What are specific clinical applications of deep learning AI that NetApp enables in healthcare organizations?
Answered at 20:40 in IoT Trojectory Podcast - Episode 19 - Interview with Esteban Rubens - Healthcare AI · IoT Trajectory Podcast Series
What are specific clinical applications of deep learning AI that NetApp enables in healthcare organizations?
Early sepsis prediction is one major use case where deep learning models trained on EHR data including vital signs and medical history can warn clinical staff four to six hours before a patient will experience sepsis, allowing earlier intervention. Readmission prevention is another application where keeping patients in the hospital one or two days longer can reduce readmission risk significantly. Additionally, deep learning is being used to combat clinical burnout by streamlining data entry, avoiding repetition, and prioritizing alarms to reduce alert fatigue. Nuance's ambient computing technology uses natural language processing to transcribe clinical conversations without requiring clinicians to stare at screens.
Full answer and detailsRead the whole answer, key facts and every other question from this episode
In short
Early sepsis prediction is one major use case where deep learning models trained on EHR data including vital signs and medical history can warn clinical staff four to six hours before a patient will experience sepsis, allowing earlier intervention.
Esteban Rubens, Leader for the Healthcare AI Practice at NetApp · IoT Trojectory Podcast - Episode 19 - Interview with Esteban Rubens - Healthcare AI · IoT Trajectory Podcast Series · answer at 20:40
Key takeaways
- Early sepsis prediction is one major use case where deep learning models trained on EHR data including vital signs and medical history can warn clinical staff four to six hours before a patient will experience sepsis, allowing earlier intervention.
Quote to remember
“We want to help people get to the point in which they can do their jobs. We want to remove frustration from the process.”
— Esteban Rubens
The full answer · about 4 min
Early sepsis prediction is one major use case where deep learning models trained on EHR data including vital signs and medical history can warn clinical staff four to six hours before a patient will experience sepsis, allowing earlier intervention. Readmission prevention is another application where keeping patients in the hospital one or two days longer can reduce readmission risk significantly. Additionally, deep learning is being used to combat clinical burnout by streamlining data entry, avoiding repetition, and prioritizing alarms to reduce alert fatigue. Nuance's ambient computing technology uses natural language processing to transcribe clinical conversations without requiring clinicians to stare at screens.
Key facts
- Data scientists spend 75 to 90 percent of their time on data preparation and non-science-specific tasks
- Clinical staff can receive a sepsis warning four to six hours before a patient experiences sepsis using deep learning models
- Readmission risk can potentially be reduced by 80 percent in some cases by keeping patients in hospital one to two days longer
- Healthcare industry experiences widespread clinical staff burnout across North America, Western Europe, and Asia
- 100 gigabit per second networking is now available for healthcare AI infrastructure projects