How to leverage AI to boost care management success

Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! Watch here.

Sixty percent of American adults live with at least one chronic condition, and 12% with five or more. They spend exponentially more on healthcare than those without any chronic conditions. For instance, 32% of adults with five or more chronic conditions make at least one ER visit each year. On top of that, 24% have at least one inpatient stay, in addition to an average of 20 outpatient visits — up to 10 times more than those without chronic conditions. In fact, 90% of America’s $4 trillion healthcare expenditures are for people with chronic and mental health conditions, according to the Centers for Disease Control and Prevention (CDC).

The fundamental way healthcare organizations reduce these costs, improve patient experience and ensure better population health is through care management. 

In short, care management refers to the collection of services and activities that help patients with chronic conditions manage their health. Care managers proactively reach out to patients under their care and offer preventative interventions to reduce hospital ER admissions. Despite their best efforts, many of these initiatives provide suboptimal outcomes.

Why current care management initiatives are ineffective

Much of care management today is performed based on past data

For instance, care managers identify patients with the highest costs over the previous year and begin their outreach programs with them. The biggest challenge with this approach, according to our internal research, is nearly 50-60% of high-cost patients were low-cost in the previous year. Without appropriate outreach, a large number of at-risk patients are left unattended with the reactive care management approach. 


MetaBeat 2022

MetaBeat will bring together thought leaders to give guidance on how metaverse technology will transform the way all industries communicate and do business on October 4 in San Francisco, CA.

Register Here

The risk stratification that the care management team uses today is a national model

These models are not localized, so understanding the social determinants of individual locations is not considered.

The care management team’s primary focus is chiefly on transition of care and avoiding readmissions

Our experience while working with different clients also points to the fact that readmissions contribute only 10-15% of total admission. The focus on proactive care management and avoiding future avoidable emergency room and hospital admission is lacking. This is key to success in value-based care models.

In any given year, high-cost patients can become low-cost

Without such granular understanding, outreach efforts can be ineffective in curbing the cost of care.

How AI can boost care management success

Advanced analytics and artificial intelligence (AI) open up a significant opportunity for care management. Health risks are complex, driven by a wide range of factors well beyond just one’s physical or mental health. For example, a person with diabetes is at higher risk if they also have low-income and limited access to medical services. Therefore, identifying at-risk patients’ needs to consider additional factors to encompass those most in need of care.

Machine learning (ML) algorithms can evaluate a complex range of variables such as patient history, past hospital/ER admissions, medications, social determinants of health, and external data to identify at-risk patients accurately. It can stratify and prioritize patients based on their risk scores, enabling care managers to design their outreach to be effective for those who need it most. 

At an individual level, an AI-enabled care management platform can offer a holistic view of each patient, including their past care, current medication, risks, and accurate recommendations for their future course of action. For the patient in the example above, AI can equip care managers with HbA1C readings, medication possession ratio, and predictive risk scores to deliver proper care at the right time. It can also guide the care manager regarding the number of times they should reach out to each patient for maximum impact.

Unlike traditional risk stratification mechanisms, modern AI-enabled care management systems are self-learning. When care managers enter new information about the patient — such as latest hospital visit, change in medication, new habits, etc. — AI adapts its risk stratification and recommendations engine for more effective outcomes. This means that the ongoing care for every patient improves over time.

Why payers and providers are reluctant to embrace AI in care management

In theory, the impact of AI in care management is significant — both governments and the private sector are bullish on the possibilities. Yet, in practice, especially among those who use the technology every day, i.e., care managers, there appears to be reluctance. With good reason.

Lack of localized models

For starters, many of today’s AI-based care management solutions aren’t patient-centric. Nationalized models are ineffective for most local populations, throwing predictions off by a considerable margin. Without accurate predictions, care managers lack reliable tools, creating further skepticism. Carefully designed localized models are fundamental to the success of any AI-based care management solution.

Not driven by the care manager’s needs

On the other hand, AI today is not ‘care manager-driven’ either. A ‘risk score’ or the number indicating the risk of any patient gives little to the care manager. AI solutions need to speak the user’s language, so they become comfortable with the suggestions. 

Healthcare delivery is too complex and critical to be left to the black box of an ML algorithm. It needs to be transparent about why each decision was made — there must be explainability that is accessible to the end-user. 

Inability to demonstrate ROI

At the healthcare organizational level, AI solutions must also demonstrate ROI. They must impact the business by moving the needle on its key performance indicators (KPIs). This could include reducing the cost of care, easing the care manager’s burden, minimizing ER visits, and other benefits. These solutions must provide healthcare leaders with the visibility they need into hospital operations as well as delivery metrics.

What is the future of AI in care management?

Despite current challenges and failures in some early AI projects, what the industry is experiencing is merely teething troubles. As a rapidly evolving technology, AI is adapting itself to the needs of the healthcare industry at an unprecedented pace. With ongoing innovation and receptiveness to feedback, AI can become the superpower in the armor of healthcare organizations.

Especially in proactive care management, AI can play a significant role. It can help identify at-risk patients and offer care that prevents complications or emergencies. It can enable care managers to monitor progress and give ongoing support without patients ever visiting a hospital to receive it. This will, in turn, significantly reduce the cost of care for providers. It will empower patients to lead healthy lives over the long term and promote overall population health.

Pradeep Kumar Jain is the chief product officer at HealthEM AI.


Welcome to the VentureBeat community!

DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.

If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.

You might even consider contributing an article of your own!

Read More From DataDecisionMakers

Repost: Original Source and Author Link


The Windows 11 2022 update is here, but should you care?

Can you believe it’s been almost a year since Windows 11 launched? Back then, I was surprised that Microsoft was practically rushing a new version of Windows out the door. But, as I noted , Windows 11 ended up refining Microsoft’s desktop formula fairly well. My opinion hasn’t changed much since then (and yes, I’m still frustrated by the cleaner but less usable taskbar). Today, Microsoft will start rolling out the Windows 11 2022 update, the operating system’s first major revision, which brings better security, accessibility and a handful of gaming improvements. Mostly, though, it’ll make it easier for you to get future updates more quickly.

How do I get the Windows 11 2022 Update?

No surprise here: Head over to Windows Update in your Settings app and see if your computer is eligible for an upgrade. As usual, Microsoft says it’s taking a “measured and phased” approach, which means there’s a chance you won’t see the update immediately. The company will also highlight potential conflicts on your system — e.g., an incompatible app, an out of date driver — that will prevent you from getting the refreshed OS. This advice applies to both Windows 11 and Windows 10 users, though the latter should double-check their computer with the to ensure their hardware is compatible. (Check out our Windows 11 review for more details on upgrading from Windows 10.)

What’s this about faster updates?

Panos Panay, Microsoft’s Chief Product Officer, that the company was aiming to deliver “continuous innovation” and more frequent Windows 11 updates outside of the major annual release. That begins with the 2022 update. The company has “significantly reduced” the size of updates (around 450MB for many folks), as well as reduced their installation time, John Cable, the head of Windows Servicing and Delivery, said in a blog post.

The Windows 11 2022 update will also be more carbon aware, allowing you to schedule installations for times when your local grid is relying on cleaner energy sources like wind, hydro and solar. This functionality won’t be available everywhere, and we’re still waiting to hear more about how Microsoft will keep track of electric grid statistics. But theoretically, it’s a smart way to cut down on extraneous carbon emissions (and it’s something I’d love to see on phones, tablets and other devices).

Tabbed Explorer windows in the Windows 11 2022 update


So where are the new features?

At first glance, it’ll be difficult to tell you’re running the new update. It doesn’t bring any major UI changes, though Microsoft says it’ll be adding tabs to File Explorer in October. I’ve been using an early build of that feature, and it’s definitely helped to reduce my window clutter when moving between SD cards, OneDrive and my downloads folder. (Don’t worry, you can still fill your screen with multiple Explorer windows if you prefer.) 

You’ll also be able to tweak the Start Menu further by either adding more pinned apps, or more recommendations. Additionally, Microsoft is bundling the , which looks like a huge improvement over the existing Windows tool (and certainly lightyears ahead of Movie Maker).

Clipchamp video editor in the Windows 11 2022 update


Much like the improved Windows Update experience, the vast majority of new features in the 2022 update are under the hood. Those include a slew of accessibility upgrades that : system-wide live captions, which will initially appear at the top of your screen to help you feel engaged during video chats; natural sounding voices for the Narrator screen reader; as well as a preview of improved voice commands for using your PC and transcription.

More so than most tech companies, Microsoft has been over the last decade. That includes launching the , setting up a , and announcing a five-year commitment toward bridging the “Disability Divide.” The features debuting in this Windows 11 update are all driven by members of the Windows Accessibility team, giving them all a personal touch.

The new Focus Sessions experience, for example, is spearheaded by Alexis Kane, a product manager who has ADHD. She noticed how notifications were giving her more anxiety and disrupting her workflow, so she helped to create a way to minimize them without disabling notifications entirely. Focused Sessions reduces the noise of those alerts, but it also disables Task Bar badges and lets you time work sprints with the clock app.

Repost: Original Source and Author Link


Notable aims to improve AI in health care with new $100M

This article is part of a VB special issue. Read the full series: AI and the future of health care

Notable, an intelligent automation company focused on health care, today announced it received a $100 million series B funding round. The investment, led by ICONIQ Growth with participation from Greylock Ventures, Oak HC/ FT, and F-Prime, will be used to expand access to more health care providers and enhance its capabilities, so partners achieve a higher return on investment.

The reality is that many health care providers still use repetitive, manual workflows, which cost over $1 trillion in administrative overhead per year. A patient may spend seven minutes with a physician – but that visit could result in hundreds of minutes of administrative work per clinician, according to Pranay Kapadia, cofounder, and CEO of Notable. Using AI, Notable can eliminate more than 700 minutes of that administrative work, including creating clinical documentation and adding billing codes for the insurance claim processing.

The investment points to a larger industry trend toward using AI to improve patient care and streamline processes. Care sites like Intermountain Healthcare and CommonSpirit Health already use Notable, which automates everything from patient scheduling and check-in to post-visit follow-up, as well as creating clinical documentation and adding billing codes.

Demand for AI continues to increase as patients expect a digital-first experience due to the COVID-19 pandemic, as well as the “great resignation” that has left every industry — including health care — short-staffed. “Technology needs to drive ten times the efficiency at a quarter of the cost,” said Kapadia.

“Technology is the future of everything, and health care is no exception,” said Andrew J. Scott, founding partner of 7percent Ventures. “Artificial intelligence is already having a positive impact. Companies like Kheiron Medical can already perform mammography analysis for breast cancer better than a human.”

7percent Ventures invests in AI technology including Limbic, which uses AI for mental health triage and support, and Kherion Medical, which provides improved breast cancer diagnosis. These “are the sorts of transformative technologies that have a positive impact and improve the way we live,” he said.

Will AI Provide All Diagnoses?

Going all-in on AI in a health care setting may speed up a diagnosis – but it also takes away a physician’s autonomy in making the diagnosis and recommending treatment, according to Robert Wachter, MD, professor, and chair of the Department of Medicine at the University of California, San Francisco.

“There are a lot of sources of pushback, from the physician’s ego to worries about malpractice and who is liable, to ethical issues around AI,” such as whether the data is biased, he said. For example, the data may note that patients of one race don’t need as much medication as patients of another, without taking into account that particular patient’s situation.

AI will tackle more tractable problems like workflows before heading into the more difficult ones like diagnosis and prognosis, but there won’t be a real “AI moment,” Wachter said. “You start …where the stakes are less high, with business and operational problems.”

Instead, AI will augment what physicians are doing and provide options, including triage, but ultimately leave the decision up to the physician’s discretion.

“I see AI working silently behind the scenes of the busy clinician,” said Chris Larkin, chief technology officer at Concord Technologies. “The models will continue to gather data on patient diagnosis and trajectories and update the clinician when it’s appropriate. This is more like modern avionics, working on behalf of the pilot of the aircraft.”

For example, ICU nurses hear thousands of patient alarms on their shifts, many of which are false. AI can help the nurses decide which ones are most pressing based on the patient’s diagnosis and attend to them first, Larkin said.

Some clinicians already are using AI and machine learning exactly this way. “I’ve used VIDA Insights as an AI agent to assist me in interpreting chest CTs,” said John Newell, MD, professor of radiology and biomedical engineering, director of the Radiology Image Phenotyping Laboratory, and the co-director of the Iowa Institute for Biomedical Imaging.

Additionally, AI can help lower costs for both patients and health care organizations while providing better care. “If AI can help us to diagnose disease earlier and with more accuracy, the impact on reducing the cost of patient care can be significant,” Newell said.

“For example, a patient with early-stage COPD spends about $1,600 [per] year on care versus a patient with advanced-stage COPD who spends nearly $11,000 [per] year. COPD is often diagnosed later in the disease process, so any tools that can help providers identify it early can have a massive impact on population health care costs.”

Despite the opportunities AI provides for the health care industry, humans will always be needed — and AI doesn’t aim to entirely displace them. “With all the AI in the world, [there’s still] a certain level of empathy that comes in health care,” Kapadia said, noting that, like comforting a child with a sore throat, AI isn’t needed for that.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


Medical device leader Medtronic joins race to bring AI to health care

Medtronic, the world’s largest medical device company, is significantly increasing its investments into AI and other technologies, in what it says is an effort to help the health care industry catch up with other industries.

While many other industries have embraced technology, health care has been slower. Studies reveal that only 20% of consumers would trust AI-generated health care advice.

VentureBeat interviewed Torod Neptune, Medtronic’s senior vice president and chief communications officer, and Gio Di Napoli, president of Medtronic’s Gastrointestinal Unit, to discuss the company’s vision of the future of health care technology.

Digital transformation in health care

Neptune spoke about Medtronic’s transition beyond traditional med tech to more innovative solutions using AI. He noted that health care technology — through its unusual scale and ability to harness data analytics, algorithms, and intelligence — plays a significant role in solving big problems in the AI field.

Artificial intelligence increases the detection of early cancer by 14% compared to normal colonoscopy, Di Napoli said. This is very important because “every percentage of increase in detection reduces the risk of cancer by 2%,” he said.

Building on Medtronic’s medical devices already serving millions (like its miniature pacemaker, smart insulin pump, and more), the company’s plan to make health care more predictive and personal led to the development of GI Genius Intelligent Endoscopy Module (granted USFDA de novo clearance on April 9, 2021, and launched on April 12, 2021).

Medical equipment arranged in shelves on a cart, with a large monitor on top that shows an intestinal scan in progress.

Above: Medtronic says its GI Genius Intelligent Endoscopy Module is the first-to-market computer-aided polyp detection system powered by artificial intelligence.

The GI Genius module is the first and only artificial intelligence system for colonoscopy, according to Medtronic, assisting physicians in detecting precancerous growths and potentially addressing 19 million colonoscopies annually. The company says the module serves as a vigilant second observer, using sophisticated AI-powered technology to detect and highlight the presence of precancerous lesions with a visual marker in real time.

Investing in innovative health care

Medtronic has launched more than 190 health care technology products in the past 12 months. It also invests $2.5 billion yearly on research and development (R&D). Medtronic’s CEO, Geoff Martha, recently announced a 10% boost in R&D spending by FY22.

This enormous investment, the largest R&D increase in company history, underscores Medtronic’s focus on innovation and technology.

The company says it plans to expand the number of patients it serves each year, with the goal being 85 million by FY25.

According to Di Napoli, “AI is here. And it’s here to stay.”

A new era of health care

Speaking further about health care technology, Di Napoli says, “I can tell from my personal experience within the gastrointestinal business that there is a need for training and getting to know artificial intelligence as a partner and not as an enemy. And I think it’s critical for companies like ours to keep collecting data to improve our algorithms, to improve how our customers decide based on this data, and also improve patients outcomes with this.”

Although data collection comes with security concerns and privacy issues, Di Napoli says that the company is in constant communication with the FDA to understand the process to put in place to protect sensitive data for the future.

Neptune believes that technology and data drive patient empowerment in a much more significant way, based on more comfortable user adoption over the last 20 months. He said, “I think the pandemic has enabled more comfort and consideration, and there’s a global shift and willingness to engage and adopt new technological solutions.”


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


Using digital twins in health care to stave off the grim reaper

All the sessions from Transform 2021 are available on-demand now. Watch now.

VentureBeat caught up with NTT Research Medical & Health Informatics Lab director Dr. Joe Alexander, who elaborated on his view of the future of “bio digital twins,” which promise to improve precision medicine and bring digital transformation to the health care industry.

Japanese telecom giant NTT has launched a major initiative to improve digital health through precision medicine using digital twin technology. This project is part of NTT Research, a new R&D hub focused on basic research. The goal is to address long-term technological challenges with solutions that, once achieved, can positively impact wider ranges of businesses and many parts of our lives. These projects are not tied to specific product roll-out plans but could lead to much more significant long-term improvements than conventional incremental research conducted by enterprises.

The why behind the digital twin application

VentureBeat: What exactly is medical and health informatics — where does it fit into the landscape of other enterprise medical software like EHRs, diagnostics, telemedicine, and research?

Dr. Joe Alexander: Medical informatics is the sub-discipline of health informatics that directly impacts the patient-physician relationship. It focuses on the information technology that enables the effective collection of data using technology tools to develop medical knowledge and to facilitate the delivery of patient medical care. The goal of medical informatics is to ensure access to critical patient medical information at the precise time and place it is needed to make medical decisions. Medical informatics also focuses on the management of medical data for research and education.

The acquisition, storage, retrieval, and use of health care information to foster better collaboration among a patient’s various health care providers is the study of health informatics. It plays a critical role in the push toward health care reform. Health informatics is an evolving specialization that links information technology, communications, and health care to improve the quality and safety of patient care. EHRs help providers better manage care for patients and are an important part of health informatics.

Telemedicine has more to do with the access and sharing of medical information for the purpose of treating patients remotely. The term “diagnostics” can be applied to any process or device that involves techniques for (medical) diagnoses.

One current area of research that is of particular interest to our team is precision cardiology. This includes the cardiovascular bio digital twin as well as heart-on-a-chip technologies.

Research at MEI Labs does not currently target EHR software development or telemedicine per se. Our work does support remote monitoring, diagnostics, and advanced therapeutics.

VentureBeat: What is the bio digital twin initiative, and how do you plan to advance it?

Alexander: A bio digital twin is an up-to-date virtual representation (an electronic replica) which provides real-time insights into the status of a real-world asset to enable better management and to inform decision-making. This concept has been applied to the preventive maintenance of jet engines and may be applied as well to the predictive maintenance of health.

The Bio Digital Twin (BioDT) initiative aims to individualize and revolutionize health care by use of BioDT technologies. We will first realize precision cardiology on multiple scales through development of a cardiovascular BioDT (CV BioDT) and heart-on-a-chip platforms. The CV BioDT is at the whole organ physiological system level, whereas the heart-on-a-chip is at a microfluidics level, making use of an individual’s stem cells to make in vitro organs.

For the CV BioDT, we will begin with acute conditions (acute myocardial infarction and acute heart failure) and progress to chronic cardiovascular conditions and their co-morbidities and complications. The latter requires heavy dependence on organ systems other than the heart. Ultimately, based on our accumulating knowledge of underlying physiological and pathophysiological mechanisms (together with advanced sensing technologies), we will be able to move into wellness and prevention.

Can digital twins in health care save a life?

VentureBeat: What is the value of a digital twin, and how does it build upon other technologies for capturing and managing medical data or simulating things?

Alexander: We expect that our bio digital twin will best enable individualized care. By reproducing an individual’s entire physiology based on causal mechanisms, we should be able to predict health issues as well as provide recommendations for therapies in complex patients through “what if” scenario testing.

Autonomous therapies — delivered by the bio digital twin — become possible, where the physician would simply monitor autonomous devices. Virtual clinical trials in populations of bio digital twins also become feasible and would dramatically accelerate drug (or vaccine) development.

What we are proposing is not evolutionary, but revolutionary. An ambitious project of this scope and scale will take time. We will certainly need continuously to inventory the evolving trajectories of clinical and technology landscapes for facilitatory impact points.

VentureBeat: Why did you decide to start with the heart, and how will this complement other, similar efforts?

Alexander: We started with cardiovascular disease because it is the global leading cause of death. One of the principal missions of NTT Research is to provide long-term benefits to humanity; this is fundamental to deciding what projects to pursue.

Our immediate cardiovascular disease targets will be acute myocardial infarction (AMI) and acute heart failure (acute HF). We will pursue chronic heart failure and other conditions afterwards.

VentureBeat: What’s next in digital twins and why?

Alexander: Following development of the CV BioDT, our next pursuit will be neurodegenerative diseases, e.g., Alzheimer’s disease and Parkinson’s disease. Our reasoning here is similar: neurodegenerative diseases are the 2nd leading cause of death, at least in the U.S.

Organs on a chip

VentureBeat: What kinds of things are you working on with nano and microscale sensors and electrodes?

Alexander: MEI Lab is developing “organ-on-a-chip” microfluidics platforms as well as three-dimensionally transformable and implantable electrodes. This work involves the exploration and examination of new materials that include nanofibers and nanofiber-based paper electrodes.

VentureBeat: Which ones show the most promise in the short term and possibility in the long term?

Alexander: This is a difficult question for me to answer since I am not directly involved in the research. However, all our targets tend to be long term. Based on current progress, microscale three-dimensionally transformable electrodes for sensing are more promising in the shorter term, followed by similar types of electrodes for both stimulating and sensing. Organ-on-a-chip platforms will likely mature in the longer term.

VentureBeat: What are some of the key developments in digital biomarkers, wearable technologies, and remote sensing you are exploring?

Alexander: While we are in an ongoing background process of doing a clinical and technical landscape inventory of such devices, we have not yet developed a strategy within the MEI Lab to point us in any particular directions. Our focus right now is on acute conditions where patients are hospitalized and well-instrumented for access to the directly observable data necessary for early model building, verification, and validation.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


HeyRenee is the next home health care startup for Heal founders

All the sessions from Transform 2021 are available on-demand now. Watch now.

HeyRenee has raised $3.8 million for a personal health care concierge — the latest home health care business from the founders of Heal.

Husband and wife team Renee Dua and Nick Desai started Heal in 2014 after a bad experience in an emergency room with their son. They created a startup that let doctors make house calls to see patients in their homes or via remote telemedicine appointments. Mobile health care became a huge trend during the pandemic and grew to serve patients in a number of stages.


Over seven years, Heal raised more than $120 million. But Desai started thinking about what he wanted to do next and left in March. Dua stayed on longer to finish up her work as the company’s chief medical officer. Then she also left. For a time, they focused on taking care of their family, and then they started thinking about their next startup.

That’s where HeyRenee comes in. It’s a service that has similar vibes to Heal in that it focuses on patients and using digital technology to provide better health care. In this case, HeyRenee focuses on helping the elderly, the underserved, those with chronic health conditions, and others manage their patient care. The aim is to use the concierge service to tie together all of the patients’ medical needs — from prescriptions to doctor visits — through one digital helper.

Renee Dua founded HeyRenee after taking care of her father.

Above: Renee Dua cofounded HeyRenee after taking care of her father.

Image Credit: HeyRenee

Open platform

Los Angeles-based HeyRenee will be an open platform that will eventually work with every provider, partner, and point solution to curate the necessary combination of services for each patient’s specific needs.

Quiet Capital led the oversubscribed $3.8 million funding round, with Mucker Capital, Fika Ventures, Tau Ventures, Global Founders Capital, and SaaS Venture Capital also participating. HeyRenee is using seed proceeds to curate digital health partners, build a team of product and engineering leaders, and win early customers.

Dua, a practicing nephrologist, said in a statement that it’s “impossibly difficult for all of us, certainly older, sicker Americans, to follow the many instructions from their doctors.” She said those instructions are the recipe for leading happier, healthier lives, but people need help managing them.

She added that HeyRenee’s aim is to build something to finally slow the progression and exploding costs of easily treated chronic diseases, like obesity, diabetes, hypertension, and mental health issues, by easing the burden of managing health care.

About 85% of the people who used telehealth options during the COVID-19 pandemic in 2020 had a household income over $150,000. However, the true potential of the digital health revolution is to transform care for those with the fewest financial resources, the company said. HeyRenee aims to do that by demystifying and integrating previously disconnected point solutions and providers to work together in a data-driven symphony for a delightfully easy patient-centric experience, Desai said.

Nick Desai is cofounder of HeyRenee.

Above: Nick Desai is cofounder of HeyRenee.

Image Credit: HeyRenee

Health care helper

HeyRenee intends to ease the burden of health care coordination — from appointment scheduling, in-home services, and medication delivery to telehealth and the monitoring of symptoms and vital signs via one easy-to-use app. HeyRenee won’t provide an actual personal caretaker to care for a specific patient, but such a caretaker might use HeyRenee to manage a patient’s care, Desai said.

Tau Ventures managing partner Amit Garg said in a statement that his firm invests in AI-first companies and that having a moat around data is key. He touted the founders’ experience and said he believes HeyRenee will help improve the lives of patients.

And the founders reminded us that the business is personal. In the aftermath of a surgery on her father, Dua saw his memory was greatly affected. He had new cognitive difficulties because the hospital stay wiped him out. She became his caretaker and is bringing this knowledge to HeyRenee because she believes everyone needs a “Renee” as a best friend on their health care journey.

The platform is expected to launch in 2022.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


How AI can enable better health care outcomes

Join executive leaders at the Data, Analytics, & Intelligent Automation Summit, presented by Accenture. Register here.

Artificial intelligence isn’t just a tool for pure tech — health care providers can use it too. Clinical practice and AI go together, three top health care leaders at national enterprises agreed during a panel at Transform 2021 hosted by VentureBeat general manager Shuchi Rana.

Using data to reduce medical waste and over-testing can help hospital systems save money, said Dr. Doug Melton, head of clinical and customer analytics at Evernorth, a subsidiary of insurance giant Cigna. “Before, we had unsupervised learning, and it was harder to do. You had to be prescriptive in your hypotheses,” Melton said.

AI has the potential to help clinicians improve patient outcomes, said Dr. Taha Kass-Hout, director and chief medical officer at Amazon Web Services. Medical records can be a great source of data to develop algorithms, speech recognition, and decision-making tools that could help doctors and nurses identify risk factors for serious illnesses such as congestive heart failure.

Early breast and lung cancer detection is another outcome that not only helps patients, but also benefits enterprise leaders. At Evernorth, Melton’s team used machine learning to analyze pre-certifications for radiology and past claims data, identifying who was at higher risk of developing more serious health issues down the line. ML improves prevention and holistic management, Melton said, and improves cost savings for both the patient and provider by as much as 3 times.

Data analytics are also key to reducing other hospital costs, said Dr. Joe Colorafi, system VP of clinical data science and analytics at Commonspirit Health. By crunching the numbers, researchers can find which hospital stays last too long and when clinicians are over-assigned to a patient.

Collecting additional data from users can also help providers determine a holistic health care plan, Melton said. For instance, information on stressors in patients’ lives and other social determinants of health, such as access to fresh food and stable housing, can anchor plans to improve health outcomes. “When we do that, I think we can have whole-person medicine instead of acute care management,” Melton said.

Think of AI as a toolbox to understand the information presented to health care providers, Kass-Hout said. Using machine learning to narrow down symptoms and diagnoses also means building a repository of information to improve health systems. For instance, the accuracy of Amazon Web Services’ model to predict congestive heart failure increased by 4% as the algorithms took in notes about how physicians were treating the condition and monitoring patients for symptoms.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


21 ways medical digital twins will transform health care

Where does your enterprise stand on the AI adoption curve? Take our AI survey to find out.

The health care industry is starting to adopt digital twins to improve personalized medicine, health care organization performance, and new medicines and devices. Although simulations have been around for some time, today’s medical digital twins represent an important new take. These digital twins can create useful models based on information from wearable devices, omics, and patient records to connect the dots across processes that span patients, doctors, and health care organizations, as well as drug and device manufacturers.

It is still early days, but the field of digital twins is expanding quickly based on advances in real-time data feeds, machine learning, and AR/VR. As a result, digital twins could dramatically shift how we diagnose and treat patients, and help realign incentives for improving health. Some proponents liken the current state of digital twins to where the human genome project was 20 years ago, and it may require a similar large-scale effort to take shape fully. A team of Swedish researchers recently wrote, “Given the importance of the medical problem, the potential of digital twins merits concerted research efforts on a scale similar to those involved in the HGP.”

While such a “moon shot” effort may not be immediately underway, there are many indicators that digital twins are gaining traction in medicine. Presented here are 21 ways digital twins are starting to shape health care today, broken roughly into personalized medicine, improving health care organizations, and drug and medical devices and development. In fact, many types of digital twins span multiple use cases and even categories; it is these cross-domain use-cases that form a major strength of digital twins.

Personalized medicine

Digital twins show tremendous promise in making it easier to customize medical treatments to individuals based on their unique genetic makeup, anatomy, behavior, and other factors. As a result, researchers are starting to call on the medical community to collaborate on scaling digital twins from one-off projects to mass personalization platforms on par with today’s advanced customer data platforms.

1. Virtual organs

Several vendors have all been working on virtual hearts that can be customized to individual patients and updated to understand the progression of diseases over time or understand the response to new drugs, treatments, or surgical interventions. Philip HeartModel simulates a virtual heart, starting with the company’s ultrasound equipment.  Siemens Healthineers has been working on a digital twin of the heart to improve drug treatment and simulate cardiac catheter interventions. European startup FEops has already received regulatory approval and commercialized the FEops Heartguide platform. It combines a patient-specific replica of the heart with AI-enabled anatomical analysis to improve the study and treatment of structural heart diseases.

Dassault launched its Living Heart Project in 2014 to crowdsource a virtual twin of the human heart. The project has evolved as an open source collaboration among medical researchers, surgeons, medical device manufacturers, and drug companies. Meanwhile, the company’s Living Brain project is guiding epilepsy treatment and tracking the progression of neurodegenerative diseases. The company has organized similar efforts for lungs, knees, eyes, and other systems.

“This is a missing scientific foundation for digital health able to power technologies such as AI and VR and usher in a new era of innovation,” Dassault senior director of virtual human modeling Steve Levine told VentureBeat. He added that this “could have an even greater impact on society than what we have seen in telecommunications.”

2. Genomic medicine

Swedish researchers have been mapping mice RNA into a digital twin that can help predict the effect of different types and doses of arthritis drugs. The goal is to personalize human diagnosis and treatment using RNA. The researchers observed that medication does not work about 40% to 70% of the time. Similar techniques are also mapping the characteristics of human T-cells that play a crucial role in immune defense. These maps can help diagnose many common diseases earlier when they are more effective and cheaper to treat.

3. Personalized health information

The pandemic has helped fuel the growth of digital health services that help people assess and address simple medical conditions using AI. For example, Babylon Health‘s Healthcheck App captures health data into digital twins. It works with manually entered data such as health histories, a mood tracker, symptom tracker, and automatic capture from fitness devices and wearables like the Apple Watch. The digital twin can provide basic front-line information or help guide priorities and interactions with doctors to address more severe or persistent conditions.

4. Customize drug treatment

The Empa research center in Switzerland is working on digital twins to optimize drug dosage for people afflicted by chronic pain. Characteristics such as age and lifestyle help customize the digital twin to help predict the effects of pain medications. In addition, patient reports about the effectiveness of different dosages calibrate digital twin accuracy.

5. Scanning the whole body

Most approaches to digital twins build on existing equipment to capture the appropriate data, while Q Bio’s new Gemini Digital Twin platform starts with a whole-body scan. The company claims to capture a whole-body scan in 15 minutes without radiation or breath holds, using advanced computational physics models that are more precise than conventional MRI for many diagnoses. The company has received over $80 million from Andreessen Horowitz, Kaiser Foundation Hospitals, and others. Q Bio is also developing integrations to improve these models using data from genetics, chemistry, anatomy, lifestyle, and medical history.

6. Planning surgery

A Boston hospital has been working with Dassault’s digital heart to improve surgical procedure planning and assess the outcomes afterward. The digital twins also help them to generate the shape of a cuff between the heart and arteries.

Sim&Cure’s Sim&Size is a digital twin to help brain surgeons treat aneurysms using simulations to improve patient safety. Aneurysms are enlarged blood vessels that can result in clots or strokes. These digital twins can improve the ability to plan and execute less invasive surgery using catheters to install unique implants. Data from individual patients helps customize simulations that run on an embedded simulation package from Ansys.  Preliminary results have dramatically reduced the need for follow-up surgery.

Improving health care organizations

Digital twins also show promise in improving the way health care organizations deliver care. Gartner coined the term digital twin of the organizations to describe this process of modeling how an organization operates to improve underlying processes.

In most industries, this can start by using process mining to discover variations in business processes. New health care-specific tools can complement these techniques.

7. Improving caregiver experience

Digital twins can also help caregivers capture and find information shared across physicians and multiple specialists. John Snow Labs CTO David Talby said, “We’re generating more data than ever before, and no one has time to sort through it all.” For example, if a person sees their regular primary care physician, they will have a baseline understanding of the patient, their medical history, and medications. If the same patient sees a specialist, they may be asked many of the same repetitive questions.

A digital twin can model the patient and then use technologies like NLP to understand all of the data and cut through the noise to summarize what’s going on. This saves time and improves the accuracy of capturing and presenting information like specific medications, health conditions, and more details that providers need to know in context to make clinical decisions.

8. Driving efficiency

The GE Healthcare Command Center is a major initiative to virtualize hospitals and test the impact of various decisions on changes in overall organizational performance. Involved are modules for evaluating changes in operational strategy, capacities, staffing, and care delivery models to objectively determine which actions to take. For example, they have developed modules to estimate the impact of bed configurations on care levels, optimize surgical schedules, improve facility design, and optimize staff levels. This allows managers to test various ideas without having to run a pilot. Dozens of organizations are already using this platform, GE said.

9. Shrinking critical treatment window

Siemens Healthineers has been working with the Medical University of South Carolina to improve the hospital’s daily routine through workflow analysis, system redesign, and process improvement methodologies. For example, they are working to reduce the time to treat stroke patients. This is important since early treatment is critical but requires the coordination of several processes to perform smoothly.

10. Value-based health care

The rising cost of health care has many nations exploring new incentive models to better align new drugs, interventions, and treatments with outcomes. Value-based health care is one approach that is growing in popularity. The basic idea is that participants, like drug companies, will only get compensation proportionate to their impact on the outcomes. This will require the development of new types of relationships across multiple players in the health delivery systems. Digital twins could provide the enabling infrastructure for organizing the details for crafting these new types of arrangements.

11. Supply chain resilience

The pandemic illustrated how brittle modern supply chains could be. Health care organizations immediately faced shortages of essential personal protection equipment owing to shutdowns and restrictions from countries like China. Digital twins of a supply chain can help health care organizations model their supply chain relationships to understand better how to plan around new events, shutdowns, or shortages. This can boost planning and negotiations with government officials in a pinch, as was the case in the recent pandemic. A recent Accenture survey found that 87% of health care executives say digital twins are becoming essential to their organization’s ability to collaborate in strategic ecosystem partnerships.

12. Faster hospital construction

Digital twins could also help streamline construction of medical facilities required to keep up with rapid changes, such as were seen in the pandemic. Atlas Construction developed a digital twin platform to help organize all the details for health care construction. The project was inspired long before the pandemic when Atlas founder Paul Teschner saw how hard it was to get new facilities built in remote areas of the world. The platform helps organize design, procurement, and construction processes. It is built on top of the Oracle Cloud platform and Primavera Unifier asset lifecycle management service.

13. Streamlining call center interactions

Digital twins can make it easier for customer service agents to understand and communicate with patients. For example, a large insurance provider used a TigerGraph graph database to integrate data from over 200 sources to create a full longitudinal health history of every member. “This level of detail paints a clear picture of the members current and historical medical situation,” said TigerGraph health care industry practice lead Andrew Anderson.

A holistic view of all diagnosis claims prescriptions, refills, follow-up visits, and outstanding claims reduced call handling time by 10%, TigerGraph claimed, resulting in over $100 million in estimated savings. Also, shorter but more relevant conversations between the agents and members have increased Net Promoter Score and lowered churn.

Drug and medical device development

There are many ways that digital twins can improve the design, development, testing, and monitoring of new medical devices and drugs. The U.S. FDA has launched a significant program to drive the adoption of various types of digital approaches. Regulators in the U.S. and Europe are also identifying frameworks for including modeling and simulation as sources of evidence in new drug and device approvals.

14. Software-as-a-medical device

The FDA is creating the regulatory framework to allow companies to certify and sell software-as-a-medical device. The core idea is to generate a patient-specific digital twin from different data sources, including lab tests, ultrasound, imaging devices, and genetic tests. In addition, digital twins can also help optimize the software in medical devices such as pacemakers, automated insulin pumps, and novel brain treatments.

15. Classifying drug risks

Pharmaceutical researchers are using digital twins to explore the heart risks of various drugs. This could help improve drug safety of individual drugs and drug combinations more cost-effectively than through manual testing. They have built a basic model for 23 drugs. Extending this model could help reduce the estimated $2.5 billion required to design, test, get approved, and launch new drugs.

16. Simulating new production lines

Siemens worked with several vaccine manufacturers to design and test various vaccine production line configurations. New mRNA vaccines are fragile and must be precisely combined using microfluidic production lines that precisely combine nanoscale-sized particles. Digital twins allowed them to design and validate the manufacturing devices, scale these processes, and accelerate its launch from 1 year down to 5 months.

17. Improve device uptime

Philips has launched a predictive maintenance program that collates data from over 15,000 medical imaging devices. The company is hoping that digital twins could improve uptime and help their engineers customize new equipment for the needs of different customers. In addition, it is hoping to apply similar principles across all of its medical equipment.

18. Post-market surveillance

Regulators are beginning to increase the emphasis for device makers to monitor the results of their equipment after-sales as part of a process called post-market surveillance. This requires either staffing expensive specialists to maintain the equipment or embedding digital twins capabilities into the equipment. For example, Sysmex worked with PTC to incorporate performance testing into its blood analyzer to receive a waiver from these new requirements, PTC CTO Steve Dertien told VentureBeat. This opened the market for smaller clinical settings closer to patients, which can speed diagnosis.

19. Simulating human variability

Skeletons and atlases commonly depict the perfect human. However, real-life humans typically have some minor variations in their muscles or bones that mostly go unnoticed. As a result, medical device makers struggle with how common anatomical variations among people may affect the fit and performance of their equipment. Virtonomy has developed a library of common variations to help medical equipment makers test conduct studies on how these variations may affect the performance and safety of new devices. In this case, they simulate the characteristics representing common variations in a given population rather than individuals.

20. Digital twin of a lab

Modern drug development often requires testing out thousands or millions of possibilities in a highly controlled environment. A digital twin of the lab can help to automate these facilities. It can also help to prioritize tests in response to discoveries. Digital twins could also improve the reproducibility of experiments across labs and personnel in the same lab. In this quest, Artificial recently closed $21.5 million in series A funding from Microsoft and others to develop lab automation software. The company is betting that unified data models and platforms could help them jump to the front of the $10 billion lab automation market.

21. Improving drug delivery

Researchers at Oklahoma State have been working with Ansys to develop a digital twin to improve drug delivery using models of simulated lungs as part of the Virtual Human System project. They found that only about 20% of many drugs reached their target. The digital twins allowed them to redesign the drug’s particle size and composition characteristics to improve delivery efficiency to 90%.


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


AI Weekly: WHO outlines steps for creating inclusive AI health care systems

Where does your enterprise stand on the AI adoption curve? Take our AI survey to find out.

This week, the World Health Organization (WHO) released its first global report on AI in health, along with six guiding principles for design, development, and deployment. The fruit of two years of consultations with WHO-appointed experts, the work cautions against overestimating the benefits of AI while highlighting how it could be used to improve screening for diseases, assist with clinical care, and more.

The health care industry produces an enormous amount of data. An IDC study estimates the volume of health data created annually, which hit over 2,000 exabytes in 2020, will continue to grow at a 48% rate year over year. The trend has enabled significant advances in AI and machine learning, which rely on large datasets to make predictions ranging from hospital bed capacity to the presence of malignant tumors in MRIs. But unlike other domains to which AI has been applied, the sensitivity and scale of health care data makes collecting and leveraging it a formidable challenge.

The WHO report acknowledges this, pointing out that the opportunities brought about by AI are linked with risks. There’s the harms that biases encoded in algorithms could cause patients, communities, and care providers. Systems trained primarily on data from people in high-income countries, for example, may not perform well for low- and middle-income patients. What’s more, unregulated use of AI might undermine the rights of patients in favor of the commercial interests or governments engaged in surveillance.

The datasets used to train AI systems that can predict the onset of conditions like Alzheimer’s, diabetes, diabetic retinopathy, breast cancer, and schizophrenia come from a range of sources. But in many cases, patients aren’t fully aware their information is included. In 2017, U.K. regulators concluded that The Royal Free London NHS Foundation Trust, a division of the U.K.’s National Health Service based in London, provided Google’s DeepMind with data on 1.6 million patients without their consent.

Regardless of the source, this data can contain bias, perpetuating inequalities in AI algorithms trained for diagnosing diseases. A team of U.K. scientists found that almost all eye disease datasets come from patients in North America, Europe, and China, meaning eye disease-diagnosing algorithms are less certain to work well for racial groups from underrepresented countries. In another study, researchers from the University of Toronto, the Vector Institute, and MIT showed that widely used chest X-ray datasets contain racial, gender, and socioeconomic biases.

Further illustrating the point, Stanford researchers found that some AI-powered medical devices approved by the U.S. Food and Drug Administration (FDA) are vulnerable to data shifts and bias against underrepresented patients. Even as AI becomes embedded in more medical devices — the FDA approved over 65 AI devices last year — the accuracy of these algorithms isn’t necessarily being rigorously studied, because they’re not being evaluated by prospective studies.

Experts argue that prospective studies, which collect test data prior to rather than concurrent with deployment, are necessary, particularly for AI medical devices because their actual use can differ from the intended use. For example, most computer-powered diagnostic systems are designed to be decision-support tools rather than primary diagnostic tools. A prospective study might reveal that clinicians are misusing a device for diagnosis, leading to outcomes that might deviate from what’s expected.

Beyond dataset challenges, models lacking peer review can encounter roadblocks when deployed in the real world. Scientists at Harvard found that algorithms trained to recognize and classify CT scans could become biased toward scan formats from certain CT machine manufacturers. Meanwhile, a Google-published whitepaper revealed challenges in implementing an eye disease-predicting system in Thailand hospitals, including issues with scan accuracy.

To limit the risks and maximize the benefits of AI for health, the WHO recommends taking steps to protect autonomy, ensure transparency and explainability, foster responsibility and accountability, and work toward inclusiveness and equity. The recommendations also include promoting well-being, safety, and the public interest, as well as AI that’s responsive and sustainable.

The WHO says redress should be available to people adversely affected by decisions based on algorithms, and also that designers should “continuously” assess AI apps to determine whether they’re aligning with expectations and requirements. In addition, the WHO recommends both governments and companies address disruptions in the workplace caused by automated systems, including training for health care workers to adapt to the use of AI.

“AI systems should … be carefully designed to reflect the diversity of socioeconomic and health care settings,” the WHO said in a press release. “They should be accompanied by training in digital skills, community engagement, and awareness-raising, especially for millions of healthcare workers who will require digital literacy or retraining if their roles and functions are automated, and who must contend with machines that could challenge the decision making and autonomy of providers and patients.”

As new examples of problematic AI in health care emerge, from widely deployed but untested algorithms to biased dermatological datasets, it’s becoming critical that stakeholders follow accountability steps like those outlined by the WHO. Not only will it foster trust in AI systems, but it could improve care for the millions of people who might be subjected to AI-powered diagnostic systems in the future.

“Machine learning really is a powerful tool, if designed correctly — if problems are correctly formalized and methods are identified to really provide new insights for understanding these diseases,” Dr. Mihaela van der Schaar, a Turing Fellow and professor of machine learning, AI, and health at the University of Cambridge and UCLA, said during a keynote at the ICLR conference in May 2020. “Of course, we are at the beginning of this revolution, and there is a long way to go. But it’s an exciting time. And it’s an important time to focus on such technologies.”

For AI coverage, send news tips to Kyle Wiggers — and be sure to subscribe to the AI Weekly newsletter and bookmark our AI channel, The Machine.

Thanks for reading,

Kyle Wiggers

AI Staff Writer


VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member

Repost: Original Source and Author Link


Responsible AI in health care starts at the top — but it’s everyone’s responsibility (VB Live)

Presented by Optum

Health care’s Quadruple Aim is to improve health outcomes, enhance the experiences of patients and providers, and reduce costs — and AI can help. In this VB Live event, learn more about how stakeholders can use AI responsibly, ethically, and equitably to ensure all populations benefit.

Register here for free.

Breakthroughs in the application of machine learning and other forms of artificial intelligence (AI) in health care are rapidly advancing, creating advantages in the field’s clinical and administrative realms.  It’s on the administrative side — think workflows or back office processes — where the technology has been more fully adopted. Using AI to simplify those processes creates efficiencies that reduce the amount of work it takes to deliver health care and improves the experiences of both patients and providers.

But it’s increasingly clear that applying AI responsibly needs to be a central focus for organizations who use data and information to improve outcomes and the overall experience.

“Advanced analytics and AI have a significant impact in how important decisions are made across the health care ecosystem,” says Sanji Fernando, SVP of artificial intelligence and analytics platforms at Optum. And, so, the company has guidelines for the responsible use of advanced analytics and AI for all of UnitedHealth Group.

“It’s important for us to have a framework, not only for the data scientists and machine learning engineers, but for everyone  in our organization — operations, clinicians, product managers, marketing — to better understand  expectations  and how we want to drive breakthroughs to better support our customers, patients, and the wider health care system,” he says. “We view the promise of AI and its responsible use as  part of our shared responsibility to use these breakthroughs appropriately for patients, providers, and our customers.”

The guideline focuses on making sure everyone is considering how to appropriately use advanced analytics and AI, how these models are trained, and how they are monitored and evaluated over time, he adds.

Machine learning models, by definition, learn from the available data that’s being created throughout the health care system. Inequities in the system may be reflected in the data and predictions that machine learning models return. It’s important for everyone to be aware that health inequity may exist and that models may reflect that, he explains.

“By consistently evaluating  how models may classify or infer, and looking at how that affects folks of different races, ethnicities, and ages, we can  be more aware of where some models may require consistent examination to best ensure they are working the way we’d like them to,” he says. “The reality is that there’s no magic bullet to ‘fix’ an ML model automatically, but it’s important for us to understand and consistently learn where these models may impact different groups.”

Transparency is a key factor in delivering responsible AI. That includes being very clear about how you’re training your models, the appropriate use of data used to train an algorithm, as well as data privacy. When possible, it also means understanding how specific features are being identified or leveraged within the model. Basics like an age or date are straightforward features, but the challenge arises with paragraphs of natural language and unstructured text. Each word, phrase or paragraph can be considered a feature, creating an enormous number of combinations to consider.

“But understanding feature importance — the features that are more important to the model — is important to provide better insight into how the model may actually be  working,” he explains. “It’s not true mathematical interpretability, but it gives us a better awareness.”

Another important factor is being able to reproduce the performance and results of a model. Results will necessarily change when you train or retrain an algorithm, so you want to be able to trace that history, by being able to reproduce results over time. This ensures the consistency and appropriateness of the model remains constant (and allows for potential adjustments should they be needed).

There’s no shortage of tools and capabilities available across the field of responsible AI because there are so many people who are passionate about making sure we all use AI responsibly. For example, Optum uses an open-source bias audit tool from the University of Chicago. But there are any number of approaches and great thinking from a tooling perspective, Fernando says, so it’s really becoming an industry best practice to implement a policy of responsible AI.

The other piece of the puzzle requires work and a commitment from everyone in the ecosystem: making responsible use everyone’s responsibility, not just the machine learning engineer or data scientist.

“Our aspiration is that every employee understands these responsibilities and takes ownership of them,” he says, “whether UHG employees are using ML-driven recommendations in their day-to-day work, designing new products and services, or they’re the data scientists and ML engineers who can evaluate models and understand output class distributions, we all have a shared responsibility to ensure these tools are achieving the best and most equitable results for the people we serve.”

To learn more about the ways that AI is impacting the delivery and administration of health care across the ecosystem, the benefits of machine learning for cost savings and efficiency, and the importance of responsible AI for every worker, don’t miss this VB Live event.

Don’t miss out!

Register here for free.

You’ll learn:

  • What it means to use advanced analytics “responsibly”
  • Why responsible use is so important in health care as compared to other fields
  • The steps that researchers and organizations are taking today to ensure AI is used responsibly
  • What the AI-enabled health system of the future looks like and its advantages for consumers, organizations, and clinicians


  • Brian Christian, Author, The Alignment Problem, Algorithms to Live By and The Most Human Human
  • Sanji Fernando, SVP of Artificial Intelligence & Analytics Platforms, Optum
  • Kyle Wiggers, AI Staff Writer, VentureBeat (moderator)

Repost: Original Source and Author Link