TensorHealth-NewsLetter on Medical AI

Exploring the Intersection of Artificial Intelligence and Healthcare Innovation

Opinion AI: Rethinking AI Benchmarks-Adapting to the Complexity of Agentic Systems

Microsoft, Meta and Anthropic have all recently announced plans to build AI agents that can execute tasks for humans autonom­ously on their behalf. To do this effect­ively, the sys­tems must be able to per­form increas­ingly com­plex actions, using reas­on­ing and plan­ning. How to evaluate these complex models? This article argues the need for new bench­marks.

News bits and bytes:

iKraph is the largest biomedical knowledge graph, extracting insights from 34M PubMed abstracts and 40 databases to advance drug discovery and disease research.

TxGemma is an open-source LLM suite designed for therapeutic development, achieving SOTA performance across 64 therapeutic tasks and improving explainability in AI-driven drug discovery.

Another huge AI breakthrough in drug delivery: Delete (Deep Lead Optimization Enveloped in Protein Pocket) tackles lead optimization by refining drug candidates directly within their 3D protein pockets.

Genentech and Roche's AI-driven Lab-in-the-loop (LitL) pipeline accelerates antibody discovery up to 100× faster by automating design, selection, and optimization. Their ensemble of generative models leverages biophysical priors for realistic, developable sequences.

OpenAI and start-ups race to generate code and transform software industry. Artificial intelligence groups are releasing models for computer programming that are reshaping coding jobs.

AutoGen v0.4: Reimagining the foundation of agentic AI for scale, extensibility, and robustness

An Open-Source AI Agent for Doing Tasks on the Web
NNetNav learns how to navigate websites by mimicking childhood learning through exploration.

The Collaborative Community on Ophthalmic Innovation-On-Demand Content

Health and Wellness

6 simple exercises to add to your next at-home workout by Kaiser Permanente

Upcoming Events

Keynote: Multimodal generative AI for precision health by Microsoft Research Forum