Open praeducer opened 2 years ago
praeducer
commented
2 years ago Our system is to complicated for us to do this in one research phase or a single diagram. Therefor, this tasks needs to be broken up as I am doing in the DevOps Milestone Planning wiki.
praeducer
commented
2 years ago To align with our #bizdev goals, we eventually want to adapt COINSTAC’s approach to hospital settings similar to what BrainForge has done.
It’s a competitive space, but good research is being done we can leverage. This paper describes how to combine multiple hospital data sets and multiple machine learning approaches to increase accuracy of results like predictions.
Our approach uses generalizable data pre-processing and machine learning modeling stages that can be readily adapted for research and deployment in healthcare environments.
We show that this framework can consistently and robustly produce models that outperform similar single-source approaches across various healthcare demonstrations (by 6–33%), including 10 distinct chest pathology diagnoses, along with length-of-stay and 48 h mortality predictions.
We also quantify the contribution of each modality and data source using Shapley values, which demonstrates the heterogeneity in data modality importance and the necessity of multimodal inputs across different healthcare-relevant tasks.
The generalizable properties and flexibility of our Holistic AI in Medicine (HAIM) framework could offer a promising pathway for future multimodal predictive systems in clinical and operational healthcare settings. https://www.nature.com/articles/s41746-022-00689-4
Outcomes
Related tasks: