Events Add an event Speakers Talks Collections
 
Duration 03:25
16+
Video

C15 Adoption of a Deep Learning Risk Scale Predictive Model to Reduce 7-day Readmission of...

John Morrison
Assistant Professor of Pediatrics at The Johns Hopkins University School of Medicine
+ 6 speakers
  • Video
  • Table of contents
  • Video
Machine Learning for Healthcare
August 8, 2020, Online, Los Angeles, CA, USA
Machine Learning for Healthcare
Request Q&A
Machine Learning for Healthcare
From the conference
Machine Learning for Healthcare
Request Q&A
Video
C15 Adoption of a Deep Learning Risk Scale Predictive Model to Reduce 7-day Readmission of...
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Add to favorites
90
I like 0
I dislike 0
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
  • Description
  • Transcript
  • Discussion

About the talk

Adoption of a Deep Learning Risk Scale Predictive Model to Reduce 7-day Readmission of Respiratory Patients at a Pediatric Center

About speakers

John Morrison
Assistant Professor of Pediatrics at The Johns Hopkins University School of Medicine
Ali Jalali
Senior Data Scientist at Biofourmis
Hannah Lonsdale
Clinical Research Associate at The Johns Hopkins University
Paola Dees
Physician at Johns Hopkins All Children's Hospital
Brittany Casey
Pediatric Hospital Medicine Fellow at Johns Hopkins All Children’s Hospital
Mohamed Rehman
Eric Kobren Professor of Applied Health Informatics at The Johns Hopkins University School of Medicine
Luis Ahumada
ДолжностьDirector Center for Pediatric Data Science and Analytic Methodology at Johns Hopkins All Children's Hospital

Dr. Morrison specializes in the care of hospital patients as a pediatric hospitalist in the Department of Medicine at Johns Hopkins All Children’s Hospital. His clinical interests include the treatment of common pediatric conditions requiring hospitalization with a special interest in children with medical complexity. He is also interested in developing and practicing evidence-based pediatric medicine with an emphasis on patient-centered, cost-effective care. Dr. Morrison received his medical degree and his doctorate of philosophy, pathology and microbiology from the University of Nebraska Medical Center. He graduated from the inaugural Johns Hopkins All Children’s residency and completed a subspecialty fellowship in pediatric hospital medicine at Johns Hopkins All Children's Hospital. He is certified in pediatric advanced life support and the neonatal resuscitation program. Dr. Morrison’s research interests focus on characterizing mechanisms of antibiotic tolerance and discovering novel adjunct antimicrobial agents. He also has an interest in the discovery and development of biomarkers for acute respiratory illnesses and studying the utility of diagnostic testing. Dr. Morrison received the inaugural Allen W. Root, M.D. Award for Continuous Excellence in Residency in 2017, the inaugural Janet G. Root Award for Outstanding Research by a Fellow in 2019, and the Johns Hopkins All Children’s Hospital Patient Safety & Quality Award in 2019.

View the profile

Senior Data Scientist and Personalized Predictions Engine Team Lead at Biofourmis. Responsible for leading a talented team of data scientists focusing on utilizing data from various sources to predict patient deterioration or state of health. Experienced Doctor of Philosophy (Ph.D.) Data Scientist with deep knowledge and understanding of Data Science, Machine Learning, Mathematical Modeling, Data Visualization, Python, R, Matlab, Amazon Web services. Hands on experience on retrieving data, data pre-processing, and model development. Research professional with a strong track record of peer reviewed publications.

View the profile

Dr. Dees serves as the medical director for Utilization Management and physician adviser supporting the Utilization Management, Case Management and Health Information Management teams at Johns Hopkins All Children’s Hospital. In her role, she collaborates with many teams across the institution to ensure our facility is complying with health regulations and policies while providing high quality, evidence-based, comprehensive and family-centered care. Dr. Dees practices as a pediatric hospitalist at Johns Hopkins All Children’s Hospital. She cares for patients with common pediatric conditions, as well as children with medical complexity, who require hospitalization. Her clinical interests include improving discharge planning and evaluating processes that may help reduce or prevent readmissions. Dr. Dees joined Johns Hopkins All Children’s Hospital in 2011. She is a graduate of the Florida State University College of Medicine and completed her pediatric residency at the University of South Florida College of Medicine and All Children's Hospital. Dr. Dees is a member of the Alpha Omega Alpha Medical and Gold Humanism Honor Societies. She has received the Teaching Award from the University of South Florida Pediatric Residency program and the Outstanding Faculty Award from the Johns Hopkins All Children’s Hospital Pediatric Hospital Medicine Fellowship Program. Dr. Dees is board certified in pediatrics and hospital medicine. She is fluent in Spanish.

View the profile

Mohamed Rehman, M.D., is chair of the Department of Anesthesia at Johns Hopkins All Children’s Hospital and professor of anesthesiology and critical care and pediatrics with the Johns Hopkins University School of Medicine. Nationally recognized for his medical and clinical informatics expertise. Previously Dr. Rehman was a professor of clinical anesthesiology and critical care and professor of pediatrics at the University of Pennsylvania School of Medicine. He held numerous leadership roles at the Children’s Hospital of Philadelphia, including director of transplant anesthesia, and was the anesthesia team leader for the world’s first bilateral hand transplant and several conjoint twin separations. Dr. Rehman is board certified in anesthesiology, with subspecialty certification in critical care medicine and pediatric anesthesia. He also holds certifications from the American Board of Pediatrics and the American Board of Preventive Medicine, clinical informatics subspecialty. A graduate of Mysore Medical College, Mysore, India, he completed a pediatric residency at Chicago’s Cook County Hospital, an anesthesia residency at the University of Miami, and a fellowship in pediatric anesthesiology and critical care at Children’s Hospital of Philadelphia. He is the author of more than 50 original research publications and review articles and more than 70 scientific abstracts.

View the profile

Health Data Science professional and researcher at the enterprise level with novel and practical experience in biomedical and healthcare informatics, with demonstrated expertise in strategic visioning and effective leadership. Currently leading JHACH enterprise Data Science efforts for clinical care, research & operations. Extensive knowledge and hands‐on experience in the areas of data/text mining, web mining, machine learning, information retrieval, social network analysis and knowledge representation, and extensive expertise in: • Parametric and Nonparametric Bayesian statistics. • Classification (Neural Networks, Naive Bayes Classifier, Bayesian Networks, Decision Trees, SVM, etc.) • Clustering Analysis. • Natural Language Processing NLP. • Information retrieval and Information extraction. • Knowledge organization and representation (Ontology, metadata, and social tags). • Demonstrated ability to develop innovative data science approaches, methods, and algorithms. • Experienced in analyzing large‐scale imperfect real world datasets. Domain Specialties: Healthcare Informatics, Bioinformatics, Advanced Analytics, Artificial Intelligence, Data Mining, Machine Learning, Knowledge Management, Information Visualization, Visual Analytics, Intelligent Systems, Case Based Reasoning (CBR).

View the profile
Share

Open John Hopkins All Children's Hospital and on behalf of our team would like to share our experience to date with developing a predictive model to help reduce 7-Day respiratory readmission. Everything Children's Hospital there is increasing regulatory pressure at both state and federal levels to prevent and reduce readmissions. Unfortunately, little is known about how to safely accomplished it. Our approach is multifaceted, while many of our efforts were focused upon improving communication. We ultimately, thought to identify patients at high risk for raid Mission before they were ever

discharged. Our team developed a deep learning neural network for predicting 7 and 30 day readmission. Am I don't take data directly from the electronic medical record and compiles, a record for each patient. This information is then emailed in a report to our clinicians every morning prior to Our model had an area under the curve as a 0.76. The truth table, show the results of testing a model for three classes. Classes zero is no red Mission class one at 7. Mission and class to its 30-day readmission. The development of his model

provided a unique opportunity, for our teams, and not only evaluate how to implement a tool, but also SS performance against the clinical intuition have our providers. Respectively assess the risk stratification agreement between our hospitalist and the reveal Alpharetta prior to implementing the model. We at the hospital to assess, a patient has a high medium or low respiration Mission using percent how cut-off similarly assigned by the model. Overall, the provider in machine learning tool, provide an equivalent Batman. In 52.8% of patients, there was only slight agreement between the

provider and Tool as measured by a 0.104. 7% of patients in the study had what we deemed a clinically significant disagreement between low and high risk of red Mission there for the overall agreement of the readmission risk stratification. 53% of providers, that access to a reed mission risk or was at least somewhat likely to influence their decision. Making at the point-of-care, this suggests, there may be limited validity of army people. We assess the performance of man versus machine. In designating

patients at high risk for admission. Both the provider and the tool development has low practical, sensitivity, and positive predictive value for Respiratory. I'm sorry for already mission in this exploratory analysis, I'll do the Tilt performance was better. However, the results are limited by the small sample size and future, more adequately power and studies are necessary. Overall days of a predictive model alone is likely insufficient to change seven-day readmissions. On a pediatric medicine service, be readily accessible and easy to

navigate. A predictive augment a multidisciplinary approach to proactive. Thank you so much for your time and attention be well.

Cackle comments for the website

Buy this talk

Access to the talk “C15 Adoption of a Deep Learning Risk Scale Predictive Model to Reduce 7-day Readmission of...”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Ticket

Get access to all videos “Machine Learning for Healthcare”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Ticket

Interested in topic “Artificial Intelligence and Machine Learning”?

You might be interested in videos from this event

February 4 - 5, 2021
Online
26
106
ai, application, bot, chatbot, conversation, data, design, healthcare, ml

Similar talks

Hannah Lonsdale
Clinical Research Associate at The Johns Hopkins University
+ 5 speakers
Ali Jalali
Senior Data Scientist at Biofourmis
+ 5 speakers
Luis Ahumada
ДолжностьDirector Center for Pediatric Data Science and Analytic Methodology at Johns Hopkins All Children's Hospital
+ 5 speakers
Mohamed Rehman
Eric Kobren Professor of Applied Health Informatics at The Johns Hopkins University School of Medicine
+ 5 speakers
Walid Habre
Head Anesthesiological Investigations Unit at Université de Genève
+ 5 speakers
Nicola Disma
Director of Unit for Research & InnovationConsultant paediatric Anaesthetist at IRCCS Ospedale Pediatrico Giannina Gaslini
+ 5 speakers
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Ruijun Chen
Assistant Professor in Translational Data Science and Informatics at Geisinger
+ 8 speakers
Victor Rodriguez
MD-PhD Student at Columbia University
+ 8 speakers
Lisa Liu
PhD Candidate at Columbia Department of Biomedical Informatics
+ 8 speakers
Elliot Mitchell
Graduate Research Assistant at Columbia University Irving Medical Center
+ 8 speakers
Amelia Averitt
Manager, Clinical Informatics at Regeneron Pharmaceuticals
+ 8 speakers
Oliver Bear Don't Walk IV
PHD Student at Columbia University Graduate School of Arts and Sciences
+ 8 speakers
Shreyas Bhave
Software Engineering Intern at Counsyl
+ 8 speakers
Tony Sun
PhD Student at Columbia University in the City of New York
+ 8 speakers
Phyllis Thangaraj
Resident Physician at Yale New Haven Hospital
+ 8 speakers
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Konan Hara
Intern at TXP Medical Co.
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Buy this video

Video
Access to the talk “C15 Adoption of a Deep Learning Risk Scale Predictive Model to Reduce 7-day Readmission of...”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Conference Cast

With ConferenceCast.tv, you get access to our library of the world's best conference talks.

Conference Cast
949 conferences
37757 speakers
14408 hours of content
John Morrison
Ali Jalali
Hannah Lonsdale
Paola Dees
Brittany Casey
Mohamed Rehman
Luis Ahumada