Reisel González Pérez’s Post

Growing up with a radiographer mum, I've been able to witness the evolution of radiology in recent history. I remember many times helping to manually develop x-ray film in dark rooms using developer/fixer/wash and have witnessed firsthand the digitisation of imaging too. It’s exciting now to see this evolution continue with the exploration of AI in radiology. Microsoft Research has recently posted an article that delves into how GPT-4, an advanced AI model, has the potential to reshape radiology reporting. GPT-4 has shown remarkable possibilities in processing radiology reports, offering new insights in disease classification and findings summarisation. It’s important to know that this integration of AI into healthcare is more than just automation; it's about enhancing diagnostic accuracy, treatment planning, and patient engagement. As someone who lives and breathes IT, I am excited about the transformative power of technology, this evolution is particularly exciting and reminds me of why I love working in IT - the potential to improve lives through technology. I’ll be watching this one closely and can’t wait to see the profound impact it could have on improving patient care. https://lnkd.in/gkSFY_sN #HealthcareInnovation #AITransformation #Radiology #DigitalHealth #OpenAI #MicrosoftAI #Microsoft

The AI Lens: Transforming Medical Diagnosis Hi there! I recently published a new article as part of my computer vision series. This time I focused on Semantic Segmentation and its applications in medical imaging and how it's changing the way we care for patients. I'd love to share my experience in this research field and inspire others to contribute as well. In the article, I explained the basic principles behind Semantic Segmentation and explored how this technology is enhancing diagnostic accuracy and opening up new possibilities for patient care. It's truly amazing to see how far we've come and how much more we can achieve with AI in medicine. It would be great if you could take a look at my latest article and share your views. Engaging in conversations about such groundbreaking technology is what drives the industry forward. #HealthcareInnovation #AIDiagnostics #MedicalTechnology #SemanticSegmentation

Generative AI has shown great promise in generating high-quality medical images from low-quality input data. For example, a generative adversarial network (GAN) can be trained to generate high-quality images of organs from low-resolution ultrasound images. https://lnkd.in/euT3yJyk #generativeai #artificialintelligence #dataengineering #machinelearning

Progress happening in MedPaLM2 generative AI  1. Introduced MedLM for Chest X-rays, which has the potential to help transform radiology workflows by helping with the classification of chest X-rays for a variety of use cases. 2. Seeing promising results from fine-tuned models on complex tasks such as report generation for 2D images like X-rays, as well as 3D images like brain CT scans 3. Introduced AMIE (Articulate Medical Intelligence Explorer), a research AI system built on an LLM and optimized for diagnostic reasoning and clinical conversations. #ai #healthcare #medpalm2 #google https://lnkd.in/gAv3HJxc

Great Summary about #AI in #LifeSciences & #Healthcare by Eric Topol, MD, issued a few weeks ago:   👉 Hippocratic AI released the first safety-focused #LLM for real-time patient conversations. Topol shares details about how that LLM operates   👉 AI-enabled discovery of antibiotics 💊, with a first new class in decades   👉 Pathology 🔬 LLMs: a step towards general-purpose pathology unimodal model   👉 Radiology 🎞 & AI: incremental performance support with AI is highly variable and unpredictable   👉 AI Health Misinformation 🤥: 4 LLM assessed and show differences across models. Overall this confirms potential for LLMs to propagate health misinformation.   🎤 Topol’s conclusion: “We can't get to A.I.-powered healthcare without compelling prospective clinical trial evidence in the real world of medical care, with diverse participants. That’s what we’re missing now, and hopefully will start to see soon.”   https://lnkd.in/gC_x6vX6

GPT-4’s potential in shaping the future of radiology - In recent years, AI has been increasingly integrated into healthcare, bringing about new areas of focus and priority. One notable advancement in this area involves on medical competency exams and benchmark datasets. https://lnkd.in/gWm-P3pt #Microsoft #C9 #Cloud9 #GPT #LLM #AI #ML #AIroadmap #HealthTech #Healthcare #Cloud9 #C9

Google's groundbreaking new multimodal medical AI, ELIXR, could revolutionize medical imaging 🚀 Combining language and image analysis, ELIXR is lightweight and perfect for disease classification, semantic search, and radiology report verification 🏥 Its modular design allows fine-tuning for specific tasks, making it adaptable for various applications. However, maximizing AI's potential in medicine demands ongoing research and collaboration among experts, healthcare providers, and institutions 🌟 Article: https://lnkd.in/g6JunmBW Blog: https://lnkd.in/gpaMGZBz . . . #MedicalAI #Innovation #HealthcareResearch #aiadvancements #ai #google

GPT-4’s potential in shaping the future of radiology - In recent years, AI has been increasingly integrated into healthcare, bringing about new areas of focus and priority. One notable advancement in this area involves on medical competency exams and benchmark datasets. https://lnkd.in/eigj6NSG #Microsoft #C9 #Cloud9 #GPT #LLM #AI #ML #AIroadmap #HealthTech #Healthcare #Cloud9 #C9

I'm proud to share the release of two new AI models added to Azure AI Health Insights service. Radiology Insights is an AI model that provides quality checks with feedback on errors and mismatches to ensure critical findings are surfaced and presented using the full context of a radiology report. In addition, follow-up recommendations and clinical findings with measurements documented by the radiologist are flagged. Patient-friendly reports is an AI model providing patients with more accessible versions of their clinical reports. The simplified report explains diagnoses, symptoms, anatomies, procedures, and other medical terms in accessible language while preserving medical accuracy and precision. Hadas Bitran, Uri Einav, Jean-Luc Verschelde, Rachel Wities, Franck Atlan, Joeri Van der Vloet, Ann Vanhooren, Yochai Lehman, Ksenya Kveler, Aaron (Ari) Bornstein, Ran Efrati Asaf Levi, Shay Slobodkin, Reut Gross, Asaf Arnon, Shahar Yekutiel, Tal Baumel, Yehudit Meged, Tamar Sidi Maoz Bert Hoorne https://lnkd.in/d-nXk-iH

Medical image #annotation plays a vital role in #radiology by labeling and marking images to train AI models and assist #radiologists in diagnosis. These annotations provide crucial information for accurate image interpretation, identifying and marking areas in X-rays, CT scans, MRI, ultrasounds, and text-based medical records. The annotated data is then utilized to train ML models with deep learning algorithms, enhancing #medical #diagnosis accuracy. Accurate annotation of text, notes, and metadata is essential for developing valuable ML models in the field. #MedicalImaging #AI #Radiology #ML #HealthTech Connect2Cure Healthcare Services (C2C) Click to know more: https://lnkd.in/gYkviY6M