that utilize FPGAs, and propose and execute efficient AI applications for meeting the needs of the market. Specifically, solutions that integrate EdgeCortix's Dynamic Neural AcceleratorIP DNA-F200 & DNA-F100 (FPGA bitstreams) with Alveo U50 FPGA accelerator cards, along with EdgeCortix's MERA dataflow compiler will be offered.

Naohide Yabuki, President of PALTEK Corporation, had the following to say about this partnership:

"Currently, various AI services are on the market, but most of them are processed on CPUs and GPUs in the cloud. However, there is a need for low latency in fields such as robots, smart cities, and Industry 4.0, which we are targeting. Along with this, edge AI with FPGAs is increasing as many customers demand low power consumption and long-term supply. EdgeCortix focuses on edge AI and provides solutions specialized for lower latency and lower power consumption. We are very much looking forward to achieve synergistic effects by proposing such edge AI focused solutions together with the Xilinx products we handle. "

Sakyasingha Dasgupta, CEO of Edgecortix, Inc., had the following to say about this partnership:

"We are pleased to establish this partnership with PALTEK Corporation and are confident that it will enable growth in edge AI hardware acceleration business involving FPGAs. We feel very strongly that, with complete integrated solutions leveraging our unique technology, together with PALTEK Corporation, we can enhance the practical application of edge AI products and the overall customer experience. Since 1982, PALTEK Corporation has a wealth of experience in electronic product development and sales, supporting customers from proposal stages to mass production. Leveraging this experience, we will offer our AI processor IP and compiler technology for extremely fast inference on edge devices and datacenters using FPGAs, requiring very little effort for solution deployment."

About Dynamic Neural Accelerator IP DNA-F200 from Edge Cortix

DNA-F200isthe latest member of the EdgeCortix Dynamic Neural Acceleratordataflow architecture basedIPfamilyfor deep neural network(DNN)inference applicationon FPGAs.It is designed for the latest Xilinx ALVEOU50/U50LVadaptable accelerator cardswith HBM support.DNA-F200(3.7 INT8 TOP/s at 300 MHz) and its predecessor DNA-F100 (2.3 INT8 TOP/s at 275 MHz) area high-performanceconvolution neural network (CNN)inference IP optimized forultra-low latency, energy-efficient and high throughput workloads for streaming data.Especially, designed forblazingly fastedge AI applications.DNA-F200/F100runs with highly optimized instructions setfor INT8 bit batch-size 1 inferenceand supports all mainstream convolutional neural networks, such as ResNet, YOLO, SSD, MobileNet, FPN,MonoDepthetc.It is supported on any Xilinx or custom board based on Xilinx FPGAs that has a Vitisplatform, out of the box. This includes bothMPSoCclass and datacenter class ALVEOFPGAs.

EdgeCortixoffers the DNA-F200 bitstream along with their proprietary MERA compiler. This enables nearly zero effort deployment of deep neural networks designed in most popular frameworks likePytorchandTensorflow-lite directly on the FPGA. Developed on top of open-source machine learning compiler Apache TVM*3, MERA enables machine learning engineers to optimize and run networks designed for CPUs or GPUs, with INT8 bit quantization out of the box on the DNA IP implemented onXilinxALVEOFPGAs.The compiler automatically identifies which parts of the neural network can be offloaded to the accelerator and automatically offloads new or unknown operators to the host processor.The MERA compiler also comes with a built-in simulator and interpreter. After compilation, customers can use these tools for cycle-accurate performance simulation without testing on the hardware as well as to quantify the impact of INT8 bit quantization on network accuracy.

Further details of DNA-F200 and the MERA dataflow compiler is available from: https://www.edgecortix.com/

Terms used

1. Alveo U50 Data Center Accelerator Card - An accelerator card from Xilinx that demonstrates excellent performance for high-speed processing such as DNN inference. For more information, please visit https://www.xilinx.com/products/boards-and-kits/alveo/u50.html

2. Dynamic Neural Accelerator An AI hardware architecture provided by EdgeCortix focusing on energy-efficient, low latency AI inference on embedded and telco edge devices. DNA-F100/F200 are specifics versions of the IP family targeting FPGAs.

3. Apache TVM is an open source, end-to-end machine learning compiler for CPUs, GPUs and accelerators. https://tvm.apache.org/

4. Vitis is an integrated software framework optimized for Xilinx FPGAs and Versal ACAP hardware platforms. https://www.xilinx.com/products/design-tools/vitis/vitis-platform.html

5. Dynamic Neural Accelerator and MERA are registered trademarks or trademarks of EdgeCortix in Japan and other countries.

6. The Xilinx name and other brand names mentioned in this press release are registered trademarks or trademarks of Xilinx in the United States and other countries. All other names belong to their respective owners.

About PALTEK Corporation

Since its founding in 1982, PALTEK has been selling semiconductor products in Japan and overseas to electronics manufacturers, as well as providing contract design services for hardware and software, and as a partner in customer product development, from specification studies to trial production. They support development and mass production. PALTEK will contribute to the development of their customers by providing optimal solutions for them based on their corporate philosophy of "coexistence with diverse entities."

For more information about PALTEK, please visit https://www.paltek.co.jp/en/index.html

About Edgecortix, Inc.

EdgeCortix, Inc. was founded in 2019 with the corporate mission to "bring cloud-level performanceto the embedded edge, for low latency, low cost, and energy-efficient deep neural network inference". The company's strength is in its unique artificial intelligence processor technology designed with an integrated hardware and software co-design approach. EdgeCortix has raised a total investment of 525 Million Yen till date from investors in Japan, Singapore and USA. It has a proven track record of achievements with existing partnerships with multiple companies in the electronic manufacturing industry. Taking a software centric approach to AI hardware IP creation, their Dynamic Neural Accelerator IP core and MERA compiler is designed to work with little effort across custom ASICs and FPGAs.

For more information about Edgecortix, please visit https://www.edgecortix.com

SOURCE Edgecortix, Inc.

http://www.edgecortix.com

Original post:

EDGECORTIX partners with PALTEK to bring Edge AI Hardware Acceleration Solutions to Market - PRNewswire

Computer science professor Tina Eliassi-Rad says shes proud to be named on an industry list of 100 Brilliant Women in AI Ethics, which identifies her as one of the top thinkers in the male-dominated field of artificial intelligence. But shes even prouder of what the carefully-curated list represents.

Part of the issue in a field such as computer science is that women and other under-represented minorities arent always seen. Initiatives like this one show that there are a lot of women who are qualified to do this work, says Eliassi-Rad.

Mia Shah-Dand, the CEO of the Oakland, California-based research firm Lighthouse3, created the annual list in 2018. Shah-Dand says she wanted to provide a rebuttal to technology leaders who complained that they couldnt find accomplished, diverse women to hire.

I was a little frustrated with all the times I would hear, There just arent enough qualified women, says Shah-Dand. Its the same old excuse. Well, we have an entire directory of qualified women now. There is no excuse. At this point in 2021, if you have only men on your staff, its intentional.

According to recent research by the World Economic Forum, women hold only 26% of data and artificial intelligence jobs across the globe, and even fewer have senior roles.

Shah-Dand says she included Eliassi-Rad on her 2021 list because of the professors extensive research on racial, gender and other baked-in biases in artificial intelligence algorithms.

Her emphasis on algorithmic accountability and fairness was particularly interesting, says Shah-Dand.

Algorithms, which scan large amounts of data and find whatever information its creators want, are increasingly part of our everyday lives. For example, credit card fraud departments use algorithms to detect abnormal spending, while social media algorithms use viewer interests to determine which ads to run.

Eliassi-Rads research at Northeastern focuses on the unseen but overwhelming influence that artificial intelligence algorithms can make in peoples lives, especially in social media.

Part of the problem with algorithms is that they can impact life-altering decisions if theyre used in criminal justice or even your credit score, says Eliassi-Rad. Microlenders, or individuals who issue small loans, will often check a candidates Facebook and Twitter feeds when deciding whether to grant a loan. A chance connection with someone who has defaulted on a loan could trigger a denial, says Eliassi-Rad.

Sometimes if you dont get the right loan in life, you cant better yourself, she says.

Eliassi-Rads career in computer science was sparked by her fathers early work with autonomous vehicles. She avidly read the many magazines he brought home and decided computer science was the perfect balance between math and electrical engineering. Her focus recently sharpened as she learned about the different class, race, and gender biases in machine learning.

She likens the data used in algorithms to an iconic photo of a police officers German shepherd attacking a Black high school student during a 1963 civil rights event in Birmingham, Alabama.

The German shepherd isnt racist, its the people teaching the dog, Eliassi-Rad says. Even if the data used in an algorithm isnt biased, the algorithm may still produce biased findings.

As you are developing an algorithm you are making choices, and those choices have consequences, Eliassi-Rad says.

Eliassi-Rad and Shah-Dand say the list of top women in AI ethics does more than provide a roster of qualified computer science professionals who also happen to be female, LGTBQ, or women of color. It creates a community to foster networking and support while providing role models for future generations.

Its sort of like a sisterhood, says Eliassi-Rad, who received an Outstanding Mentor Award from the Office of Science at the US Department of Energy in 2010. I hope young women see this and think, I can be somebody like this person.

For media inquiries, please contact media@northeastern.edu.

Originally posted here:

She was named one of the 100 most brilliant women in AI ethics - News@Northeastern

The U.S. Food and Drug Administration has authorized the use of CLEW Medical's artificial intelligence tool to predict hemodynamic instability in adult patients inintensive care units, the company announced on Wednesday.

The tool, CLEWICU, uses AI-based algorithms and machine learning models to identify the likelihood of occurrence of significant clinical events for ICU patients.

CLEW says the clearance is the FDA's first for such a device.

"AI can be a powerful force for change in healthcare, enabling assessment of time-critical patient information and predictive warning of deterioration that could enable better informed clinical decisions and improved outcomes in the ICU," said Dr. David Bates, medical director of clinical and quality analysis in information systems at Mass General Brigham and CLEW Advisory Board member, in a statement.

WHY IT MATTERS

Hemodynamic instability is a common COVID-19 complication, so CLEWICU's predictive capabilities could prove especially useful during the ongoing pandemic particularly given ICUs' strained resources around the country.

By analyzing patient data from various sources, including electronic health records and medical devices, CLEWICU provides a picture of overall unit status and helps identify individuals whose conditions are likely to deteriorate.

According to the company, the system notifies users of clinical deterioration up to eight hours in advance, enabling early intervention. The system also identifies low-risk patients who are unlikely to deteriorate, thus potentially enabling better ICU resource management and optimization.

"CLEW's AI-based solution is a huge leap forward in ICU patient care, providing preemptive and potentially lifesaving information that enables early intervention, reduces alarm fatigue and can potentially significantly improve clinical outcomes," said Dr. Craig Lilly of University of Massachusetts Medical School in a statement.

THE LARGER TREND

The FDA granted emergency use authorization to CLEWICU back this past June. The tool was among several AI-powered technology innovations developed, or modified, in response to the ongoing pandemic.

Mayo Clinic Chief Information OfficerCris Ross said in December that AI has been crucial in understanding the pandemic. He noted the variety of COVID-19-specific use cases, while he also flaggedthe risk of algorithmic bias.

"We know that Black and Hispanic patients are infected and die at higher rates than other populations. So we need to be vigilant for the possibility that that fact about the genetic or other predisposition that might be present in those populations could cause us to develop triage algorithms that might cause us to reduce resources available to Black or Hispanic patients because of one of the biases introduced by algorithm development," said Ross.

ON THE RECORD

"We are proud to have received this landmark FDA clearance and deliver a first-of-its-kind product for the industry, giving healthcare providers the critical data that they need to prevent life-threatening situations," said Gal Salomon, CLEW CEO, in a statement.

Kat Jercich is senior editor of Healthcare IT News.Twitter: @kjercichEmail: kjercich@himss.orgHealthcare IT News is a HIMSS Media publication.

More here:

FDA issues landmark clearance to AI-driven ICU predictive tool - Healthcare IT News

HealthTensor, an artificial intelligence company creating software to help augment medical decision-making, has raised a $5 million in a seed round of financing led by Calibrate Ventures, TenOneTen Ventures and Susa Ventures.

WHY IT MATTERS

The round also includes hospitals and physicians, including a medical officer at Amazon Health. Funds will be used to scale the company's software engineering and implementation team to keep up with demand from major health systems, the vendor said.

HealthTensor's software functions between physicians and the troves of raw medical data from any given patient, which often is more than any individual doctor can handle. The company uses advanced algorithms to do AI-enabled diagnosiswith the aim of ensuring no medical condition is overlooked. The software was designed with the physician workflow in mind, enabling frictionless adoption of the product by users, the company contended.

"HealthTensor makes me a better doctor because it allows me to spend less time in front of the computer and more time in front of the patient," said Dr. Tasneem Bholat, an early user of HealthTensor's software. "HealthTensor synthesizes all the data from the patient's chart, saving me from doing chart biopsy and surfacing diagnoses I might have otherwise missed."

The company's software currently is integrated within several hospitals and will expand to more in the coming months, the vendor reported.

THE LARGER TREND

The use of AI in healthcare has been on the rise throughout 2020. According to some experts, 2021 could be a big year for AI and machine learning.

"AI had become mythical, but 2021 looks set to be the year where it may come into its own in the health sector, along with the use of automation," said Dr. Sam Shah, chief medical strategy officer at Numan and former director of digital development at NHSX. "During the next year, we are likely to see more solutions that support, not only imaging, but also the quality of reporting,as well as the greater use of natural language processing.

"The combination of these technologies will help improve efficiency in health systems as they begin to recover from the pandemic," he said.

ON THE RECORD

"We think of HealthTensor as an AI-powered medical resident that is focused specifically on the tedious, data-driven aspects of medicine, which is what computers do best," said Eli Ben-Joseph, cofounder and CEO of HealthTensor.

"Many doctors are forced to spend a majority of their day focused on data aggregation from medical records, which leads to missed diagnoses, patient dissatisfaction and physician burnout. HealthTensor frees up the physician to focus on the conceptual and emotional aspects of medicine, which is what humans do best."

"HealthTensor makes doctors' lives easier and helps provide better patient care, ultimately generating revenue for hospitals, making it one of the rare startups that has massive global potential for both patients and healthcare providers," said Jason Schoettler, general partner at Calibrate Ventures.

Twitter:@SiwickiHealthITEmail the writer:bsiwicki@himss.orgHealthcare IT News is a HIMSS Media publication.

More here:

HealthTensor raises $5M for its AI-based medical diagnosis tools - Healthcare IT News

For all its comparisons to the human brain, AI still isnt much like us. Maybe thats alright. In the animal kingdom, brains come in all shapes and sizes. So, in a new machine learning approach, engineers did away with the human brain and all its beautiful complexityturning instead to the brain of a lowly worm for inspiration.

Turns out, simplicity has its benefits. The resulting neural network is efficient, transparent, and heres the kicker: Its a lifelong learner.

Whereas most machine learning algorithms cant hone their skills beyond an initial training period, the researchers say the new approach, called a liquid neural network, has a kind of built-in neuroplasticity. That is, as it goes about its worksay, in the future, maybe driving a car or directing a robotit can learn from experience and adjust its connections on the fly.

In a world thats noisy and chaotic, such adaptability is essential.

The algorithms architecture was inspired by the mere 302 neurons making up the nervous system of C. elegans, a tiny nematode (or worm).

In work published last year, the group, which includes researchers from MIT and Austrias Institute of Science and Technology, said that despite its simplicity, C. elegans is capable of surprisingly interesting and varied behavior. So, they developed equations to mathematically model the worms neurons and then built them into a neural network.

Their worm-brain algorithm was much simpler than other cutting-edge machine learning algorithms, and yet it was still able to accomplish similar tasks, like keeping a car in its lane.

Today, deep learning models with many millions of parameters are often used for learning complex tasks such as autonomous driving, Mathias Lechner, a PhD student at Austrias Institute of Science and Technology and study author, said. However, our new approach enables us to reduce the size of the networks by two orders of magnitude. Our systems only use 75,000 trainable parameters.

Now, in a new paper, the group takes their worm-inspired system further by adding a wholly new capability.

The output of a neural networkturn the steering wheel to the right, for instancedepends on a set of weighted connections between the networks neurons.

In our brains, its the same. Each brain cell is connected to many other cells. Whether or not a particular cell fires depends on the sum of the signals its receiving. Beyond some thresholdor weightthe cell fires a signal to its own network of downstream connections.

In a neural network, these weights are called parameters. As the system feeds data through the network, its parameters converge on the configuration yielding the best results.

Usually, a neural networks parameters are locked into place after training, and the algorithms put to work. But in the real world, this can mean its a bit brittleshow an algorithm something that deviates too much from its training, and itll break. Not an ideal result.

In contrast, in a liquid neural network, the parameters are allowed to continue changing over time and with experience. The AI learns on the job.

This adaptibility means the algorithm is less likely to break as the world throws new or noisy information its waylike, for example, when rain obscures an autonomous cars camera. Also, in contrast to bigger algorithms, whose inner workings are largely inscrutable, the algorithms simple architecture allows researchers to peer inside and audit its decision-making.

Neither its new ability nor its still-diminutive stature seemed to hold the AI back. The algorithm performed as well or better than other state-of-the art time-sequence algorithms in predicting next steps in a series of events.

Everyone talks about scaling up their network, said Ramin Hasani, the studys lead author. We want to scale down, to have fewer but richer nodes.

An adaptable algorithm that consumes relatively little computing power would make an ideal robot brain. Hasani believes the approach may be useful in other applications that involve real-time analysis of new data like video processing or financial analysis.

He plans to continue dialing in the approach to make it practical.

We have a provably more expressive neural network that is inspired by nature. But this is just the beginning of the process, Hasani said. The obvious question is how do you extend this? We think this kind of network could be a key element of future intelligence systems.

At a time when big players like OpenAI and Google are regularly making headlines with gargantuan machine learning algorithms, its a fascinating example of an alternative approach headed in the opposite direction.

OpenAIs GPT-3 algorithm collectively dropped jaws last year, both for its sizeat the time, a record-setting 175 billion parametersand its abilities. A recent Google algorithm topped the charts at over a trillion parameters.

Yet critics worry the drive toward ever-bigger AI is wasteful, expensive, and consolidates research in the hands of a few companies with cash to fund large-scale models. Further, these huge models are black boxes, their actions largely impenetrable. This can be especially problematic when unsupervised models are trained on the unfiltered internet. Theres no telling (or perhaps, controlling) what bad habits theyll pick up.

Increasingly, academic researchers are aiming to address some of these issues. As companies like OpenAI, Google, and Microsoft push to prove the bigger-is-better hypothesis, its possible serious AI innovations in efficiency will emerge elsewherenot despite a lack of resources but because of it. As they say, necessity is the mother of invention.

Image Credit: benjamin henon / Unsplash

Originally posted here:

New 'Liquid' AI Learns Continuously From Its Experience of the World - Singularity Hub

Data science, machine learning, and AI experts highlight the top AI and ML trends they expect to shape the data science industry in 2021

The ongoing impact of Covid-19 is still affecting organizations nearly a year since the pandemic began, with business leaders continuing to leverage technology in order to navigate the crisis. According to Oxylabs dedicated AI and ML advisory board, some of the most important trends in 2021 will include the increased use of ethical AI for diversity, accountability, and model explainability, alongside increased instances of AI solving challenging real-world problems.

Oxylabs advisory board comprises the leading figures in the machine learning, AI, and data science industries and its members outline what they believe are the most important data science predictions for the year ahead:

Firstly, Pujaa Rajan, Machine Learning Engineer at Stripe, USA Ambassador at Women in AI andGoogleDeveloper MLExpert, believes COVID-19 will instigate a renewed enthusiasm for the application of edge AI in the healthcare industry and the use of ethical AI:

Covid-19 defined 2020 and although development in healthcare has historically been slower than other industries due to regulation this year will see a focus on edge AI in the healthcare industry and other industries. This will lead to the ability to run ML models locally, and tiny ML, resulting in smaller sized ML models that fit on smaller devices like phones. Businesses will focus on these specific, technical areas because they are related to data privacy and security, which the general public and government increasingly care about.Model explainability and interpretability is a space that the government, healthcare companies and finance companies are all actively exploring because of technical curiosity and business motivations. Many leaders will also finally prioritise AI ethics, diversity, inclusion, model explainability, and model interpretability after public outrage at many bad, biased, and unethical applications of AI. On the other hand, the biggest AI news last year was OpenAIs GPT-3, so I expect continued innovation in large NLP models. Software and hardware are like yin and yang. Since the larger models will need more efficient hardware, neural network accelerators will be a hot space.

Ali Chaudhry, PhD researcher, Artificial Intelligence atUCL, sees AI as having have more of a contribution in solving challenging real-world problems in 2021:

I think there will be more focus on fairness, transparency, accountability and explainability in AI systems this year, hence, we can expect more regulations from governments around the globe. We will also see AIs contribution in solving more challenging real-world problems, similar to the protein folding problem that was recently solved by AI. In terms of AI techniques that are set to emerge, there will be more real-world applications of Reinforcement Learning (RL) algorithms and RL will also retain its top position in academia.

Another prediction comes from Gautam Kedia, Machine Learning Engineering Manager at Stripe, ex-Applied Scientist Lead at Microsoft, previously Head of Applied ML at Lyft. He considers how AI-generated content could finally become mainstream across multiple sectors:

AI-generated content will become mainstream and in the next few years, I expect truly generative models to be producing logos, short stories, stock images, voiceovers and workouts, DALL-E is just a start and I believe this content will gradually start to pass the Turing Test. Self-driving cars will also take another step forward and I expect Waymo to start a taxi service directly competing with Uber & Lyft. Tesla will also release the much-awaited Full Self Driving computer.

Finally, Jonas Kubilius, AI researcher, Marie Skodowska-Curie Alumnus, and Co-Founder of Three Thirds is optimistic about the implementation of AI in healthcare but also has fears that AI investment may suffer:

Im certainly optimistic about AI-driven solutions making a greater impact in the healthcare sector and drug discovery, however, my only concern is the economic impact of the global COVID-19 pandemic. It may well be that there is a slowdown of investments in AI-driven solutions and research labs, forcing companies to justify any investments they make and focus very clearly on problems where AI brings a clear added value. With ever increasing pressure on governments and organisations to take action in regard to climate change, I expect to see more AI-driven solutions being leveraged in this field. Particularly in the areas that could benefit from the optimisation of manufacturing and logistics processes to reduce the impact they have on the environment.

Sign up for the free insideBIGDATAnewsletter.

Join us on Twitter:@InsideBigData1 https://twitter.com/InsideBigData1

See the original post:

AI Solving Real-world Problems and AI Ethics Among Top Trends for 2021, According to Oxylabs' AI and ML Advisory Board - insideBIGDATA