Implantable Device for Spinal Cord Injury
The HEPIUS Lab includes team members from the JHU School of Medicine, School of Engineering, Applied Physics Lab, and multiple companies. Together, we are creating a system of interactive devices based on ultrasound technology for continuous and active monitoring of patients with spinal cord injury. My contribution is to enable automated calculation of blood perfusion at the injury site.
Rapid, Intraoperative Foreign Body Object Detection
Figure source: Laura Roy
Leaving behind surgical items in the body burdens both patients and hospitals with millions of dollars spent every year. In this study, we have explored the powerful combination of deep learning, intraoperative ultrasound imaging and app development to demonstrate, in both animal and human studies, the ability to automatically detect and localize foreign body objects in neurosurgery.
Citation: H. G. Abramson et al., "Automatic Detection of Foreign Body Objects in Neurosurgery Using a Deep Learning Approach on Intraoperative Ultrasound Images: From Animal Models to First In-Human Testing." Frontiers in Surgery, 2022. https://doi.org/10.3389/fsurg.2022.10400.
Patent application filed (co-inventor); licensed an option agreement.
Automating Cardiac Image Segmentation
Segmenting the left ventricle myocardium is a critical first step to analyzing heart pathologies. We developed a deep learning segmentation algorithm tailored to clinical application. This work out of the Trayanova Lab at JHU has been published in the Cardiovascular Digital Health Journal and was selected for poster presentation at the 2020 American Heart Association Conference.
Citation: D. M. Popescu, H. G. Abramson, et al. “Anatomically-Informed Deep Learning on Contrast-Enhanced Cardiac MRI for Scar Segmentation and Clinical Feature Extraction.” Cardiovascular Digital Health Journal, 2021. https://doi.org/10.1016/j.cvdhj.2021.11.007.
Patent application filed (co-inventor).
Magnet Tracking for Prosthetics
We developed the fastest multiple magnet tracking algorithm with implications in minimally-invasive prostheses, augmented reality, and other applications.
Citation: C. R. Taylor, H. G. Abramson and H. M. Herr, "Low-Latency Tracking of Multiple Permanent Magnets," in IEEE Sensors Journal, vol. 19, no. 23, pp. 11458-11468, 1 Dec.1, 2019, doi: 10.1109/JSEN.2019.2936766.