Healthcare Data Matters Podcast

The COVID pandemic resulted in a high incidence of patients with respiratory distress. While the use of mechanical ventilation helped some patients, others experienced acute lung failure, leading to a steady increase in the need for ECMO.

The challenge with patients on ECMO is that even the smallest movement can be life-threatening, resulting in the need for constant bedside supervision. Staff at Johns Hopkins needed to monitor these complex patients while finding ways to reduce frontline caregiver risk.

Learn more about:

• Initiated virtual rounding so providers could monitor patients from offices, conference rooms, and home
• Augmented bedside staff with an eye-piece camera and microphone to provide an added layer of monitoring support to virtual rounding staff
• Enabled access to all real-time and retrospective waveform data from the bedside so care teams could remotely automate trends, recognize patterns, plot correlations, and more accurately and quickly assess patients and reduce risk

For a full replay of the webinar with accompanying slides please visit: https://healthcaredatamatters.com/on-demand/

Information: Season 1, Episode 8 Aired live on May 25, 2021 Audio: 8:27

Medicine and healthcare can use advanced data science technology to predict patient deterioration and risk.  However, challenges in data aggregation, access, and sharing can slow adoption and use of these technologies.  

Listen in to hear from Dr. Nagy and Dr. Bergmann from Johns Hopkins Medicine and learn more about how they have solved the data challenge and are working with institutions across the country to accelerate the field.  

Learn more about: 

• The Johns Hopkins Precision Medicine Platform and how it is used for tracking patient hospitalization trajectories during the pandemic 
• Leveraging the platform for collaboration and deploying a severe Risk prediction algorithm rapidly at the start of the pandemic
• Future plans to help accelerate the vision of patient-centered AI through common data models and collaboration networks

For a full replay of the webinar with accompanying slides please visit: https://healthcaredatamatters.com/on-demand/

Information: Season 1, Episode 7 Aired live on April 14, 2021 Audio: 10:12

Data science, machine learning, and AI can be applied to a wide spectrum of healthcare tasks, from modeling the spread of COVID across a city to arrhythmia detection on an individual patient.

Using these tools requires data and expertise from both engineering and medical communities. While these groups often collaborate on worthwhile projects, it can be difficult to start new projects across industries and institutions due to a variety of factors including geography, contracting, and regulations. 

Motivated by a desire to train the next generation of data workers, the Data2Knowledge Lab at Rice University has created an ecosystem that brings together clinical and engineering researchers in the service of answering healthcare questions with data. 

Join Genevera Allen, PhD, the founder and director of the D2K Lab, as she discusses how she developed the program, leveraging the largest industries in Houston to find interesting problems and foster research opportunities. Particular projects with institutions in the Texas Medical Center have included:

• Detecting arrhythmias in pediatric populations
• NLP chatbot for actively monitoring COVID symptoms with University of Health Science Center at Houston
• Detecting cardiac instability in children prior to cardiac arrest (with Texas Children’s Hospital)
• Finding genomic signatures in age-related macular degeneration (with Baylor College of Medicine)
• Impact of COVID-19 on EMS services (with City of Houston Fire Department)

Learn more about how the program was built and best practices for working with academic universities in your area to create a similar program. 

For a full replay of the webinar with accompanying slides please visit: https://healthcaredatamatters.com/on-demand/

Information:  Season 1, Episode 6 Aired live on March 16, 2021 at 12:30pm CT Audio: 9:07

Sedation is a necessary part of managing any patient during and after surgery, but also has well-known risks, including respiratory depression. When it comes to neonates and babies, sedation management is even more complex due to the size of these patients, their underdeveloped systems, and their inability to communicate. These factors not only lead to other short-term risks such as severe respiratory failure, bronchopulmonary dysplasia, hemodynamic failure, necrotizing enterocolitis and longer duration of mechanical ventilation, but studies also show that repeated or lengthy use of sedation can negatively impact baby’s brains, leading to a lifetime of challenges that can include learning disorders, behavioral dysfunction, ADHD and more. 

Join Dr. Achuff as she demonstrates how the team at Texas Children’s Hospital (TCH) is leveraging bedside data and advanced analytics to reduce short- and long-term risks for these complex patients. Hear how the visual analytics tools developed at TCH have:

• Reduced opioid exposure by 90%
• Reduced benzodiazepine exposure by 28%
• Reduced the number of patients on four or more medications by 50%

Also learn how the visual analytics have been adapted for COVID to enable clinical distancing and virtual rounding to protect care teams and improve both care coordination and patient outcomes. And, find out how TCH is using real-time patient data to create groundbreaking research in the space of hemodynamics, medication management and more.

For a full replay of the webinar with accompanying slides please visit: https://healthcaredatamatters.com/on-demand/

Information:  Season 1, Episode 5 Aired live on February 2, 2021 at 12:00pm CT Audio: 10:02

The specialty of neonatal hemodynamics has the potential to reduce death and brain injury among premature babies. While widely adopted in other countries as a standard of care, the specialty is beginning to gain traction in the U.S. because of the promise it holds in enabling increased precision in the identification of patients at greatest risk of cardiovascular morbidity, a more targeted approach to use of medical therapy, and enhanced longitudinal appraisal of response to treatment.

For a full replay of the webinar with accompanying slides please visit: www.healthcaredatamatters.com

Information: Season 1, Episode 4 Aired live on December 8, 2020 at 12:30 CT Audio: 9:25

Up to 48% of post-op pediatric cardiac patients with congenital heart defects develop post-operative arrhythmias. Junctional ectopic tachycardia (JET) is considered the most common type of tachycardia seen during early post-operative care, but there are currently no automated detection algorithms for JET.

Dr. Jain  discusses how he leveraged both data and expertise across academic and industry partners to create algorithms to detect JET. Hear how this approach can accelerate clinically impactful research.

For a full replay of the webinar with accompanying slides please visit: www.healthcaredatamatters.com

Information: Season 1, Episode 3 Aired live on November 10, 2020 at 12:30 CT Audio: 8:05

Neonates with critical congenital heart disease have an immature cardiovascular systems and increased risks for hemodynamic instability and sudden cardiac arrest. Hemodynamic management of these patients helps decrease that risk. Dr. Lasa discusses how he and his colleagues at Texas Children’s Hospital use data and remote monitoring tools to better manage these complex patients in everyday patient care at one of the largest cardiac intensive care units dedicated to neonates with congenital heart disease in the country.

Hear how they are also leveraging that data to accelerate research and develop and deploy patient centered AI at scale. Dr. Javier Lasa is the Cardiac Critical Care Intensivist, Cardiovascular ICU, Texas Children’s Hospital

For a full replay of the webinar with accompanying slides please visit: www.healthcaredatamatters.com

Information: Season 1, Episode 2 Aired live on October 20, 2020 at 12:30 CT Audio: 8:55

Patients with hypoplastic left heart syndrome (HLHS) have a high risk of critical deterioration. And since their physiologic profile differs from other patient populations, specialized clinical training is required to care for these patients. Join Dr. Craig Rusin as he discusses how he and his colleagues use data to predict these events before they happen.

For a full replay of the webinar with accompanying slides please visit: www.healthcaredatamatters.com. Dr. Rusin's research paper: Prediction of imminent, severe deterioration of children with parallel circulations using real-time processing of physiologic data

Information: Season 1, Episode 1 Aired live on October 1, 2020 at 12:30 CT Audio: 7:14