Background Image

Health Tech New Year’s Resolutions – Gerald J. Wilmink from CarePredict shares his thoughts for the new year

2018 was a pivotal year for digital health technologies with major headlines of industry disruption dominating the news agenda.     

 As we enter 2019, we’ve asked the leaders and innovators of the industry to look forward, and share their new year’s resolutions for 2019. They reveal the technologies they are most excited about, the biggest challenges currently facing health tech, and those they see having the biggest impact on the industry.

Intro

Today we hear from Gerald “Jerry” Wilmink, Chief Business Officer at CarePredict. CarePredict was the first-to-market to use machine learning, smart wearables, and unique kinematics to quantify daily activities performed by older adults and predict health conditions including increased fall risks, depression, and urinary tract infections (UTIs). Jerry is an American health entrepreneur, biomedical engineer, inventor, and keynote speaker. He has a broad and deep range of technical, operational and product commercialisation expertise which he gained from founding several health technology startups and from growing the first Terahertz biosensing laboratory in the Department of Defense (DoD) advanced research sector.  

Here’s what Jerry Wilmink had to say…

Artificial Intelligence (AI) is in full swing – powering and impacting our everyday lives

“Whether you are using Google maps to find the shortest commute to work, the use of AI has disrupted nearly every industry, and it is now making a big impact and taking over healthcare. Many of the major issues within our health care system are now being addressed using a family of AI techniques known as deep learning models or deep neural nets.” 

 “These methods allow us to feed raw data into a computer and it automatically discovers the proper representation needed for classification, detection and even diagnosis.”

In 2019, AI comes of age and begins to be used to solve narrow business and healthcare challenges

“In 2018 we saw a significant rise in AI technologies, tools, platforms and applications. In 2019, AI comes of age and begins to be used to solve narrow business and healthcare challenges. The major things that I predict in 2019 include:

Neural network interoperability

Many different frameworks are used to develop neural network models. However, once a model is trained on one framework it is difficult to move it to another framework. The adoption of AI will accelerate when models trained on one framework can be reused on another framework. Many of the technology behemoths: Microsoft, Facebook, Amazon, have joined forces to solve this problem by creating the Open Neural Network Exchange (ONNX). ONNX will be a key to 2019. 

AI-Enabled Chips

Deep learning requires powerful computer processors to quickly and efficiently perform massive calculations. For example, over 100 million parameters are calculated for a commonly used convolutional neural network of 16 hidden layers. Until the advent of the graphical processing unit, central processing units were used to train neural networks.  Central processing units (CPUs) calculate the parameters one at a time, whereas graphics processing units (GPUs) can calculate all of the operations in parallel.  

When GPUs hit the computer scene in the late 2000s, they were used for training neural networks. GPUs can train neural networks considerably quicker than the fastest CPUs. For instance, a CPU may take 150 hours to train a convolutional neural network, whereas a GPU only requires two hours. Deep neural nets are fundamentally different than other machine learning techniques in that their performance continues to increase with both increases in data and model size. Major chip manufacturers such as Intel, ARM, and NVIDIA are developing AI-enabled chips which will allow for faster query processing and predictive analytics which are required in many healthcare applications.”

Software has a voracious appetite

“Software may be eating the world, but software has a voracious appetite and requires a steady diet of rich, valuable, accurate, meaningful data. Therefore, insightful machine learning algorithms require access to differentiated data.”  

“For instance, data on a senior’s daily step counts may not be the most important data for predictions for when a senior is at elevated risk for cardiac arrest. Health technologies that use proprietary connected hardware can collect unique, differentiated, valuable data which can be fed to machine learning algorithms to address real health major problems.” 

“Data has become a commodity; however, valuable, rich, useful data is in high demand, but few health technologies are collecting such data.  Specifically, I am excited to see growth in predictive, preventative health technologies that address senior falls. Major corporations have developed reactive technologies to detect the presence of a fall, but such approaches are solving the wrong problem. Globally, nearly 650,000 fatal falls occur annually. This makes falls the second leading cause of unintentional injury death.”  

“Falls are a significant public health problem.  Falls are also the single most significant inflection point in aging.   People over the age of 70 have a 25% of death within a year of falling and breaking a major bone and majority die within 5 years of the incident. Not breaking a bone can mean living a normal lifespan. Previous technologies focused on reacting to a fall, CarePredict is leveraging its unique proprietary data to predict and potentially prevent senior falls.”

We’re in the midst of the Fourth Industrial Revolution

“In recent years, many technologies have been developed and implemented to achieve the elusive goal of bringing the “Quadruple Aim” from concept to reality. The Quadruple aims include improving clinician satisfaction, enhancing the patient care experience, improving outcomes, and reducing the cost of care.” 

“We presently are in the midst of the Fourth Industrial Revolution, a technological revolution that is based on information and communication technologies, that fuse the boundaries between the physical, digital and biological spaces. Health technologies developed in recent years and in the future during this period have already and will continue to modernise the healthcare industry and deliver on the “Quadruple Aim.” This technological revolution is marked by the several key elements: ubiquitous nature of wireless connectivity, internet of things IoT (eg. wearables, smart devices) equipped with miniaturised sensors that serve as the gateway to the collection of big data, cloud computing, and artificial intelligence (AI).” 

“Health technologies developed during this current revolution are now providing scenarios where medical information can be gathered at the point of care, analysed in the cloud or at the edge using algorithms, which can then provide real-time, predictive, person-centered actionable insights. Data-driven predictions underpin personalized medicine, and such insights are key to providing better, more cost-effective care. Putting sophisticated health technology tools at the fingertips of physicians, nurses and caregivers empowers them to deliver on the “The Quadruple Aim.”

Deep neural nets disrupting the healthcare sector are dominating the news agenda

“Some of my favourite news stories from the past twelve months include these studies which illustrate that such models can identify cancer up to 50% faster and with performance on par with leading radiologists and dermatologists.”  

“These algorithms can potentially save our health care system billions of dollars annually, by providing a preliminary diagnosis before a patient sees a specialist or visits an emergency room. Another recent study showed that a deep learning system that analyzed OCT images could detect over 50 eye diseases as accurately as a doctor. This software was issued the first FDA permit for an AI diagnostic system.”  

“In addition to diagnosing diseases, other deep learning platforms are being commercialised to provide preventative, person-centered senior care.  For example, CarePredict has commercialised a system that helps predict when seniors are at a higher fall risk or showing changes in activity and behavior foreshadowing  a urinary tract infection (UTI) or Depression.”