Very insightful article. When you consider AI in the next post, I hope you'll also look at the oft-neglected richness of data that can exist in the DICOM realm ; not just for radiology but also potentially ECG and visible light modalities too.
💯I have a piece on multi-modal and multi-task AI coming in a few weeks! Integrating signals from multiple data sources is one of the core strengths of neural networks, and multi-tasking actually makes models *better* — a fact that's under-appreciated IMO.
Great. I will be keen to read that and have some friends looking into a clinical problem that needs multimodal data into an AI model, they will also be keen to read it.
Very insightful article. When you consider AI in the next post, I hope you'll also look at the oft-neglected richness of data that can exist in the DICOM realm ; not just for radiology but also potentially ECG and visible light modalities too.
💯I have a piece on multi-modal and multi-task AI coming in a few weeks! Integrating signals from multiple data sources is one of the core strengths of neural networks, and multi-tasking actually makes models *better* — a fact that's under-appreciated IMO.
Great. I will be keen to read that and have some friends looking into a clinical problem that needs multimodal data into an AI model, they will also be keen to read it.
Great post, excited to read more about how AI will impact healthcare.