Zhihong Zou

How can #blockchain transform healthcare data management? Abderahman Rejeb, Mike G., Anjum Khurshid, Alex Norta, Jim Poteet, Suresh Sivagnanam and I believe that the changes will be #disruptive and profound. In our newest publication in Blockchain in Healthcare Today - DMV Chapter, we identify important areas related to current #challenges, #stakeholders, #legal environment, #trust and #privacy, #benefits, #data #integrity, data #security, data #interoperability, data #sharing and #transfer as well as numerous #future #developments. This paper should not only help #researchers to identify interesting areas but is also relevant for #practitioners to better understand the potentials of blockchain for the #healthcare #industry. The open access paper can be found here: https://lnkd.in/dBrgqJ7r This paper is based on a #panel #discussion that can be watched here: https://lnkd.in/dQamzKNp We hope that we can raise awareness and contribute to a more efficient #healthcaresystem that will benefit #patients and the #industry. Many thanks to John Russo and Tory Cenaj for their ongoing support.

36

5 Comments

The work that Anthropic is doing to unravel the black-box nature of #LLMs is brilliant, foundational, and highly commendable. They are addressing a key challenge in any neural network -- explainability. Not only can this research lead to safer and more-accurate models, but the ability to explain WHY a model generated a specific response is critical in many regulated industries. Bravo!

19

Healthcare will be impacted the most by AI 🏥 An AI agent just outperformed nurses in understanding lab reports, drug toxicity, and impact on Zoom videos… 🙄 Soon AI, not humans, will be analyzing lab reports, diagnose symptoms, and recommend medication. AI will be far more robust and error-free in this critical area than humans!! AI can completely transform healthcare in the near term, with AI doctors and nurses and medical devices that perform surgeries… 👩‍⚕️🤖 The big obstacle for AI adoption in this space is red-tape, but the institutions that adopt AI and cut through all the red tape will be the ultimate winners 🙏 #healthcare #healthtech #digitalhealth #healthcareai #aiinhealthcare #hospitals #nurses #patientflow #ai #artificialintelligence #deeplearning #alzacare #

9

4 Comments

Wondering how to improve exponentially the performance of your LLMs???? Your solution could be CRITIC! 🤖🗣️🔧 CRITIC allows LLMs, which are essentially “black boxes” to validate and progressively amend their own outputs in a manner similar to human interaction with tools. More specifically, starting with an initial output, CRITIC interacts with appropriate tools to evaluate certain aspects of the text, and then revises the output based on the feedback obtained during this validation process. This offers an element of getting external feedback; in many cases this gives much better results than only using LLMs to critique itself. You should take a look to this cheatsheet by Val Andrei Fajardo, ML Engineer at LlamaIndex, that guides you through all the most relevant components of CRITIC. 📝 You can take a look to the paper below 👉 https://lnkd.in/dhskQ8rw #ai #genai

57

6 Comments

Excellent report from General Dynamics Information Technology John P. Curran regarding adoption and trends of AI in the Govt. I used Meta to generate summary below and included link to original article. Main AI use cases are chat and creating/updating computer code. Note also the 23% failure rate of AI projects due to user expectation misalignment. This can be overcome with thorough protoyping and establishing use case metrics. - Successful AI project completion timelines are narrowing, with an average of 14 months from pilot to production. - Future demand for AI projects is building around chat and coding-related functions. - 54% of surveyed AI experts said projects currently under development involve improving computer code. - 59% of projects involve "research and understanding", which may relate to creating chat functions. - 23% of AI projects fail due to misalignment between AI solutions and user expectations. - Partnership efforts are required for successful AI projects. Data diversity is crucial, and incorporating diverse data sources can lead to better AI outcomes. Trends: - AI adoption is increasing across the Federal government, with a growing focus on chat and coding-related functions. - Agencies are leveraging cloud-native services, tooling, and other capabilities to accelerate AI project development. - There is a growing interest in translating older code to newer languages, such as Python or cloud-based analytic tools. - Domain-specific chat functionality is becoming a priority, with many agencies planning to implement chat services in the next two years. Statistics: - 325 AI experts surveyed across the Federal government - 14 months: average time from pilot to production for AI projects - 54%: percentage of projects currently under development that involve improving computer code - 59%: percentage of projects that involve "research and understanding" - 23%: percentage of AI projects that fail due to misalignment between AI solutions and user expectations. Bill Casola John Grim Leslie Anne McIndoe Kim Giffin Michael S. Freeman III Tobias Quaranta https://lnkd.in/eqvmAVFG

4

The rise of AI in medicine is exemplified by #Google's #MedGemini. But the blend of optimism and skepticism is here. Can #AI truly match the nuanced understanding that doctors offer? Historically, people frustrated with conventional medical advice turned to "Doctor Google" often with mixed results. Today, AI tools like #ChatGPT are stepping into this arena. With again mixed results. Now they promise that "Med-Gemini destroys GPT's benchmark and outperforms doctors". It's critical to remember that AI is a tool — not a replacement for human expertise. As we marvel at AI's potential to analyze vast data sets and even engage in complex reasoning, the technology's current limitations must be acknowledged. AI can suggest, but not substantiate, it can propose, but not empathize. At TenderHut, we recognize the transformative potential of AI across industries, including healthcare. Our AI-powered healthcare solutions analyze medical results, generate alerts, and assist with bespoke dictionaries for illnesses, medications, and procedures. I agree that Med-Gemini, represents a significant leap forward in clinical diagnostics and has vast potential for real-world application. ➡ It is designed to handle various types of medical data, including images (such as X-rays and CT scans), genomics, and other relevant information. It excels in understanding these different modalities simultaneously. ➡ Unlike general-purpose AI models, Med-Gemini is specifically fine-tuned for medical use. It leverages its core capabilities from the Gemini family to reason about complex medical scenarios. This means it can provide more accurate and context-aware predictions. ➡ Med-Gemini-2D sets a new standard for AI-based CXR report generation. Expert evaluation shows that it outperforms previous results by a significant margin. For both normal and abnormal cases, a substantial percentage of AI-generated reports are considered equivalent to or better than radiologists’ reports. ➡ Med-Gemini-3D achieves the first-ever large multimodal model-based report generation for 3D CT volumes. While 53% of AI reports are considered clinically acceptable, further research is needed to match expert radiologist reporting quality. ➡ Med-Gemini-2D performs exceptionally well in CXR VQA tasks, surpassing state-of-the-art or baseline models in 17 out of 20 tasks. ➡ It also excels in image classification tasks related to histopathology, ophthalmology, and dermatology. It outperforms baselines across 18 out of 20 tasks. ➡ Med-Gemini-Polygenic improves disease risk prediction beyond the standard linear polygenic risk score-based approach. It even generalizes to genetically correlated diseases not explicitly trained for. This marks an exhilarating phase in #ClinicalDiagnostics. I believe that AI is indeed a superb advancement, but its role should be to enhance human capabilities, not to replace them. #HealthTech #MedicalAI #PatientCare #Technology #Healthcare

24

Totally agree with you, Anthony. The name of the game is to get the right context as tight as possible to the question or task on hand. Graphs are one good way to do this, but you should use all the tools in your belt. With our solution, we use graphs, tables, key value pairs, geospatial information, regular search, vectors, and so on. When you combine all these things, the probability of getting the right context as much higher.

5

Finally we have Llama 3 - the best open source LLM today. It comes in three flavors: 8B, 70B, and 400B. he 70B model performs similarly to Claude Sonnet (which powers the free Claude) Its still not there, compared with GPT-4, but open source models are getting there. #llms #genai #generatieveai #capgemini #llm #opensourceai #capgeminiindia #ai #artificialintelligence #software  #leaderboard #benchmark #benchmarking #llama3 #meta

26

1 Comment

This Forbes article couldn’t have said it better. People first, data second, then AI. By fostering collaboration, AI systems are not only technically capable but also ethically aligned with human values and the objectives of companies. Embracing the synergy between individuals and data unleashes the complete potential of AI. #ArtificialIntelligence #AI #ResponsibleAI

33

2 Comments

Can we have LLM routers? Unify has created a solution which abstracts the LLM being used - allowing us to use the best of breed. https://unify.ai/ An interesting concept to combine multiple LLMs #llms #genai #generatieveai #capgemini #llm #opensourceai #capgeminiindia #ai #artificialintelligence #software

6

1 Comment

🤓 𝗟𝗲𝘁'𝘀 𝗴𝗲𝘁 𝗻𝗲𝗿𝗱𝘆! Take a look at this Amazon Web Services (AWS) blog about "Project Heal" – an AI-powered toolkit designed for the early detection and mitigation of medical misinformation. The blog does an excellent job of explaining some the technical components of the system. Here are some highlights: 𝗗𝗲𝘁𝗲𝗰𝘁𝗶𝗻𝗴 𝗮𝗻𝗱 𝗖𝗹𝗮𝘀𝘀𝗶𝗳𝘆𝗶𝗻𝗴 𝗠𝗶𝘀𝗶𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻 “One of the key functionalities of Project Heal will be the ability to classify and detect emerging misinformation before it has an opportunity to proliferate.” “Ingested information will be evaluated using trained ML models, enabling the tool to classify the likelihood of a statement having misleading content and allowing for categorization based on the statement’s entities and context.” “Misleading statements will then be evaluated by a subsystem to help score the severity of the threat to human health.” “The detection engine will be built using graph neural networks, which will be supported by a large, scalable, fully managed graph database (Amazon Neptune). Amazon Comprehend will be used to support keyword and entity extraction from content.” “A human feedback loop will consistently audit and improve the model using Amazon Augmented AI (Amazon A2I).” “Both detection and scoring will be supported by Amazon SageMaker, a fully managed service to prepare data and build, train, and deploy ML models for any use.” “Additionally, generative AI will support the summarization and grouping of related misinformation content.” 𝗣𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗲𝗱 𝗖𝗼𝘂𝗻𝘁𝗲𝗿 𝗠𝗲𝘀𝘀𝗮𝗴𝗶𝗻𝗴 “One of the most innovative features of Project Heal will be the platform’s ability to generate tailored communication responses to misleading information.” “To achieve this, Project Heal will use generative AI foundation models (FMs) through Amazon Bedrock, such as Amazon Titan.” “Supporting evidence, built from trusted sources of information, will be available to users via Retrieval-Augmented Generation (RAG), an approach that reduces some of the shortcomings of large language model (LLM)-based queries.” “Through this technique, Amazon Bedrock is able to generate more personalized messaging by combining trusted information and user preferences.” 𝗧𝗵𝗲 𝗕𝗼𝘁𝘁𝗼𝗺 𝗟𝗶𝗻𝗲 This project illustrates the complexity of developing AI-driven solutions and the importance of integrating them thoughtfully to tackle real-world challenges such as medical misinformation. 🙂 With these insights, I hope I’ve satisfied your nerdy needs! https://lnkd.in/e3gzR7XG

32

4 Comments

While LLMs show immense potential in healthcare, there is a gap between its promise and actual application. As with any innovation journey, we must develop the necessary regulations and practices to close it. The piece below does an amazing job of highlighting the missing pieces to this AI puzzle. I think the biggest takeaway here is to not let our excitement get the best of us The longevity of these technologies will require a lot of thoughtful trial and error. #Healthcare #LLMs #AI

5

#ChatGPT continues to improve and so does CampTek Software’s usage of it. We are discovering ways to solve problems more efficiently and faster than ever before with the use of this #GenerativeAI capablity. https://lnkd.in/gsvPqkSi

5

Must see this new article from Prophecy >> The Case for Data Transformation Copilots (#AI-powered companions that leverage #GenerativeAI to boost productivity) — How they will change ETL by evolving far beyond conversational interfaces: https://lnkd.in/eidJ69YK ➕ See the Prophecy Data Transformation Copilot in Action at the Databricks Data + AI Summit June 10-13!

29

2 Comments

I wrote an article for infoq with practical examples of using generative AI to accelerate: * Learning and discovery * Articulating and understanding requirements * Maintaining alignment with stakeholders I think the hype over autonomous LLM agents is missing the immediate benefits today for seasoned professionals “who can describe what they want the LLM to complete and critically evaluate the result. These tools enable us to swiftly cross divides of domain language and scale large repetitive tasks down to interesting ones on a human scale. When used carefully, they even facilitate work fundamentally about human interaction.”

6

Costs for generative AI models are going in the wrong direction relative to every technological advancement of the past ten years. Costs should be going down by 10x in subsequent generations not up by 10x. LLMs can’t be the final form when it comes to AI architectures.

1

📢 Who Invented the Term QoE (Quality of End User Experience)? 📢 Ever wondered who coined the term QoE (Quality of End User Experience) and pioneered solutions to measure it, relating it to QoS (Quality of Service) for the first time nearly 20 years ago? Discover the innovator behind this groundbreaking concept and the impact it has had on the telecommunications industry. Watch the video below to learn more! 🎥👇 #QoE #drdavidsoldani https://lnkd.in/gcvF8Tux

11

1 Comment

An interesting WSJ article today. The savings should be measured in potential efficiency gains, not just reducing parameters. TRAINING MODELS: Of course 10 billion parameters take less compute cost than a 1 trillion parameter LLM. But there is only 1% of trained knowledge by definition. If we aggregate for an apples-to-apples basis 🍎🍏, there is no savings on a “knowledge-adjusted” basis. INFERENCING: Inferencing off of a smaller model by definition allows for efficiency gains. It also allows for private data to be used internally and create more accuracy/precision, which will drive adoption. There are a wide range of issues of how to access the right model, at the right time, in an efficient manner to allow for the mass market general user who at one moment wants to say write a poem, and the next moment wants to write a thank you letter with a generative graphic. CONCLUSION: Comercial users will benefit from smaller model efficiency gains, but adoption still means massive compute capacity needs. Mass market users will require a complex back-end provisioning to marry “mass” knowledge with “small” models — a null set? What are your thoughts? Happy 4th 🇺🇸 all.

6

1 Comment

Anthropic, the AI safety and research company, just unveiled the next generation of their Claude models: Claude 3.5 that is said to be more powerful than GPT-4o in coding and reasoning! #Claude #Anthropic #GenAI #ArtificialIntelligence #AI #OpenAI

8