AI Case Study
Google Health improves predictions of hospital patient medical outcomes by using deep neural networks trained on 46 billion data points
Google Health three flavours of deep neural networks to help predict hospital outcomes, such as patient likelihood of death, when they will be discharged or readmitted or final diagnosis could be invaluable for personalising healthcare treatment. Analysing 46b anonymised data points they demonstrated that deep learning methods are capable of accurately predicting outcomes with the most impactful being the ability to predict patient deaths 24-48 hours before current methods.
Healthcare Providers And Services
"The biggest challenge for AI researchers looking to train their algorithms on electronic health records, the source of the data, is the vast, disparate, and poorly-labelled pieces of data contained in a patient’s file, the researchers write. In addition to data points from tests, written notes have traditionally been difficult for automated systems to comprehend; each doctor and nurse writes differently and can take different styles of notes.
To compensate for this, the Google approach relies on three complex deep neural networks that learn from all the data and work out which bits are most impactful to final outcomes. After analyzing thousands of patients, the system identified which words and events associated closest with outcomes, and learned to pay less attention to what it determined to be extraneous data. Typically, AI scientists have to carefully tinker with how their system interprets the data after it’s built, like which number of layers are needed to make the decision most accurately."
"While the results have not been independently validated, Google claims vast improvements over traditional models used today for predicting medical outcomes. Its biggest claim is the ability to predict patient deaths 24-48 hours before current methods, which could allow time for doctors to administer life-saving procedures." They also were able to predict 30-day unplanned readmission.
"...formulated three deep learning neural-network model architectures that take advantage of such data in different ways: one based on recurrent neural networks (LSTM), one on an attention-based time-aware neural network model (TANN), and one on a neural network with boosted time-based decision stumps."
R And D
Core Research And Development
"Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality." Being able to predict hospital outcomes, such as patient likelihood of death, when they will be discharged or readmitted or final diagnosis could be invaluable for personalising healthcare treatment.
"Google obtained de-identified data of 216,221 adults, with more than 46 billion data points between them. The data span 11 combined years at two hospitals, University of California San Francisco Medical Center (from 2012-2016) and University of Chicago Medicine (2009-2016)." The data spans from pre-admission to discharge.