Dr.Khaled Aboeldahab’s Post

https://lnkd.in/dU-pMnEA  This work showed that GPT-4 performance is comparable with that of physicians on official medical board residency examinations. Model performance was near or above the official passing rate in all medical specialties tested. Given the maturity of this rapidly improving technology, the adoption of LLMs in clinical medical practice is imminent. Although the integration of AI poses challenges, the potential synergy between AI and physicians holds tremendous promise. This juncture represents an opportunity to reshape physician training and capabilities in tandem with the advancements in AI.

Large language models (LLM) are becoming powerful tools to process unstructured data (documents, images, reports, etc). This trend will lead to new ways of discovering more accessible knowledge to the community. The generated insight could also benefit many downstream applications. As LLMs continue to surge, more works are expected to address daunting tasks like data cleaning, data labeling, data integration, and eventually outcome predictions. Some impressive healthcare case studies from top labs led by Akshay Chaudhari, Faisal Mahmood, Yang Xie, Tianlong Chen etc, that I have seen: 💡 EHR record summarization: https://lnkd.in/gNRnHWZp 💡 Extracting structured data from clinical note:  https://lnkd.in/gd9FByHe 💡 Knowledge graph and relational information processing: https://lnkd.in/g2_uT5bw 💡 "Talk to your images": image and language multimodal analysis: https://lnkd.in/gts_FE8k 💡 Accelerating medical text annotation via LLM  https://lnkd.in/gt4txvyi Connect with me for quality content in: #LLM #AI #ML #healthcare #GenAI

As a preliminary step toward integrating AI into medical practice, it is imperative to ascertain whether model performance is comparable with that of physicians.

In a recent Substack article, Eric Topol, physician-scientist, author, and editor, summarises three "new deep learning cardiopulmonary imaging studies, one transformer model (aka generative A.I., or large language model, LLM), some progress in virtual scribes and tomorrow (Part Two) a slew of new papers that are coming out." Read 👉Medical A.I. is on a tear here 👉 https://hubs.li/Q01Yj57v0 #ArtificialIntelligence

Foundation models are starting to emerge as a powerful primitives for building AI applications in healthcare and life sciences. At Bitfount, we're working on enabling the next generation of foundation models by combining them with federated data science. These new Federated Foundation Models (FFMs) are trained on 10x the data and are safer, more accurate and less biased while allowing access to siloed healthcare data. https://lnkd.in/eD8arunw

After four successful volumes in this series- Check out the call for 5th volume. If you are in the field of AI in Emergency and Critical Care- consider submitting manuscript/review/SR/MA below. (Deadline - August 27th, 2024) Frontiers

🚀 Exciting news in the world of AI and healthcare! A new study introduces an interpretable CNN-Multilevel Attention Transformer for rapid recognition of pneumonia from chest X-ray images, aiming to provide high-speed analytics support for medical practice. The framework emphasizes interpretability and computational efficiency, addressing key challenges in deep learning-based pneumonia recognition. This innovative approach has shown promising results in COVID-19 recognition tasks and demonstrates the potential to enhance clinical medical practice. #AI #Healthcare #PneumoniaRecognition #MedicalImaging #InnovativeResearch

AI can help doctors make better decisions and save lives.

AI beats doctors in clinical reasoning but is more error-prone: Artificial intelligence (AI), via the ChatGPT-4 program, can outperform attending physicians in clinical reasoning #EarthDotCom #EarthSnap #Earth

Check out recently published work by our group titled "Deep learning model utilizing clinical data alone outperforms image-based model for hernia recurrence following abdominal wall reconstruction with long-term follow up" The worked concluded that "The clinical-only DLM outperformed both the image-only DLM and the mixed-value DLM in predicting recurrence." https://lnkd.in/eurZtBNR