KELLI MURRAY, MedSpeaks
Life in 2017 is a real double-edged sword. On the one hand we have modern privileges like an open and free Internet, the ability to see and talk to our loved ones no matter where they are on the planet, and self-driving cars so close we can taste it. But on the other, we have the dark reality that we exist in a time when terror is being utilized on a global scale, most often surprising us when we least expect it.
Terror threats are a reality we must face and the weapons in use to cause harm involve bombs, semi-automatic guns, knives, and vehicles. Schools, churches, concert stadiums, parades, employers, and nightclubs are now targets; places where you'd least expect mass murders to occur. For us locals, the Pulse nightclub shooting in Orlando left the innocence of the "Happiest Place on Earth" marred by something sinister and unpredictable.
Last week, our Health Innovators' event focused on how organizations are preparing emergency response personnel, from hospitals to the battlefield. We talked with Eric Alberts, Manager of Emergency Preparedness at Orlando Health about the field he works in and protecting those he serves. He's responded to a variety of emergencies including a terrorist attack, 4 hurricanes, tropical systems, communicable diseases, wildfires, multi-type vehicle accidents, and mass casualty incidents.
Eric iterated that Orlando's risk for mass incidents is high and that teams have to be ready with a depth of knowledge to react when situations arise. To prepare staff, mock tragedies are performed 3 times each year. One of these exercises recruits nearly 800 volunteers to stand in as injured victims and distraught family members to challenge the clinical and ancillary teams in as realistic setting as possible. Clinicians practice dynamic procedures on simulators while patients and families scream and yell in various dialects and languages. The goal of scenario exposure, such as an active shooter exercise, is to move staff out of their instinctual flight mode and into a fight mode of response.
So what are some of the biggest hurdles when it comes to this new mindset and requirement for emergency response training? According to Eric, "you just cannot make things happen fast enough," referring to the speed of learnings, converting knowledge into competencies, and also decision-making within Orlando Health and across involved agencies. David Rogers, CEO of Allogy, also added that within organizations, "people are territorial and protective, and that really limits accessibility. Making information accessible to teams helps speed up the process to activate expertise and inform a lot of people at once. It also helps debrief on new standards of care but that's hard to do if bureaucracy is not compressed."
Before the Pulse tragedy, we didn't have the idea to connect a network of police, fire department, hospitals, the FBI, and others... Eric Alberts
To overcome these challenges, Eric and David surmise that what communities and organizations like ours need are:
2-way mass communication tools for notifications and acknowledgements
engagement to move new standards, processes and procedures into practice
bring training to where people are through mobile technology; shifting away from classrooms and traditional learning management systems (LMS)
urgency and prioritization for decision-making
funding to support emergency preparedness infrastructures
Ultimately, the greatest challenge and opportunity in medical training is finding ways to reduce the friction and efficiently bridge information and knowledge with demonstrable competencies. At the top of Eric's wish list is having a mass notification system, but the solutions he has seen thus far has been price prohibitive. Perhaps, Orlando Health, a nationally recognized star for emergency trauma management and Allogy, the mobile learning platform behind the Defense Health Agency's, Deployed Medicine app which is used to improve readiness and performance of military medical personnel, can find a way to lead a new path to emergency training.
As a community that sits at high risk for mass casualty incidents from both natural and manmade threats, we as citizens and healthcare leaders have a vested interest and role in helping push responsible solutions forward. As Eric flatly stated, "The threats are real. It's not a matter of if it happens again, it's a matter of when. We need to be ready."
For more information about volunteering for Orlando Health's mock trainings, go to www.orlandohealth.com and to learn more about Allogy's mobile learning platform, go to www.allogy.com.
Revolutionizing Cancer Care Part 1
BETH RUDLOFF, MedSpeaks
Hello Innovators! I am excited to start a series on some of the wonderful innovations that are helping in the world of cancer care. This particular innovator recently spoke at our Miami event on November 30, 2017 along with two other Florida-based cancer technologies, DermaSensor and DosiSoft.
Anabetsy Rivero is the CEO of Metastatic AI, a new company that is changing the way that biopsy results are given to patients. Anabetsy has a really interesting background - she was invited to conduct cancer and stem cell research at Dr. Robert Sackstein's lab at Harvard Medical School. The Sackstein lab further fueled her passion to cure cancer and improve the human condition. During her years doing research in cancer, genomics, and protein-protein interactions, she realized that learning programming was essential to bioinformatic analysis. She then taught herself computer programming and got accepted to the masters in computer science at Nova Southeastern University, where she specialized in data science and artificial intelligence. During her time at NSU, she met Dr. Saeed Rajput and, soon after, they started building a deep learning library that would eventually become Metastatic AI.
Let's start with how biopsies usually work, using the example of breast biopsies since that is what Anabetsy is working with currently. Breast biopsies are usually done to determine if a mass that is palpable and/or seen on a mammogram is cancerous. A needle is inserted into the mass and cells are drawn out by the physician and sent to the pathology department for review. The pathologist looks at the size and shape of the cells and the cell's nuclei in the biopsy as a factor in determining whether a patient has cancer. This is painstaking work that can take 24 hours, or up to a week depending on the testing the pathologist needs to do, because of course it is very important that the diagnosis is accurate.
At the University of California in Irvine, they have a database of 569 breast biopsies with these measurements and the final diagnosis. And that is where Anabetsy started to think about her hypothesis.
Could artificial intelligence and machine learning fast track the diagnosis from these pathology measurements? And how accurate would this diagnosis be?
As I have learned, artificial intelligence is when a computer does not just store and retrieve information, but it also mimics human learning patterns. Machine learning, a component of AI (artificial intelligence), occurs when "algorithms enable systems to take in new information, and apply those 'learnings' to make changes to its analytical engine. The more information fed into the system, the better it is at guiding decisions", according to Frost and Sullivans' 2017 white paper called The Artificial Intelligence Revolution has Arrived in Healthcare.
When Anabetsy fed a randomized sample of the pathology measurement data from the University of California's breast biopsy database into her machine learning tool and added mathematical calculations, the tool could learn how to predict which biopsies would have cancer. She then retested with another sample to help determine which measurements were most important, and verified the machine learning with a third sample, which predicted the biopsy results with a 99 percent accuracy.
The computer analysis of an individual biopsy takes nanoseconds to perform.
This discovery could change cancer diagnosing timeframes to minutes instead of days, alleviating a great deal of stress for the patients with negative biopsies and getting treatment faster for those with cancer. Also, the more data the tool learns from, the more specific it can get. It also would be exciting to integrate genomic data and other data (demographics, lab results, risk factors, etc.) that could give us more information about the correlations with a cancer.
Metastatic AI is another exciting tool to add to the fight against a deadly disease. For more visit: http://metastaticai.com.
Deployed Medicine - a platform in use by the Defense Health Agency to trial new innovative learning models aimed at improving readiness and performance of deployed military medical personnel. The intent is to deliver personalized, dynamic learning using the most current and accessible technology, enabling a self-directed and continuous study of learnings and medical best practices. Android and iOS apps available.
Learn more at http://www.deployedmedicine.com
4D Healthware - Founded by Star Cunningham, this chronic condition management platform provides a dashboard with each patient's care plan, a record of adherence and progress and exception reports when a patient needs intervention. The communication between the platform and the patient exceeds the 20-minute requirement of non-face-to-face care coordination/management and is reimbursed under CPT-99490.
Learn more at http://www.4dhealthware.com
MACRA Year 2 Readiness Webinar | UCF HealthARCH
12/20/17 | Orlando, FL
Lake Nona IMPACTForum
2/28/18 | Orlando, FL
March 5-9, 2018 | Las Vegas, NV