Mission

AI for RT Delivery Error Detection and Mitigation

Our group is developing algorithms, tools and techniques for the automatic detection of human errors in the alignment of image-guided radiotherapy setup images. These algorithms could be used as an interlock to prevent treatment of the wrong body part, wrong patient, or with significant alignment errors (e.g. wrong vertebral body).

• Development and interinstitutional validation of an automatic vertebral-body misalignment error detector for cone-beam CT-guided radiotherapy
• Development and multi-institutional validation of a convolutional neural network to detect vertebral body mis-alignments in 2D x-ray setup images
• Proof-of-concept study of artificial intelligence-assisted review of CBCT image guidance
• Offline generator for digitally reconstructed radiographs of a commercial stereoscopic radiotherapy image-guidance system

Human Factors Research

We study how humans interact with automation and technology to improve quality, safety and efficiency of radiotherapy treatments.

• Barriers and facilitators to clinical implementation of radiotherapy treatment planning automation: A survey study of medical dosimetrists

Computational Tools for RT Image Analysis

Some of the computational tools we developed for our project which we think may have interest to others. Code publically available at links below.

• Feasibility of a deep-learning based anatomical region labeling tool for Cone-Beam Computed Tomography scans in radiotherapy
• Offline generator for digitally reconstructed radiographs of a commercial stereoscopic radiotherapy image-guidance system

Auto-Segmentation

Our group develops and clincally validates auto-segmentation algorithms, with a focus on human-machine cohesion and clinical evaluation.

• Machine-assisted interpolation algorithm for semi-automated segmentation of highly deformable organs
• Human factors in the clinical implementation of deep learning-based automated contouring of pelvic organs at risk for MRI-guided radiotherapy