Researchers Developed A Prognostic Machine Learning Model That Can Accurately Predict Survival And Recovery Outcomes Six Months After A Traumatic Brain Injury – MarkTechPost

Deep learning prognostic models combining admission CT scans and clinical data exceeded neurosurgeon forecasts in predicting 6-month death and poor outcomes following severe traumatic brain injury (TBI).
According to researchers, advanced machine learning can predict outcomes in patients with severe traumatic brain injuries.
According to a report published in Radiology, the researchers developed a deep learning model that analyses admission CT scans and clinical data from patients to predict six-month death and other adverse outcomes. According to the study, patients with severe TBI often require two weeks to come out of a coma, yet they are frequently taken off life support within 72 hours after being admitted to the hospital.
TBI is a severe public health concern. TBIs are most typically caused by falls, firearm-related injuries, motor vehicle crashes, and assaults, with over 64,000 TBI-related deaths documented in 2020. TBIs cause death, disability, and serious short- and long-term health problems because they change how the brain functions.
The research team started working on its deep learning system to satisfy this need. Two patient cohorts were used to train and validate the model: one of 537 patients treated for severe TBI at UPMC and another of 220 patients from the Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) collaboration.
Multiple patient brain scans are coupled with coma severity estimates and vital signs, blood test, and heart function information to produce the model’s predictions. The model was also trained on various picture-taking techniques, demonstrating that it can account for data inconsistencies and image quality in many types of brain imaging.
The model was compared to the International Mission on Prognosis and Analysis of Clinical Trials in TBI (IMPACT) model and three neurosurgeons’ forecasts. At six months after the traumatic occurrence, the model accurately predicted patient mortality risk and adverse outcomes, outperforming neurosurgeon forecasts.
Paper: https://pubs.rsna.org/doi/10.1148/radiol.212181
Reference: https://healthitanalytics.com/news/upmc-pitt-develop-machine-learning-model-to-predict-brain-injury-outcomes
Marktechpost is a California based AI News Platform providing easy-to-consume, byte size updates in machine learning, deep learning, and data science research
© 2021 Marktechpost LLC. All Rights Reserved. Made with ❤️ in California

source
Connect with Chris Hood, a digital strategist that can help you with AI.

Leave a Reply

Your email address will not be published.

© 2022 AI Caosuo - Proudly powered by theme Octo