Learn about the many roles AI is performing in healthcare to enable safer, more reliable and precision-based outcomes.
There can be few better uses of technology than helping improve people’s health. As the coronavirus pandemic has shown, tools like cloud-based AI and data visualization play a vital role in everything from modeling responses to sequencing the virus and the genomes of those affected.
Our collective goal is to help ensure the right tools are in the right hands to inspire medical breakthroughs. As a result, during the current crisis:
- NVIDIA is offering a free 90-day license to Parabricks (for accelerating the analysis of genome sequencing data) to any researcher studying the
- NVIDIA is debuting a series of technical talks entitled the Compute4Covid Webinar Series, exploring the available tools and how they can be used to combat COVID-19.
- Google Cloud launched a number of projects, including an AI-based chatbot to help government agencies manage the surge in communications. Researchers who require additional Google Cloud capacity for COVID-19 projects can submit proposals here.
Beyond these headlines,there are many examples of how technology – and increasingly AI – is being put to good use across the healthcare industry. In this article, you’ll find a number of scenarios that you can leverage to inspire your own innovation.
One of the more high-profile applications of AI in healthcare involves accurately detecting diseases in their early stages. This is important work; in 2015 it was estimated that misdiagnosing illness and medical error accounted for 10% of US deaths.
Incomplete medical records and large caseloads can lead to human error. With AI, doctors have access to every stored medical journal, symptom, and case study treatment for diagnostic reference. This repository of medical information is making a difference for institutions worldwide.
AI can also predict and diagnose disease at a faster rate than most medical professionals. Google’s DeepMind Health AI software is being used to notify doctors when a patient’s health deteriorates and helps with diagnosis by combining its massive set of comparable symptoms.
Then there are the more specific uses of the technology, with standout examples including:
- Early breast cancer detection: AI can review and translate mammograms 30X faster than a human with 99% accuracy and reduce the need for unnecessary biopsies.
- Detection of blood diseases: AI-enhanced microscopes can scan for harmful bacteria (e.g. E.coli) in blood samples at a much faster rate than manual methods, and learn how to identify and predict harmful bacteria in blood with 95% accuracy.
- Reduced death from sepsis: The BARDA Drive Solving Sepsis initiative is using Google Cloud to develop learning software for the early prediction of sepsis in hospital intensive care units.
- Enhanced predictive diagnoses: NVIDIA GPUs and deep learning AI have enabled researchers at NYU to analyze lab tests, X-rays, and doctors’ notes to predict ailments, such as heart failure, three months quicker than manual analysis.
Improving healthcare training
AI allows medical professionals to go through realistic simulations in a way that “simple” computer-driven training cannot. With the ability to incorporate natural language and draw instantly on a large database of scenarios, AI training routines can respond to questions, decisions, or advice from a trainee in a way that might be impossible for humans.
In addition, these training programs can quickly learn from previous responses, enabling courses to be continually refined and adjusted over time to meet evolving learning needs.
However, when AI is combined with VR and simulation technologies, the technology gets even more interesting. This can be demonstrated via medical training that helps neurosurgeons develop the skills they need before they step inside an operating room. Here are some of the results:
- AI can record and collect all instrument movements in millisecond intervals.
- Using this data, an AI mode measures key factors like instrument position and force applied, as well as outcomes such as the amount of tumor removed and blood loss.
- Virtually-enabled surgery: AI-enabled robots combined with VR allow surgeons to “shrink” and explore the inside of a patient’s body in detail.
- Spinal surgery: 3D tools can be used to visualize surgical plans, read images with AI to recognize anatomical features, and perform more precise spinal operations.
- Hospital cleaning: AI-enabled UV light robots enter different hospital rooms, assess the situation, and leave only after they’re germ-free.
- Gamified recovery: For cognitive rehabilitation, AI tracking can guide a patient’s rate of progress based on various criteria.
Realize the Power of Two
AI technology is a major driving force behind the innovation agendas of many healthcare providers. From safer procedures to more accurate diagnoses, AI is able to inspire fundamental change and best practices – particularly when you have the Power of Two. That’s the power of NVIDIA® GPUs and the Google Cloud AI platform working in partnership to inspire breakthrough performance.