Tiancheng He

Online 4-D CT estimation for patient-specific respiratory motion based on real-time breathing signals

In image-guided lung intervention, the electromagnetic (EM) tracked needle can be visualized in a pre-procedural CT by registering the EM tracking and the CT coordinate systems. However, there exist discrepancies between the static pre-procedural CT and the patient due to respiratory motion. This paper proposes an online 4-D CT estimation approach to patient-specific respiratory motion compensation. First, the motion patterns between 4-D CT data and respiratory signals such as fiducials from a number of patients are trained in a template space after image registration. These motion patterns can be used to estimate the patient-specific serial CTs from a static 3-D CT and the real-time respiratory signals of that patient, who do not generally take 4-D CTs. Specifically, the respiratory lung field motion vectors are projected onto the Kernel Principal Component Analysis (K-PCA) space, and a motion estimation model is constructed to estimate the lung field motion from the fiducial motion using the ridge regression method based on the least squares support vector machine (LS-SVM). The algorithm can be performed onsite prior to the intervention to generate the serial CT images according to the respiratory signals in advance, and the estimated CTs can be visualized in real-time during the intervention. In experiments, we evaluated the algorithm using leave-one-out strategy on 30 4-D CT data, and the results showed that the average errors of the lung field surfaces are 1.63 mm.

A minimally invasive multimodality image-guided (MIMIG) system for peripheral lung cancer intervention and diagnosis.

BACKGROUND Lung cancer is the leading cause of cancer-related death in the United States, with more than half of the cancers are located peripherally. Computed tomography (CT) has been utilized in the last decade to detect early peripheral lung cancer. However, due to the high false diagnosis rate of CT, further biopsy is often necessary to confirm cancerous cases. This renders intervention for peripheral lung nodules (especially for small peripheral lung cancer) difficult and time-consuming, and it is highly desirable to develop new, on-the-spot earlier lung cancer diagnosis and treatment strategies. PURPOSE The objective of this study is to develop a minimally invasive multimodality image-guided (MIMIG) intervention system to detect lesions, confirm small peripheral lung cancer, and potentially guide on-the-spot treatment at an early stage. Accurate image guidance and real-time optical imaging of nodules are thus the key techniques to be explored in this work. METHODS The MIMIG system uses CT images and electromagnetic (EM) tracking to help interventional radiologists target the lesion efficiently. After targeting the lesion, a fiber-optic probe coupled with optical molecular imaging contrast agents is used to confirm the existence of cancerous tissues on-site at microscopic resolution. Using the software developed, pulmonary vessels, airways, and nodules can be segmented and visualized for surgical planning; the segmented results are then transformed onto the intra-procedural CT for interventional guidance using EM tracking. Endomicroscopy through a fiber-optic probe is then performed to visualize tumor tissues. Experiments using IntegriSense 680 fluorescent contrast agent labeling αvβ3 integrin were carried out for rabbit lung cancer models. Confirmed cancers could then be treated on-the-spot using radio-frequency ablation (RFA). RESULTS The prototype system is evaluated using the rabbit VX2 lung cancer model to evaluate the targeting accuracy, guidance efficiency, and performance of molecular imaging. Using this system, we achieved an average targeting accuracy of 3.04 mm, and the IntegriSense signals within the VX2 tumors were found to be at least two-fold higher than those of normal tissues. The results demonstrate great potential for applying the system in human trials in the future if an optical molecular imaging agent is approved by the Food and Drug Administration (FDA). CONCLUSIONS The MIMIG system was developed for on-the-spot interventional diagnosis of peripheral lung tumors by combining image-guidance and molecular imaging. The system can be potentially applied to human trials on diagnosing and treating earlier stage lung cancer. For current clinical applications, where a biopsy is unavoidable, the MIMIG system without contrast agents could be used for biopsy guidance to improve the accuracy and efficiency.

Three-dimensional dendritic spine detection based on minimal cross-sectional curvature

The morphological changes of dendritic spines play a role in adaptive changes of the bram, and dysregulation of them is implicated in various neurological disorders. A systematic analysis tool for the morphology of dendritic spines, acquired via high resolution optical microscopy, should contribute to our understanding of neurophysiology as well as neuropathology. However, large numbers of high-resolution dendritic spine images make manual labeling extremely difficult and laborious. This is particularly challenging for the slender and curved dendritic spines frequently found in the striatum, dysregulation of which is highly implicated in neuropsychiatrie diseases, especially when compared to the analysis of relatively small and bulbous dendritic spines in other bram areas, including the hippocampus and cortex, synaptic dysregulation and neurodegeneration, of which are implicated in Alzheimers disease. In this paper, we present cross-sectional curvature, a new feature, for detecting the tip of dendritic spines in stnatal neurons, the spines of which are then segmented using the region growing strategy. Comparative results showed that our method was more accurate and superior to the existing ones.

A minimally invasive multimodality image-guided (MIMIG) molecular imaging system for peripheral lung cancer intervention and diagnosis

The once-promising computed tomography (CT) lung cancer screening appears to result in high false positive rates. To tackle the common difficulties in diagnosing small lung cancer at an early stage, we developed a minimally invasive multimodality image-guided (MIMIG) interventional system for early detection and treatment of peripheral lung cancer. The system consists of new CT image segmentation for surgical planning, intervention guidance for targeting, and molecular imaging for diagnosis. Using advanced image segmentation technique the pulmonary vessels, airways, as well as nodules can be better visualized for surgical planning. These segmented results are then transformed onto the intra-procedural CT for interventional guidance using electromagnetic (EM) tracking. Diagnosis can be achieved at microscopic resolution using a fiber-optic microendoscopy. The system can also be used for fine needle aspiration biopsy to improve the accuracy and efficiency. Confirmed cancer could then be treated on-the-spot using radio-frequency ablation (RFA). The experiments on rabbits with VX2 lung cancer model show both accuracy and efficiency in localization and detecting lung cancer, as well as promising molecular imaging tumor detection.

Multimodality Image-Guided Lung Intervention Systems

Medical imaging provides internal anatomical information of the human body to facilitate minimally invasive interventional procedures. Ideally, image-guided intervention requires both the device tracking and imaging to be performed in real time, and recent development of medical imaging and device-tracking techniques makes it possible to visualize both devices and patients’ anatomy during intervention. However, when real-time imaging is not applicable, patient motion tracking and image registration or motion compensation play important roles in generating more realistic image roadmaps for the guidance. Following discussion of the traditional techniques for multimodality image-guided intervention, this chapter focuses on how to integrate device tracking with multimodality imaging and introduces data fusion and dynamic image guidance in the context of image-guided bronchoscopy and percutaneous lung cancer intervention. It is expected that with advanced sensors and dynamic image modeling, more accurate real-time estimation about the interventional roadmap and more efficient, accurate, and safer intervention procedures can be achieved.

Helical Mode Lung 4D-CT Reconstruction Using Bayesian Model

4D computed tomography (CT) has been widely used for treatment planning of thoracic and abdominal cancer radiotherapy. Current 4D-CT lung image reconstruction methods rely on respiratory gating to rearrange the large number of axial images into different phases, which may be subject to external surrogate errors due to poor reproducibility of breathing cycles. New image-matching-based reconstruction works better for the cine mode of 4D-CT acquisition than the helical mode because the table position of each axial image is different in helical mode and image matching might suffer from bigger errors. In helical mode, not only the phases but also the un-uniform table positions of images need to be considered. We propose a Bayesian method for automated 4D-CT lung image reconstruction in helical mode 4D scans. Each axial image is assigned to a respiratory phase based on the Bayesian framework that ensures spatial and temporal smoothness of surfaces of anatomical structures. Iterative optimization is used to reconstruct a series of 3D-CT images for subjects undergoing 4D scans. In experiments, we compared visually and quantitatively the results of the proposed Bayesian 4D-CT reconstruction algorithm with the respiratory surrogate and the image matching-based method. The results showed that the proposed algorithm yielded better 4D-CT for helical scans.

A motion correction algorithm for microendoscope video computing in image-guided intervention

In multimodality image-guided intervention for cancer diagnosis, a needle with cannula is first punctured using CT or MRI -guided system to target the tumor, then microendoscopy can be performed using an optical fiber through the same cannula. With real-time optical imaging, the operator can directly determine the malignance of the tumor or perform fine needle aspiration biopsy for further diagnosis. During this operation, stable microendoscopy image series are needed to quantify the tissue properties, but they are often affected by respiratory and heart systole motion even when the interventional probe is held steadily. This paper proposes a microendoscopy motion correction (MMC) algorithm using normalized mutual information (NMI)-based registration and a nonlinear system to model the longitudinal global transformations. Cubature Kalman filter is thus used to solve the underlying longitudinal transformations, which yields more stable and robust motion estimation. After global motion correction, longitudinal deformations among the image sequences are calculated to further refine the local tissue motion. Experimental results showed that compared to global and deformable image registrations, MMC yields more accurate alignment results for both simulated and real data.

Transductive maximum margin classification of ADHD using resting state fMRI

Resting-state functional magnetic resonance imaging (rs-fMRI) provides key neural imaging characteristics for quantitative assessment and better understanding of the mechanisms of attention deficit hyperactivity disorder (ADHD). Recent multivariate analysis studies showed that functional connectivity (FC) could be used to classify ADHD from normal controls at the individual level. However, there may not be sufficient large numbers of labeled training samples for a hand-on classifier especially for disease classification. In this paper, we propose a transductive maximum margin classification (TMMC) method that uses the available unlabeled data in the learning process. On one hand, the maximum margin classification (MMC) criterion is used to maximize the class margin for the labeled data; on the other hand, a smoothness constraint is imposed on both labeled and unlabeled data projection so that similar samples tend to share the same label. To evaluate the performance of TMMC, experiments on a benchmark cohort from the ADHD-200 competition were performed. The results show that TMMC can improve the performance of ADHD classification using rs-fMRI by involving unlabeled samples, even for small number of labeled training data.

Reconstruction of four-dimensional computed tomography lung images by applying spatial and temporal anatomical constraints using a Bayesian model

Abstract. Current four-dimensional computed tomography (4-D CT) lung image reconstruction methods rely on respiratory gating, such as surrogate, to sort the large number of axial images captured during multiple breathing cycles into serial three-dimensional CT images of different respiratory phases. Such sorting methods may be subject to external surrogate signal noises due to poor reproducibility of breathing cycles. New image-matching-based reconstruction algorithms refine the 4-D CT reconstruction by matching neighboring image slices, and they generally work better for the cine mode of 4-D CT acquisition than the helical mode due to different table positions of axial images in the helical mode. We propose a Bayesian model (BM) based automated 4-D CT lung image reconstruction for helical mode scans. BM allows for applying new spatial and temporal anatomical constraints in the optimization procedure. Using an iterative optimization procedure, each axial image is assigned to a respiratory phase to make sure the anatomical structures are spatially and temporally smooth based on the BM framework. In experiments, we visually and quantitatively compared the results of the proposed BM-based 4-D CT reconstruction with the respiratory surrogate and the normalized cross-correlation based image matching method using both simulated and actual 4-D patient scans. The results indicated that the proposed algorithm yielded more accurate reconstruction and fewer artifacts in the 4-D CT image series.

Nonlinear motion compensation using cubature Kalman filter for in vivo fluorescence microendoscopy in peripheral lung cancer intervention

Abstract. Fluorescence microendoscopy can potentially be a powerful modality in minimally invasive percutaneous intervention for cancer diagnosis because it has an exceptional ability to provide micron-scale resolution images in tissues inaccessible to traditional microscopy. After targeting the tumor with guidance by macroscopic images such as computed tomorgraphy or magnetic resonance imaging, fluorescence microendoscopy can help select the biopsy spots or perform an on-site molecular imaging diagnosis. However, one challenge of this technique for percutaneous lung intervention is that the respiratory and hemokinesis motion often renders instability of the sequential image visualization and results in inaccurate quantitative measurement. Motion correction on such serial microscopy image sequences is, therefore, an important post-processing step. We propose a nonlinear motion compensation algorithm using a cubature Kalman filter (NMC-CKF) to correct these periodic spatial and intensity changes, and validate the algorithm using preclinical imaging experiments. The algorithm integrates a longitudinal nonlinear system model using the CKF in the serial image registration algorithm for robust estimation of the longitudinal movements. Experiments were carried out using simulated and real microendoscopy videos captured from the CellVizio 660 system in rabbit VX2 cancer intervention. The results show that the NMC-CKF algorithm yields more robust and accurate alignment results.