Automatic 3D MRI knee damage classification with 3D CNN using BigDL on Spark

Damage to the meniscus is a physically limiting injury that can lead to further medical complications. Automatically classifying this damage at the time of an MRI scan would allow quicker and more accurate diagnosis. Jennie Wang, Valentina Pedoia, Berk Norman, and Yulia Tell offer an overview of their classification system built with 3D convolutional neural networks using BigDL on Apache Spark.

Damage to the meniscus has been proposed as a precipitating event for osteoarthritis, a degenerative disease affecting millions a year with significant reduction in their quality of life. Additionally, damaged menisci assessed by magnetic resonance imaging (MRI)-based grading have been associated with greater odds of longitudinal cartilage loss than intact menisci. The grading of meniscus lesions is typically done by radiologists using a semiquantitative grading system that indicates whether or not a lesion exists and the severity of the lesion to inform further care. An automated system that can classify menisci based on the presence or absence of lesions has high clinical relevance as it would provide immediate objective results at the time of the MRI scan, eliminate intra-user variability, and enable automated comparison over time. Jennie Wang, Valentina Pedoia, Berk Norman, and Yulia Tell offer an overview of their classification system built with 3D convolutional neural networks using BigDL on Apache Spark. BigDL, a new distributed deep learning framework on Apache Spark, provides easy and seamlessly integrated big data and deep learning capabilities for big data users and data scientists. In the 3D imaging field, BigDL provides support with 3D image convolutions, 3D max pooling, and a 3D image augmentation library. Jennie, Valentina, Berk, and Yulia walk you through this complex use case, covering data preparation, model development, training, and more. Along the way, they present the challenges and novel solutions to significant problems and share insight into the ultimate deployment of BigDL as a platform and tool for enabling and implementing the MRI classification solution.