Registration-Guided Deep Learning Image Segmentation for Cone Beam CT-based Online Adaptive Radiotherapy
Authors:
Lin Ma,
Weicheng Chi,
Howard E. Morgan,
Mu-Han Lin,
Mingli Chen,
David Sher,
Dominic Moon,
Dat T. Vo,
Vladimir Avkshtol,
Weiguo Lu,
Xuejun Gu
Abstract:
Adaptive radiotherapy (ART), especially online ART, effectively accounts for positioning errors and anatomical changes. One key component of online ART is accurately and efficiently delineating organs at risk (OARs) and targets on online images, such as CBCT, to meet the online demands of plan evaluation and adaptation. Deep learning (DL)-based automatic segmentation has gained great success in se…
▽ More
Adaptive radiotherapy (ART), especially online ART, effectively accounts for positioning errors and anatomical changes. One key component of online ART is accurately and efficiently delineating organs at risk (OARs) and targets on online images, such as CBCT, to meet the online demands of plan evaluation and adaptation. Deep learning (DL)-based automatic segmentation has gained great success in segmenting planning CT, but its applications to CBCT yielded inferior results due to the low image quality and limited available contour labels for training. To overcome these obstacles to online CBCT segmentation, we propose a registration-guided DL (RgDL) segmentation framework that integrates image registration algorithms and DL segmentation models. The registration algorithm generates initial contours, which were used as guidance by DL model to obtain accurate final segmentations. We had two implementations the proposed framework--Rig-RgDL (Rig for rigid body) and Def-RgDL (Def for deformable)--with rigid body (RB) registration or deformable image registration (DIR) as the registration algorithm respectively and U-Net as DL model architecture. The two implementations of RgDL framework were trained and evaluated on seven OARs in an institutional clinical Head and Neck (HN) dataset. Compared to the baseline approaches using the registration or the DL alone, RgDL achieved more accurate segmentation, as measured by higher mean Dice similarity coefficients (DSC) and other distance-based metrics. Rig-RgDL achieved a DSC of 84.5% on seven OARs on average, higher than RB or DL alone by 4.5% and 4.7%. The DSC of Def-RgDL is 86.5%, higher than DIR or DL alone by 2.4% and 6.7%. The inference time took by the DL model to generate final segmentations of seven OARs is less than one second in RgDL. The resulting segmentation accuracy and efficiency show the promise of applying RgDL framework for online ART.
△ Less
Submitted 19 August, 2021;
originally announced August 2021.
Multimetric Event-driven System for Long-Term Wireless Sensor Operation in SHM Application
Authors:
Muhammad Zohaib Sarwar,
Muhammad Rakeh Saleem,
Jong-Woong Park,
Do-Soo Moon,
Dong Joo Kim
Abstract:
Wireless sensor networks (WSNs) are promising solutions for large infrastructure monitoring because of their ease of installation, computing and communication capability, and cost-effectiveness. Long-term structural health monitoring (SHM), however, is still a challenge because it requires continuous data acquisition for the detection of random events such as earthquakes and structural collapse. T…
▽ More
Wireless sensor networks (WSNs) are promising solutions for large infrastructure monitoring because of their ease of installation, computing and communication capability, and cost-effectiveness. Long-term structural health monitoring (SHM), however, is still a challenge because it requires continuous data acquisition for the detection of random events such as earthquakes and structural collapse. To achieve long-term operation, it is necessary to reduce the power consumption of sensor nodes designed to capture random events and, thus, enhance structural safety. In this paper, we present an event-based sensing system design based on an ultra-low-power microcontroller with a programmable event-detection mechanism to allow continuous monitoring; the device is triggered by vibration, strain, or a timer and has a programmed threshold, resulting in ultra-low-power consumption of the sensor node. Furthermore, the proposed system can be easily reconfigured to any existing wireless sensor platform to enable ultra-low power operation. For validation, the proposed system was integrated with a commercial wireless platform to allow strain, acceleration, and time-based triggering with programmed thresholds and current consumptions of 7.43 and 0.85 mA in active and inactive modes, respectively.
△ Less
Submitted 17 October, 2019;
originally announced October 2019.
