Transformer-based models have demonstrated substantial potential in medical image segmentation tasks due to their exceptional ability to capture long-range dependencies. To further enhance segmentation performance, various effective methods have been proposed, including pretraining methods (weakly supervised or self-supervised pretraining schemes), contrastive learning schemes, and knowledge distillation methods. However, segmenting esophageal cancer (EC) from CT images remains a significant challenge, partly due to the complex anatomy of EC, such as variable shapes, extensive extents, and often blurred boundaries with adjacent anatomical structures. In this study, we propose a prior-guided pretraining (PGP) regimen based on bounding boxes, which enhances the model’s ability to discern textural differences between EC and the surrounding tissues. Using Swin UNITR as the backbone, our proposed pretraining scheme demonstrates superior performance in EC segmentation compared to other schemes. To further improve the segmentation accuracy of EC, we also addressed the class imbalance and long-tail problems inherent in EC segmentation, thereby further enhancing segmentation performance.

The brain's vascular network dynamically affects its development and core functions. It rapidly responds to abnormal conditions by adjusting properties of the network, aiding stabilization and regulation of brain activities. Tracking prominent arterial changes has clear clinical and surgical advantages. However, the arterial network functions as a system; thus, local changes may imply global compensatory effects that could impact the dynamic progression of a disease. We developed automated personalized system-level analysis methods of the compensatory arterial changes and mean blood flow behavior from a patient's clinical images. By applying our approach to data from a patient with aggressive brain cancer compared with healthy individuals, we found unique spatiotemporal patterns of the arterial network that could assist in predicting the evolution of glioblastoma over time. Our personalized approach provides a valuable analysis tool that could augment current clinical assessments of the progression of glioblastoma and other neurological disorders affecting the brain.

State-of-the-art exoskeletons are typically limited by low control bandwidth and small range stiffness of actuators which are based on high gear ratios and elastic components (e.g., series elastic actuators). Furthermore, most exoskeletons are based on discrete gait phase detection and/or discrete stiffness control resulting in discontinuous torque profiles. To fill these two gaps, we developed a portable lightweight knee exoskeleton using quasi-direct drive (QDD) actuation that provides 14 Nm torque (36.8% biological joint moment for overground walking). This paper presents 1) stiffness modeling of torque-controlled QDD exoskeletons and 2) stiffness-based continuous torque controller that estimates knee joint moment in real-time. Experimental tests found the exoskeleton had high bandwidth of stiffness control (16 Hz under 100 Nm/rad) and high torque tracking accuracy with 0.34 Nm Root Mean Square (RMS) error (6.22%) across 0-350 Nm/rad large range stiffness. The continuous controller was able to estimate knee moments accurately and smoothly for three walking speeds and their transitions. Experimental results with 8 able-bodied subjects demonstrated that our exoskeleton was able to reduce the muscle activities of all 8 measured knee and ankle muscles by 8.60%-15.22% relative to unpowered condition, and two knee flexors and one ankle plantar flexor by 1.92%-10.24% relative to baseline (no exoskeleton) condition.