Incidence of lung nodules has surged due to improved lung cancer screening programs. Although localized microwave ablation (MWA) has been shown to be a minimally invasive safe, effective, cost and time efficient treatment for non-surgical candidates, it remains an underutilized curative modality for early-stage lung cancer patients, due in part to professed superiority of radiation for local tumor control, or prevention of local tumor progression (LTP). Identification of lesions that may be more amenable to effective MWA treatment may lead to improved outcomes. To aid physicians in optimizing patient selection, we developed a machine learning model to predict LTP after MWA treatment. Our model utilizes specialized 3D three-channel data: pre-ablation CT data (channel 1), post-ablation CT data depicting the resulting ablation zone (channel 2) and overlapping data of the tumor and ablation zone (channel 3). By spatially registering pre- and post-ablation CTs, we establish a clear spatial relationship between the tumor and ablation zone. Our neural network, trained on 55 MWA-treated lung-cancer patients, achieved a C-statistic (AUC) of 0.849 compared to 0.78 of prior approaches in 5-fold cross-validation. Notably, this performance was achieved without incorporating tabular features such as cancer type or ablation margin, highlighting strengths of the specialized 3D three-channels images. Combined with our past work, where we demonstrated the potential for accurate prediction of ablation zone boundaries during procedure planning, our research presents promising preliminary results for assisting physicians in predicting LTP following localized MWA treatment. The ability to identify good responders to MWA may provide a tool for patient selection, enhance patient outcome, and expand the utilization of this safe, effective treatment option.
The use of percutaneous and bronchoscopic microwave ablation to treat both primary and secondary lung tumors has been growing recently. These ablation systems are typically accompanied by an ablation planning system to optimize the treatment outcome by ensuring adequate margin in the expected ablation zone during the planning phase. The planning system utilizes pre-operative CT scan to identify the tumor and recommend microwave probe position. Radiomics is a process of converting medical images into higher-dimensional data and subsequent mining of data to reveal underlying pathophysiology for enhancing clinical decision support making. Radiomics analysis have shown promises in capturing distinct tumor characteristics and predicting prognosis of the tumor. Here, we present a new method to predict microwave ablation zones by supplementing a bioheat transfer model of microwave tissue ablation with microwave sensitive radiomics features. We hypothesize that supplementing traditional bioheat transfer modeling with microwave sensitive radiomics features will generate a more accurate and personalized ablation prediction that will lead to better treatment outcome. Inputs to the bioheat transfer modeling approach include the geometry of the target tumor, physical characteristics of the tissue, and dimensions of the microwave ablation applicator. The radiomics algorithm extracts characteristics of the targeted tumor’s size and shape, as well as texture characteristics, from pre-operative CT images. We employed cascaded segmentation based on RetinaNet and U-Net to obtain a tumor’s size and shape. Then, a segmented tumor is employed for texture analysis through a set of regression convolutional neural networks. These tumor characteristics are employed as radiomics features for more accurate dose prediction and margin for microwave ablation treatment. We present the preliminary results of a study using images from clinical lung tumor cases to predict ablation treatment outcome, with patient-specific tissue biophysical properties based on radiomics features.
Microwave ablation (MWA) is an emerging minimally invasive treatment option for malignant lung tumors. Compared to other energy modalities, such as radiofrequency ablation, MWA offers the advantages of deeper penetration within high impedance tissues such as aerated lung, shorter treatment times, and less susceptibility to the cooling heat-sink effects of air and blood flow. Previous studies have demonstrated clinical use of MWA for treating lung tumors; however, these procedures have relied upon the percutaneous application of rigid microwave antennas. The objective of our work was to develop and characterize a novel flexible microwave applicator which could be integrated with a bronchoscopic imaging and software guidance platform to expand the use of MWA as a treatment option for small (< 2cm) pulmonary tumors. This applicator would allow physicians an even less invasive, immediate treatment option for lung tumors identified within the scope of current medical procedures. It may also improve applicator placement accuracy and increase efficacy while minimizing the risk of procedural complications. A 2D-axisymmetric coupled FEM electromagnetic-heat transfer model was implemented to characterize expected antenna radiation patterns, ablation size and shape, and optimize antenna design for lung tissue. A prototype device was fabricated and evaluated in ex vivo tissues to verify simulation results and serve as proof-of-concept. Additional experiments were conducted in an in vivo animal model to further characterize the proposed system.
Reliable transbronchial access of peripheral lung lesions is desirable for the diagnosis and potential treatment
of lung cancer. This procedure can be difficult, however, because accessory devices (e.g., needle or forceps)
cannot be reliably localized while deployed. We present a fluoroscopic image-guided intervention (IGI) system
for tracking such bronchoscopic accessories. Fluoroscopy, an imaging technology currently utilized by many
bronchoscopists, has a fundamental shortcoming - many lung lesions are invisible in its images. Our IGI
system aligns a digitally reconstructed radiograph (DRR) defined from a pre-operative computed tomography
(CT) scan with live fluoroscopic images. Radiopaque accessory devices are readily apparent in fluoroscopic video,
while lesions lacking a fluoroscopic signature but identifiable in the CT scan are superimposed in the scene. The
IGI system processing steps consist of: (1) calibrating the fluoroscopic imaging system; (2) registering the CT
anatomy with its depiction in the fluoroscopic scene; (3) optical tracking to continually update the DRR and
target positions as the fluoroscope is moved about the patient. The end result is a continuous correlation of the
DRR and projected targets with the anatomy depicted in the live fluoroscopic video feed. Because both targets
and bronchoscopic devices are readily apparent in arbitrary fluoroscopic orientations, multiplane guidance is
straightforward. The system tracks in real-time with no computational lag. We have measured a mean projected
tracking accuracy of 1.0 mm in a phantom and present results from an in vivo animal study.
C-arm fluoroscopy units provide continuously updating X-ray video images during surgical procedure. The
modality is widely adopted for its low cost, real-time imaging capabilities, and its ability to display radio-opaque
tools in the anatomy. It is, however, important to correct for fluoroscopic image distortion and estimate camera
parameters, such as focal length and camera center, for registration with 3D CT scans in fluoroscopic imageguided
procedures. This paper describes a method for C-arm calibration and evaluates its accuracy in multiple
C-arm units and in different viewing orientations. The proposed calibration method employs a commerciallyavailable
unit to track the C-arm and a calibration plate. The method estimates both the internal calibration
parameters and the transformation between the coordinate systems of tracker and C-arm. The method was
successfully tested on two C-arm units (GE OEC 9800 and GE OEC 9800 Plus) of different image intensifier
sizes and verified with a rigid airway phantom model. The mean distortion-model error was found to be 0.14
mm and 0.17 mm for the respective C-arms. The mean overall system reprojection error (which measures the
accuracy of predicting an image using tracker coordinates) was found to be 0.63 mm for the GE OEC 9800.
Stereotactic body-radiation therapy (SBRT) has gained acceptance in treating lung cancer. Localization of a
thoracic lesion is challenging as tumors can move significantly with breathing. Some SBRT systems compensate
for tumor motion with the intrafraction tracking of targets by two stereo fluoroscopy cameras. However, many
lung tumors lack a fluoroscopic signature and cannot be directly tracked. Small radiopaque fiducial markers,
acting as fluoroscopically visible surrogates, are instead implanted nearby. The spacing and configuration of
the fiducial markers is important to the success of the therapy as SBRT systems impose constraints on the
geometry of a fiducial-marker constellation. It is difficult even for experienced physicians mentally assess the
validity of a constellation a priori. To address this challenge, we present the first automated planning system
for bronchoscopic fiducial-marker placement. Fiducial-marker planning is posed as a constrained combinatoric
optimization problem. Constraints include requiring access from a navigable airway, having sufficient separation
in the fluoroscopic imaging planes to resolve each individual marker, and avoidance of major blood vessels.
Automated fiducial-marker planning takes approximately fifteen seconds, fitting within the clinical workflow.
The resulting locations are integrated into a virtual bronchoscopic planning system, which provides guidance to
each location during the implantation procedure. To date, we have retrospectively planned over 50 targets for
treatment, and have implanted markers according to the automated plan in one patient who then underwent
SBRT treatment. To our knowledge, this approach is the first to address automated bronchoscopic fiducialmarker
planning for SBRT.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.