James Caras’ Post

Curious to learn more about how I'm modeling both student and grader active time on task? Come check out my talk. This has been the culmination of a lot of user research + big data + software development and I'm excited to introduce the newest feature in Data Insights! #timeontask #acsspring2024 #gradingtime #datainsights

Delighted to share that my team "Cookie Monsters" won the 2nd runner-up title in the "Biomed Datathon 2024" organized by mHealth Lab, BME-BUET yesterday. Participants were given a multi-label classification task, that is to detect 4 heart abnormalities solely from the heart audio recordings of patients. The main challenge was to make the train set (comprising only ~60 samples) usable for effective learning pipelines. Along with basic exploratory data analysis and data augmentations, our approach outlines both classical machine learning techniques (decision tree, random forest, xgboost, multi-layer perceptron) and attention-based modern deep learning techniques (audio spectrogram transformer) to achieve ~60-65% macro F1 score and ICBHI score. Solution Notebooks: https://lnkd.in/gUH5wAWu Presentation Slide: https://lnkd.in/gi4ZgXDh

Almaden Genomics’ g.nome® is the omics data science solution that provides scientists with an end-to-end approach to life sciences discovery. With g.nome, you can simply: 🔎 Select the right datasets for your research using simple queries. ✏️ Pick one of our pre-built, fully modifiable workflows to get quick results and visualizations. 📊 Dig deeper with associated tertiary analysis for a fuller understanding of underlying factors. g.nome enables you to modify any steps along the way with complete traceability and reproducibility of your results. Learn more: https://lnkd.in/dhxj7Z5r

Early Access Paper Just Posted: Contrastive Graph Pooling for Explainable Classification of Brain Networks. We propose a contrastive dual-attention block and a differentiable graph pooling method called ContrastPool to meet the fMRI-specific characteristics, such as low signal-to-noise ratio, node alignment and limited data scale. Read the paper: https://lnkd.in/grqg5xud

Learning Points! Save time and cost. Enable more data to be extracted from a single experiment. Monitor cell progression by differentiating between single generations. Read application note: https://ow.ly/SEAp50QgxsC #flowcytometry #cellproliferation #drugdiscovery #encodingdyes #multiplexedassays #cellpopulations

Uncover Insights: Customer Loyalty & Switching in Library Prep Kits 🧬 Dive into BioInformatics' newest publication, delving deep into the realm of library prep kits. Gain crucial insights into purchase decisions and supplier loyalty dynamics. 📥 Eager to Learn More? Access the report here: https://hubs.la/Q02yJZs80

😃😺 With this workshop, you will learn to carry out a supervised statistical learning project from start to finish! Register here: https://lnkd.in/ejSQgFEe #ML #randomforest #hyperparameters #supervisedlearning #tidydata

🚀 [New Article] 🚀 : An overview of classifier-free guidance for diffusion models This new blog post by Nikolas Adaloglou and Tim Kaiser presents an overview of classifier-free guidance (CFG) for diffusion models and recent advancements in CFG based on noise-dependent sampling schedules. It's been a while since our last article, so I really hope you find this useful. Part 2 is coming soon. Stay tuned. Link: https://lnkd.in/dVucTJXe

Reproducibility in multi-omics analysis boils down to standardization - but what does that look like? Join me for our August 22nd webinar. We’re hosting a session on transforming your multi-omics research with reproducible environments. Some topics we’ll cover: 🔹 Handling diverse data types, standardizing methods, and managing varying analysis pipelines. 🔹 Dealing with leftover code, unorganized tools, and difficulties collaborating on analysis. 🔹 Establishing standardized workflows, ensuring documentation and transparency. 🔹 Real-world examples of improved reproducibility in multi-omics research. We hope you can join us! https://lnkd.in/gtKXGB9n