Automated image processing can increase efficiency for a diverse range of applications from defect detection in manufacturing to tumor detection in medical images. As spatial and temporal resolution of image data increases, image datasets are becoming unwieldy for traditional approaches to image processing, which require loading images in memory. In this presentation we’ll explore a new approach for processing large image datasets, such as medical images and video, that leverages Hadoop.
We’ll cover different ways to represent images in database or Hadoop, the advantages of doing so, and demonstrate basic image processing techniques including smoothing, segmentation, object detection, and classification. No prior knowledge of image processing required – the presentation will cover the basics and highlight specific use cases from medical imaging and digital pathology.
Ailey is a Senior Data Scientist at Pivotal Inc focusing on life sciences and healthcare. She holds a Ph.D. in Biophysics from UC Berkeley where her research focused on applying novel atomic force microscopy (AFM) techniques to cell mechanics. She has worked in both the Biomedical Imaging Group and Angiogenesis Group at Genentech improving automated image processing and employing machine learning to detect tissue types and architecture for digital pathology. Ailey also teaches optics and microscopy to middle schoolers as a volunteer for Citizen Schools and works with the Clearity Foundation to maintain a database of clinical trial information relevant to Ovarian Cancer.
Comments on this page are now closed.