Industry News.
A curated collection of the most timely and relevant news, updates and information from the world of AI and machine learning.
Predicting the Response of Fungal Genes Using a New Machine-learning Approach
Scientists have developed a new machine learning model called FUN-PROSE that can predict how genes in fungi respond to changes in their environment. The model was trained on data from three different species of fungi and was able to accurately predict how genes responded to environmental changes in all three species. The researchers believe that FUN-PROSE could be used to study how genes are regulated in other organisms.
Researchers Develop Automatic Text Recognition for Ancient Cuneiform Tablets
Researchers have developed a groundbreaking artificial intelligence (AI) application that can decipher difficult-to-read texts on cuneiform tablets. The AI system uses 3D models of the tablets, which is a more reliable approach than previous methods that use photos. This new system makes it possible to search through the contents of multiple tablets to compare them with each other and opens up new possibilities for research.
NLP Tool Helps Make Radiology Reports More Readable
Researchers from the Yale School of Medicine developed a new tool that can make radiology reports more readable for patients. It discusses how the tool uses natural language processing (NLP) and prompt engineering to simplify medical language. The tool was tested on 209 radiology reports and was found to significantly improve readability. The researchers believe that this tool could be a valuable asset for improving patient communication in radiology.
AI Model Weighs Epigenetic Factors to Predict Cancer Outcomes
Researchers have developed a new AI-based approach to predicting cancer patient outcomes using epigenetic factors, which are proteins that control gene expression. The study found that tumors could be categorized into distinct groups based on epigenetic factor levels. These groupings were then used to accurately predict patient outcomes. The researchers believe that this study could pave the way for the development of novel cancer therapies
Artificial Intelligence Aids Fight Against Acute Myeloid Leukemia
UD and Nemours researchers, led by Mauricio Ferrato, utilized AI to identify effective drug therapies for acute myeloid leukemia (AML). The study, published in Bioinformatics Advances, employed machine learning on genetic data from 451 AML patients, marking progress in precision medicine and showcasing the impact of AI in healthcare.
Large Language Model Diagnoses, Triages Without Introducing Biases
GPT-4, a large language model, has demonstrated the ability to diagnose and triage patients on par with board-certified physicians without introducing racial or ethnic biases. This breakthrough could facilitate the adoption of conversational AI in healthcare settings. However, continuous monitoring of these models is crucial to ensure their performance remains consistent over time.