Mahmoud Ghandi

Benchmarks! Yes! ~ as AI models becoming these increasingly complex multimodal and generative beasts, it’s crucial that we keep benchmarks up to date, challenging, relevant and informative.

11

For drug discovery scientists, consultants and vendors … Announcing targeted high-impact initiatives at HitchhikersAI + a review paper describing the challenges in the adoption of AI & Data.  Initiatives: ✅ Find bite-size solutions for your scientific problems ✅ Build relationship opportunities with biotechs, by consultants partnering with vendors ✅ A market penetration app (LLM-driven), to help vendors diagnose why they are struggling and to make actionable suggestions ✅ Going from data to quality decisions, fast ✅ Helping scientists cost-effectively construct decisions from large sets of research papers For more info on the initiatives: https://lnkd.in/epDWQuB4 Call to Action: ❓ [SCIENTISTS] Do you have a scientific problem where you need a bite-size solution? We may be able to find something for you. All we need is for you to send us your scientific problem. ❓ [VENDORS & CONSULTANTS] Would any of the initiatives above help you engage with your customers faster? These initiatives come with the support of our Steering Committee and our friends at 20/15 Visioneers and Zifo RnD Solutions.  Together, we will leverage our networks to ensure that participating vendors & consultants are able to engage with and get feedback from scientists. Join Us [Nothing to sign, Nothing to pay]: HitchhikersAI is a grass-roots no-cost community, seeking to accelerate drug discovery and development with AI & Data.  Please join our initiatives and shape them to directly help your scientific and business needs. Contact me to get involved: raminderpal@hitchhikersai.org SLAS (Society for Laboratory Automation and Screening)  ELRIG UK  Drug Target Review  Drug Discovery World  Sanj Kumar  Lesley Granberg  Vicki Loise  Pistoia Alliance  Mary Geismann  Boston Cambridge Biotech Networks  San Diego Biotechnology Network  San Francisco Biotech Networks  myBIO Community - Biotechnology connections myBIO Community - Biotechnology connections cvijic boris  Jesse Johnson  Anne Carpenter  Abhishek Jha  Catarina (Katie) Campbell  John Overington  Becky Upton BioIndustry Association (BIA) John F. Conway Neil Wilkie Andreas Bender Janette Thomas Andre Franca, PhD Nina Truter Sujeegar Jeevanandam Eli Pollock Jamie Cartland Jorge Cortell Tim Jarrett Charles Sinclair Graham Combe Sitta Sittampalam Tony Sedgwick https://lnkd.in/eicivKZf

40

I created a question regarding use of AI/ML in diagnosing disease in USA at Manifold Markets: https://lnkd.in/gKYC6Yr7 Medicine moves slowly hence the long time window (9 years from today) Note, this requires AI/ML tests to become routine, not sometimes-used-by-some-doctors

14

2 Comments

My #positconf2024 Magnetron AI talk's now on YouTube! Magnetron AI's a great example of how Regeneron invests in AI to follow through on our shared commitment to improve patient lives. I built the patent-pending MAGNETRON AI matchmaker from scratch with Jordan Leao and Jennifer Michel completely in-house, and supercharged it with LLM (large language model) technology, to enable rapid, yet accurate, company-wide pairings between employees seeking skill training/growth in any domain and others specifically capable of facilitating it. https://lnkd.in/egMy3MP4 Magnetron's an innovatively autonomous AI recommendation engine that tirelessly thinks for itself, performing and acting on complex reasoning that is based on mind-bogglingly complex data, but fully under the control, and in the best interests, of the humans that have asked it to do so. This has resulted in consistent 10x or more acceleration of the matchmaking process (reducing 11 months of work down 1) with each cohort, even after expanding considerations to simultaneously include ALL available data (which humans simply cannot do), rather than only manually-reviewable subsets. #regeneron #ai #llm #datascience #pharmaceuticals #drugdevelopment #r #python #shiny #knowledgegraph #aiengineering #autonomoussystems #neo4j #training #continuingeducation #matchmaking #doingwellbydoinggood #posit Posit PBC

68

2 Comments

Merck doesn't stop surprising my #rstats + #biostatistics soul! About year ago I wrote about the incredible number of powerful R tools to design and analyze #clinicaltrials: Mediana ( Alex Dmitrienko, Gautier Paux ), rpact, gsDesign (Keaven Anderson), gMCP, multxpert, nlmixr, RCTdesign, MAMS, MCPMod, mmrm, emmeans, and lots, LOTS of more. /💡 At my work I use over 360 packages, about 60 on regular basis. / Currently no other open source statistical package can compete with R in this field. R especially shines in adaptive designs (MAMS - multi-arm multi-stage), simulations, and super flexible multiplicity adjustments for multi-hypothesis studies, implementing the graph-MCP framework (Maurer & Bretz, covering: fixed sequence, fallback and serial/parallel gatekeeping methods). And today I learned about the gMCPLite (Yalin Zhu , Yilong Zhang , Keaven Anderson , Nan Xiao , Kornelius Rohmeyer , Florian Klinglmueller et al.) https://lnkd.in/d_j3ZV6B - a new gem in my toolbox after their gsDesign 🤩 With all those big initiatives of the #pharmaceutical industry combined with #RConsortium, like #openpharma (Daniel Sabanés Bové), #pharmaverse - making R #cdisc speaking (Thomas Neitmann), #rvalidationhub, #camis - the #PhUse working group for the numerical validation against #SAS and other statistical packages), ➡️ opensource is in #clinicalresearch to stay! ... and to think that in 2012 I was treated (more or less explicitly) like a lunatic when only I mentioned R used in pharma...

135

3 Comments

👉🏼 Chapitre 7. Applications et éclairage éthique des Intelligences Artificielles Génératives linguistiques en Médecine 🤓 Cécile Monteil 👇🏻 https://lnkd.in/e4Fnfwqw 🔍 Focus on data insights: - The integration of Generative Artificial Intelligence systems like Large Language Models (LLM) is reshaping medical practices. - Physicians are required to adapt and acquire new competencies in technology for effective utilization of Gen AI. 💡 Main outcomes and implications: - The evolving ethical framework surrounding Gen AI in medicine necessitates increased ethical vigilance. - Healthcare professionals need to balance technological advancements with preserving their medical expertise and critical thinking skills. 📚 Field significance: - The digitalization of medicine emphasizes the importance of training physicians in new technologies to maintain quality healthcare services. - Gen AI presents both risks and opportunities that require careful consideration within the healthcare sector. 🗄️: [#GenerativeAI #Healthcare #EthicalFramework #DataInsights]

3

1 Comment

Next up on my list of books from Manning Publications Co. is Machine Learning for Drug Discovery by Noah Flynn. I've always been curious about the intersection of technology and healthcare so this book was an immediate purchase. Chapter 1 is dense with new terminology but I'm enjoying every bit of learning how the drug discovery process works and how machine learning can help redefine the problem. Traditionally, drug discovery is a chemical compound search problem. Through a virtual screening process, researchers search through the chemical space, which is estimated to contain 10^60 drug-like molecules. They then match those compounds to human protein structures which will be the targets of the drugs. There are about 10^5 of those. Combine those and you have a massive search space that we couldn't even fathom completely searching before the sun explodes. 🤯 Machine learning techniques allow researchers to convert the chemical search space to a functional search space based on known properties and features of molecules. This lends well to generative AI which can produce candidate molecules based on the desired properties of known drugs (potency, toxicity, safety, etc.) and knowing what the target protein structure is. This is known as "de novo" drug design. If you're in the space, please share any resources you find interesting.

18

3 Comments

I've seen (and been involved in) so many (too many?) conversations in the past few years about the definitions of "bio-tech" vs. "tech-bio" in the preclinical space. So I'd like to be the first (?) to declare that "tech-pharma" is also a thing to be acknowledged and reckoned with -- i.e., the exploitation of all the various recent advances in AI/ML, data, compute, etc. to accelerate, derisk, and upvalue in the much larger, more expensive, and way more baroque (and broken!) world of pharma clinicial development. Call a spade a spade. #techpharma

22

3 Comments

Interesting discussion at “AI/ML is Transforming R&D and Business” panel at BIO International Convention - led by Maude Tessier with Karen Akinsanya Joseph Mcdonald Jiye Shi and Frank Grams. What is the value of AI/ML = one way to assess: ask whether decisions made by a drug discovery/development project team with the help of genAI or physics based approaches/models enhance the probability of success, accelerate, improve outcomes. The data we use for training models is only limited to the small fraction of targets/problems what we have tried to address. The more important data sets are the additional ones being generated to allow us to target the targets that haven’t been addressed previously.

29

We are a week and a half away from our JMP Clinical webinar where we will discuss the future of clinical trial review, analysis, and reporting through interactive visualization software. Registrants will be exposed to a number of features currently available in JMP Clinical 18, and catch a brief glimpse of two features coming in Version 19. JMP Clinical provides value across R&D departments, uncovering the story hiding in plain sight and giving clear insight to research teams. Hope you and your colleagues will join us for this informative session! Registration link is below. #statistics #clinicaltrials #datavisualization #safety

29

2 Comments

Join us for our last WIB AI virtual panel discussion of the year, featuring visionary discussions exploring how AI is transforming life sciences - You don't want to miss it!

23

The data used to train an AI model are critical to the success of the model. If data sources are limited or data quality is poor, the model will be negatively impacted. We have already seen reports of how poor quality data produces nonsense results (e.g., Google search results suggesting that humans should consume one small rock each day to acquire vitamins and minerals). AI software is designed, built, trained, and tested by humans. It will never be free of bias, mistakes, or errors. If we intend to use AI as a tool rather than a toy, we must verify the reliability and validity of the results. We must consider how results are generated. We must consider the data underlying the results. We must have criteria for accepting or rejecting results. We must move beyond the hype of AI and remember that it is simply software and not in any way magical.

1

In our brand new collaborator article, Eric J. Daza, DrPH, MPS explores the evolution of his work from causal inference in biostatistics to pioneering methods for observational causal inference using n-of-1 time series data, driven by the promise of personalized digital health insights. 👇Click below to read more https://lnkd.in/eiiMQaWk #biostatistics #statistics #data #science #education #career #blog #causalInference — At biostatistics.ca our mission is to create an information-rich community around biostatistics. Do you have interesting career stories, project successes or technical skills you would like to share with the community? Send us a message and become a certified biostatistics.ca writer!

13

Leaders from Cornell University and other research institutions across New York state gathered with Governor Kathy Hochul at Cornell Tech this week to celebrate Empire AI, a historic investment in research resources to advance AI technology that can benefit society and do the greatest good. Greg Morrisett and I were pleased to represent Cornell as one of seven founding institutions in the Empire AI consortium, along with Columbia University, New York University, Rensselaer Polytechnic Institute, the State University of New York, The City University of New York and the Flatiron Institute.   We thank Governor Hochul for her vision and leadership. Cornell researchers are already adapting AI to the most urgent issues facing our world today, from agriculture and nutrition to policy and medicine. By establishing a shared academic research computing facility, Empire AI ensures that Cornell and other New York-based academic institutions can drive innovation and discovery in New York state using the most advanced AI tools to pursue responsible, purpose-driven research. https://lnkd.in/eepJiJRM

234

4 Comments