Welcome to HCLS AI/ML Content Catalog

Blogs
Workshops
Home

Contact

Copyright © 2024 | Yankos

Home > Blogs

Now Loading ...

Blogs

An international study presenting a federated learning AI platform for pediatric brain tumors
In a collaboration led by Stanford Health Department of Radiology, the AWS team worked with an international team of scientists to build FL-PedBrain, a federated learning platform developed for pediatric posterior fossa brain tumors, addressing the challenges of limited data sharing in medical AI studies. The platform enables joint tumor classification and segmentation across 19 international sites without direct data sharing. FL-PedBrain’s performance is comparable to centralized data training, with only slight decreases in classification and segmentation accuracy. Notably, the platform demonstrates improved segmentation performance on external sites and explores the impact of data heterogeneity and imbalances in real-world scenarios. Resources : Paper: An international study presenting a federated learning AI platform for pediatric brain tumors
Blogs · 2024-06-28
ESM3 support on Amazon SageMaker and AWS HealthOmics
EvolutionaryScale has made their breakthrough ESM3 model family available through Amazon SageMaker JumpStart and AWS HealthOmics. ESM3 is a frontier generative model, trained on data from 3.8 billion years of evolution, to understand and generate proteins that have never existed in nature. Life science customers can use ESM3 to design better drug candidates, reducing the time and cost required to bring lifesaving cures to patients Life science customers can run ESM3 on the best AWS service for their use case, including: Interactively explore protein design techniques on Amazon SageMaker Jumpstart Integrate ESM3 into end-to-end protein engineering workflows with AWS HealthOmics Build and scale generative AI applications for life science with Amazon Bedrock Take advantage of built-in security, customization, responsible AI, and high-performance, cost-effective infrastructure Resources : AWS Industries Blog: Revolutionizing Generative Biology with AWS and EvolutionaryScale by Matt Wood ESM3 example workflow on AWS HealthOmics ESM3 example notebook for Amazon SageMaker
Blogs · 2024-06-28
Pre-training Genomic Language Models Using AWS HealthOmics and Amazon SageMaker
Genomic language models, such as HyenaDNA, analyze and interpret genomic sequences using principles similar to natural language processing models. These models benefit from the integration of AWS HealthOmics and Amazon SageMaker. AWS HealthOmics provides efficient genomic data storage and management, while SageMaker offers scalable training and deployment capabilities. The process involves converting both public and proprietary genomic data for analysis, facilitating real-time inference and advanced genomic research. This approach enhances the accessibility and usability of complex genomic data for researchers. For more details, visit the blog post. Resources : Blog post Code Repo
Blogs · 2024-05-31
Find the Next Blockbuster with NVIDIA BioNeMo Framework on Amazon SageMaker
The blog post discusses how the NVIDIA BioNeMo framework, integrated with Amazon SageMaker, can accelerate drug discovery. It explains the challenges of drug R&D, highlighting the cost and time involved. By leveraging generative AI for drug discovery, the BioNeMo framework allows researchers to develop advanced AI solutions for analyzing biomolecular data, thus potentially reducing the time and cost associated with bringing new drugs to market. For more information, you can read the full blog post here. Resources : Blog post Code Repo
Blogs · 2024-03-11
Efficiently fine-tune the ESM-2 protein language model with Amazon SageMaker
This AWS blog post introduces a method to efficiently fine-tune the ESM-2 protein language model for predicting protein subcellular localization using Amazon SageMaker. It highlights the importance of proteins in drug development and explores the capability of large language models (LLMs) in protein sequence analysis. The post details a solution for addressing the challenges of model size and training costs, presenting techniques such as gradient accumulation and low-rank adaptation for efficient training. For a comprehensive overview, you can read the full blog post here. Resources : Blog post Code Repo
Blogs · 2024-03-06
Improving Patient Experience with Digital Front Door
Navigating healthcare plan benefits can be confusing for patients, leading to dissatisfaction and a lack of engagement. An AWS blog post introduces a digital solution to streamline this process, making it easier for patients to understand their coverage. By employing an AI-powered chatbot, the solution offers 24/7 access to plan benefits information, improving patient engagement, care management, and strengthening provider-patient relationships. For more details, visit the AWS blog post here. Resources : Blog post Demo
Blogs · 2024-02-29
Use RAG for drug discovery with Knowledge Bases for Amazon Bedrock
The blog outlines using the Retrieval Augmented Generation (RAG) technique with Knowledge Bases for Amazon Bedrock to enhance drug discovery. It details creating a knowledge base on Amazon Bedrock, integrating it with RAG for effective drug discovery research, emphasizing the ease of querying vast amounts of data. This approach aids in analyzing complex, unstructured data sets, like clinical trial documents, offering a scalable, serverless solution for innovative drug research. For more insights, visit the full blog post. Resources : Blog post
Blogs · 2024-02-29
Generative AI-Powered Clinical Intelligence - Safely Driving Better Outcomes
The blog discusses leveraging Generative AI for clinical intelligence in healthcare, focusing on unlocking insights from unstructured data like doctor’s notes while ensuring patient privacy. AI21 Labs’ Contextual Answers on Amazon SageMaker JumpStart aids in accurately answering clinical queries, mitigating risks like misinformation. This tech enhances research, patient care, and incorporates social determinants into health analysis, emphasizing responsible AI use in sensitive healthcare applications. For a detailed understanding, check the full post here. Resources : Blog post
Blogs · 2024-02-28
Interact Conversationally with AWS HealthLake
Navigating the vast sea of structured and unstructured health data poses a significant challenge for healthcare providers, impacting care quality and efficiency. AWS introduces a game-changing solution in a recent blog post: a conversational interaction model leveraging AWS HealthLake and advanced language models. This innovative approach enables healthcare professionals to query and access critical health information naturally and intuitively, promising to revolutionize patient care by making data more accessible and actionable. For a deeper dive into how AWS is transforming healthcare data interaction, check out their blog post. Resources : Blog post Demo Video
Blogs · 2023-09-23
Predicting Whether a Breast Cancer Sample is Benign or Malignant
For this model, we will build, train and deploy a Multi-layer Perceptron using the sklearn library. This is a breast cancer diagnoses dataset, where, for each sample, the sample is diagnosed as “Benign” or “Malignant”. For each sample, a number of features are given as well. The source of the dataset is the UCI Machine Learning Repository. Resources : Github
Blogs · 2021-12-20
Predict Average Medicaree Hospital Spending
Medicare is a national health insurance program, administered by the Center for Medicare and Medicaid Services (CMS). This is a primary health insurance for Americans who are aged 65 and older. Medicare has published historical data showing hospital’s average spending for Medicare Part A and Part B claims based on different claim types and claim periods covering 1 to 3 days prior to hospital admission up to 30 days after discharge from hospital admission. These hospital spending are price standardized and non-risk adjusted, since risk adjustment is done at the episode level of the claims spanning the entire period during the episode. The hospital average costs are listed against the corresponding state level average cost and national level average cost. In this notebook, the data is used to build a machine learning model using Amazon SageMaker built-in Linear Learner algorithm, which predicts average hospital spending cost based on the average state level spending and average national level spending. The predicted cost can be used for purposes of budget and for negotiating pricing with the hospitals. From the hospital’s perspective, the predicted average hospital spending provides visibility to claim financials that can be used by the hospitals to increase their efficiency and level of care. Resources : Github
Blogs · 2021-12-20
Automate Retraining of Obesity Models using SageMaker Pipelines
This workshop shows how you can build and deploy SageMaker Pipelines for multistep processes. In this example, we will build a pipeline that: Deduplicates the underlying data Trains a built-in SageMaker algorithm (XGBoost) A common workflow is that models need to be retrained when new data arrives. This notebook also shows how you can set up a Lambda function that will retrigger the retraining pipeline when new data comes in. Resources : Github
Blogs · 2021-12-20

Touch background to close