At Viz.ai, it’s incredibly affirming to hear experts discuss how AI is revolutionizing diagnostics, aligning with our own vision and experiences in the space.
Thanks for the mention Gaurav Singal, MD!
Healthcare data/tech advisor and investor. Physician and faculty @ Harvard Medical School. Fmr Chief Data Officer @ Foundation Medicine. Computer scientist @ heart.
Such a treat to write about computational diagnostics with the one and only Bapu Jena. This piece makes the case that diagnostics are the front door for AI in the clinical care arena - a thesis I’ve believed for a few years now and have seen play out many times over. It also argues that the key bottleneck is financing this entre - and starts to make the case for pharma to play a key role here.
h/t to Viz.ai, PathAI, Beacon Biosignals, Caris Life Sciences, and so many other AI diagnostics trailblazers whose pioneering work is featured here. And h/t to Bristol Myers Squibb and other pharmacos that have played a critical role in subsidizing and accelerating the clinical adoption of these tools.
Many thanks to the Los Angeles Times for sharing this with the world.
https://lnkd.in/ea_gSBZz
We're at a frontier with a vast horizon of possibility ahead.
AI may very well improve many aspects of healthcare.
The fundamentally evolving nature of medicine and healthcare has seen it become virtual, ambulatory, home-based, value-based, data-driven and technology enabled.
Quantifying the value-add from "with AI" (vs "without AI") in terms of health outcomes and health economics benefits (e.g., overall survival, progression free survival, safety, economic burden, humanistic burden etc.), to justify investment and financing, may be challenging and should be cautiously interpreted.
A crucial first step requires a structured, systematic approach to identify current and anticipated gaps/opportunities for AI integration with key decision-makers and stakeholders who are subject matter experts.
Validation and prioritization exercises must follow closely.
Investing in AI integration for public health screening and surveillance for disease prevention (prognostic and predictive modelling: Who are "high risk"/"rapid progressors"/"fast responders"? etc.) is an evident game changer for early risk identification, early diagnostics and importantly, early treatment... be it interventional or monitoring. Great short to mid term return on investment.
The bang for your buck at scale, albeit a long term ROI, will see improved timeliness of robust evidence for healthcare system governance, budgetary allocation and planning, resource allocation and strategy etc.
Let's get this right.
Healthcare data/tech advisor and investor. Physician and faculty @ Harvard Medical School. Fmr Chief Data Officer @ Foundation Medicine. Computer scientist @ heart.
Such a treat to write about computational diagnostics with the one and only Bapu Jena. This piece makes the case that diagnostics are the front door for AI in the clinical care arena - a thesis I’ve believed for a few years now and have seen play out many times over. It also argues that the key bottleneck is financing this entre - and starts to make the case for pharma to play a key role here.
h/t to Viz.ai, PathAI, Beacon Biosignals, Caris Life Sciences, and so many other AI diagnostics trailblazers whose pioneering work is featured here. And h/t to Bristol Myers Squibb and other pharmacos that have played a critical role in subsidizing and accelerating the clinical adoption of these tools.
Many thanks to the Los Angeles Times for sharing this with the world.
https://lnkd.in/ea_gSBZz
Healthcare data/tech advisor and investor. Physician and faculty @ Harvard Medical School. Fmr Chief Data Officer @ Foundation Medicine. Computer scientist @ heart.
Such a treat to write about computational diagnostics with the one and only Bapu Jena. This piece makes the case that diagnostics are the front door for AI in the clinical care arena - a thesis I’ve believed for a few years now and have seen play out many times over. It also argues that the key bottleneck is financing this entre - and starts to make the case for pharma to play a key role here.
h/t to Viz.ai, PathAI, Beacon Biosignals, Caris Life Sciences, and so many other AI diagnostics trailblazers whose pioneering work is featured here. And h/t to Bristol Myers Squibb and other pharmacos that have played a critical role in subsidizing and accelerating the clinical adoption of these tools.
Many thanks to the Los Angeles Times for sharing this with the world.
https://lnkd.in/ea_gSBZz
🧠 Unlocking AI in Clinical Research:
🏥 Discover how AI is revolutionizing #healthcare! Explore the cutting-edge applications reshaping clinical research, from drug discovery to personalized dosing.
🔗 Read more in the following article!
#ClinicalResearch#Innovation#MedicalTechnology#Elmea#Lifescience
Generative AI in Clinical Research: Revolutionizing Medical Data Analysis, Treatment Planning, and Innovation - Discover how generative AI is revolutionizing clinical research by enhancing medical data analysis, treatment planning, and innovation. Explore the potential of synthetic data generation, medical image synthesis, clinical decision support, drug discovery, and more.
https://hubs.li/Q02yyltg0
Thoughts on Fogo 2024 JAMA, "AI's Threat to Medical Profession." The lasting impression I got from this JAMA article from Agnes Fogo and colleagues is that, in so many words, you are a traitor to your species for labeling data for AI training, that you are giving away human knowledge to a machine that will ultimately destroy your livelihood. Their concluding sentence summarizes their view, "...physicians should realize that keeping AI within boundaries is essential for the survival of their profession and for meaningful progress in diagnosis and understanding of disease mechanisms." I applaud the authors for bringing up this timely and relevant topic, but they seem to forget that we really don't fully understand the way ANY treatment works. If we did, there would be no need for phase III and IV drug trials Most medical discoveries are serendipitous observations that are later studied to find the relevant underlying mechanisms (bisphosphonates for osteoporosis, sildenafil for erectile dysfunction, etc.). Discoveries from AI black box models are our superhuman serendipity. We still need reason, skill, and the scientific method to probe these discoveries that cure and reduce suffering to determine their mechanistic underpinnings. But to not use AI to perform mundane human-level tasks more efficiently, accurately, cost-effectively, and in a more accessible way (e.g. morphologic features from histopathology slides) is just protectionism of an aspect of a field that has lost its value. What are your thoughts?
Generative AI in Clinical Research: Revolutionizing Medical Data Analysis, Treatment Planning, and Innovation - Discover how generative AI is revolutionizing clinical research by enhancing medical data analysis, treatment planning, and innovation. Explore the potential of synthetic data generation, medical image synthesis, clinical decision support, drug discovery, and more.
https://hubs.li/Q02yyv3P0
Generative AI in Clinical Research: Revolutionizing Medical Data Analysis, Treatment Planning, and Innovation - Discover how generative AI is revolutionizing clinical research by enhancing medical data analysis, treatment planning, and innovation. Explore the potential of synthetic data generation, medical image synthesis, clinical decision support, drug discovery, and more.
https://hubs.li/Q02yyrbj0
Generative AI in Clinical Research: Revolutionizing Medical Data Analysis, Treatment Planning, and Innovation - Discover how generative AI is revolutionizing clinical research by enhancing medical data analysis, treatment planning, and innovation. Explore the potential of synthetic data generation, medical image synthesis, clinical decision support, drug discovery, and more.
https://hubs.li/Q02yypsT0
Ohio State University researchers just developed CURE, an AI model that can accurately estimate drug treatment effects and effectiveness without clinical trials. The model is trained on de-identified health records of over 3M patients, allowing it to gain a deep understanding of patient characteristics. The AI predictions are closely aligned with clinical trial findings in tests, showcasing the potential to generate insights that streamline drug testing.
With the ability to crunch massive medical datasets, CURE represents a significant step towards systems that can reliably estimate real-world drug effectiveness — potentially accelerating the discovery of new treatments without the cost and long timeframes of traditional clinical trials.
#ai#technology#innovationhttps://lnkd.in/gXbe_nyN