Gonzalo Alvarez@agonzalorodrigo27667

This project explores the use of deep learning techniques for multilabel thoracic disease classification using chest X-ray images. The main objective was to build an initial medical imaging pipeline capable of detecting multiple pulmonary and thoracic conditions from radiographic studies. The project was developed using PyTorch and pretrained DenseNet architectures adapted for medical imaging tasks. Different experiments were conducted using grayscale chest X-rays resized to higher resolutions in order to preserve relevant anatomical information. The dataset includes multiple thoracic conditions such as Atelectasis, Cardiomegaly, Consolidation, Edema, Effusion, Emphysema, Fibrosis, Hernia, Infiltration, Mass, Nodule, Pleural Thickening, Pneumonia, Pneumothorax, Pneumoperitoneum, Pneumomediastinum, Subcutaneous Emphysema, Tortuous Aorta and Calcification of the Aorta. The system was adapted from an initial binary classification approach into a multilabel classification pipeline using BCEWithLogitsLoss and sigmoid activation during inference. Several challenges related to medical datasets were addressed, including label alignment, class imbalance, tensor dimensionality issues and GPU optimization using CUDA in Google Colab. The project also incorporated evaluation methodologies more suitable for medical AI tasks, especially AUROC per pathology instead of relying exclusively on global accuracy metrics. This allowed a more realistic analysis of the model’s performance across different diseases. Additionally, exploratory work was performed to prepare the integration of explainability techniques such as Grad-CAM heatmaps to visualize the anatomical regions that contribute to the model predictions. This project represents a first practical exploration into AI-assisted radiology and medical computer vision, focusing on understanding the complete deep learning workflow for chest X-ray analysis rather than only optimizing benchmark performance.

Build solutions that improve how software is built. Work with an AI that understands your codebase, helping you reduce repetitive work and build with real context. ⏳ Build your project in 48 hours. 🚀 Get hands-on with IBM Bob and explore how AI fits into real development workflows before it becomes standard. 💡 Build tools and workflows that developers would actually use. ⭐ Learn directly from experts throughout the event. 🤝 Join a team or build solo. 💰 Prize pool: $10,000 🧑🏻‍💻 Register before the Kick-Off Stream and start building with Bob.

Grab your ideas, spin up compute, and start building what’s next! Start with $100 in AMD Developer Cloud credits and begin building right away. Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further. To unlock credits and join the hackathon, sign up for the AMD AI Developer Program. 🤝 Go solo or team up — ship something real. 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs. 📍 On-site Venue (May 9-10): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 📍On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑‍💻 Developers, builders, and innovators - bring your boldest ideas and let’s turn them into real AI systems on AMD.