Image Guided Therapy (IGT) is a foundational discipline that involves the integration of imaging into procedural medicine used in surgery, radiation oncology, and interventional radiology. The notion that IGT is more efficient, more effective, and less expensive than conventional surgery has been confirmed both in extensive long-term studies and in ongoing revolutionary applications in the operating room. We have laid critical groundwork with this extraordinary technology during the last two decades.
During 2015-2020 our work was divided into four technology research and development (TRD) projects. The Prostate TRD developed novel platforms to allow for molecular profiling of prostate cancer and imaging to allow for improved predictive markers for guiding biopsy and focal therapy, and by use of image processing methods for monitoring treatment effect. The Neurosurgery TRD developed mass spectrometry as a biomarker for intraoperative surgical decision making to better define tumor extent. Advanced functional mapping using fMRI+DTI was used for individualized treatment planning. The Computation TRD developed specialized multimodal image registration algorithms and custom hardware for motion compensation to improve tumor localization and functional mapping. The Guidance TRD developed novel instrument and tumor tracking technologies to accurately deliver therapy despite significant physiological motion and tissue deformation.
We are currently building upon this foundation by introducing new technologies that will realize quantifiable benefits in terms of improved surgical and patient outcomes. The integration of these technologies will enable detailed assessment of tissue architecture heterogeneity (both morphologic and metabolic), image registration methods using deep learning, uniform data curation and annotation, and the use of software and devices to enable intraprocedural guidance. These technological research and precision targeting capabilities have the potential to make a significant impact on the effectiveness of cancer therapies, while vital tissues are spared. For the period 2020-2025, our research is divided into three technology research and development (TRD) projects.
Imaging Cancer Heterogeneity, is focused on the characterization of tumors and their surroundings at multiple scales (macroscopic, microscopic, and molecular). Improvements in these technologies directly improves overall risk assessment and precise treatment of cancer.
Deep Learning for IGT, is focused on developing information theoretic approaches for weakly-supervised deep learning and on applying them to image guided brain tumor surgery, lung surgery, and transperineal in-bore prostate biopsy. Additionally TRD2 is developing an infrastructure for curating and organizing image data, annotations, and metadata for deep learning.
Intraoperative Devices for IGT, is focused on developing intraoperative devices and to deliver them using the 3D Slicer and OpenIGTLink platforms. This includes an intraoperative needle guidance system for prostate biopsy, a visualization and navigation system for thoracoscopic lung surgery, and implantable microdevices with integrated retrieval mechanisms for difficult-to-access tissues, such as the brain.