Machine Learning Engineer - DICOM/Image Processing

Key Responsibilities 1. Model Development : - Build and train machine learning and deep learning models to analyze DICOM images, particularly X-rays. - Develop algorithms to detect common radiological abnormalities (e.g., fractures, tumors, infections) with high accuracy. - Experiment with state-of-the-art techniques, including CNNs, Transformers, and GANs for medical image analysis. 2. Data Processing : - Preprocess and augment DICOM datasets (e.g., normalization, resizing, denoising, and segmentation). - Extract metadata and pixel data from DICOM files using tools like pydico SimpleITK, or dicompyler. 3. Tool Development : - Collaborate with software engineers to build an interactive radiologist assistant tool. - Integrate ML models into a user-friendly platform for real-time image analysis and reporting. 4. Research and Optimization : - Stay updated with advancements in medical imaging and AI research. - Optimize models for accuracy, performance, and scalability. - Perform rigorous validation and testing to ensure clinical-grade reliability. 5. Collaboration : - Work closely with radiologists and domain experts to understand diagnostic workflows and requirement - Collaborate with backend engineers to ensure seamless integration with PACS and other medical systems. 6. Deployment : - Deploy AI models in production environments (cloud-based or on-premise). - Monitor model performance and iterate based on feedback. Required Skills and Qualification : 1. Education : - Bachelor's or Master's degree in AI&ML,Computer Science, Data Science or biomedical tech engineering Experience : - 2 years of hands-on experience in machine learning and deep learning. - Prior experience working with DICOM images and medical imaging datasets. 2. Technical Skills : - Machine Learning Frameworks : TensorFlow, PyTorch, Keras. - Medical Imaging Tools : Pydicom, SimpleITK, OpenCV. - Programming Languages : Python (mandatory), C++ (optional). - Algorithms : Proficiency in CNNs, object detection (YOLO, Faster R-CNN), a segmentation techniques (UNet, DeepLab). - Data Processing : Experience with DICOM standards, pixel data extraction, and metadata handling. - Database Knowledge : Familiarity with databases for storing large medical datasets (e.g., PostgreSQL, MongoDB). 3. Soft Skills : - Strong problem-solving and analytical skills. - Excellent communication skills to interact with radiologists and cross-functional teams. - Ability to work independently and in a team-oriented environment. 4. Preferred : - Experience in building radiology-specific AI tool - Knowledge of PACS integration and DICOM network protocols (C-STORE, C-FIND). - Familiarity with cloud platforms like (ref:hirist.tech)

Location: bangalore, IN

Posted Date: 3/1/2025