Auto-segmentation of thoraco-abdominal organs in pediatric dynamic MRI

Purpose

Analysis of the abnormal motion of thoraco-abdominal organs in respiratory disorders such as the Thoracic Insufficiency Syndrome (TIS) and scoliosis such as adolescent idiopathic scoliosis (AIS) or early onset scoliosis (EOS) can lead to better surgical plans. We can use healthy subjects to find out the normal architecture and motion of a rib cage and associated organs and attempt to modify the patient's deformed anatomy to match to it. Dynamic magnetic resonance imaging (dMRI) is a practical and preferred imaging modality for capturing dynamic images of healthy pediatric subjects. In this paper, we propose an auto-segmentation set-up for the lungs, kidneys, liver, spleen, and thoraco-abdominal skin in these dMRI images which have their own challenges such as poor contrast, image non-standardness, and similarity in texture amongst gas, bone, and connective tissue at several inter-object interfaces.

Methods

The segmentation set-up has been implemented in two steps: recognition and delineation using two deep neural network (DL) architectures (say DL-R and DL-D) for the recognition step and delineation step, respectively. The encoder-decoder framework in DL-D utilizes features at four different resolution levels to counter the challenges involved in the segmentation. We have evaluated on dMRI sagittal acquisitions of 189 (near-)normal subjects. The spatial resolution in all dMRI acquisitions is 1.46 mm in a sagittal slice and 6.00 mm between sagittal slices. We utilized images of 89 (10) subjects at end inspiration for training (validation). For testing we experimented with three scenarios: utilizing (1) the images of 90 (=189-89-10) different (remaining) subjects at end inspiration for testing, (2) the images of the aforementioned 90 subjects at end expiration for testing, and (3) the images of the aforesaid 99 (=89+10) subjects but at end expiration for testing. In some situations, we can take advantage of already available ground truth (GT) of a subject at a particular respiratory phase to automatically segment the object in the image of the same subject at a different respiratory phase and then refining the segmentation to create the final GT. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assume to have the ground truth of the test subjects at end expiration for scenario (1), end inspiration for (2), and end inspiration for (3).

Results

Amongst these three scenarios of testing, for the DL-R, we achieve a best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen and 1.5 voxels for the liver and the thoraco- abdominal skin. The standard deviation (SD) of LE is about 1 or 2 voxels. For the delineation approach, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for the thoraco-abdominal skin. The SD of DC is lower for the lungs, liver, and the thoraco-abdominal skin, and slightly higher for the spleen and kidneys.

Conclusions

Motivated by applications in surgical planning for disorders such as TIS, AIS, and EOS, we have shown an auto-segmentation system for thoraco-abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non-standardness quite well. We are in the process of testing its effectiveness on TIS patient dMRI data.

Semiautomatic assessment of respiratory dynamics using cine MRI in chronic obstructive pulmonary disease

Purpose

The quantitative assessment of impaired lung motions and their association with the clinical characteristics of COPD patients is challenging. The aim of this study was to measure respiratory kinetics, including asynchronous movements, and to analyze the relationship between lung area and other clinical parameters.

Materials and methods

This study enrolled 10 normal control participants and 21 COPD patients who underwent dynamic MRI and pulmonary function testing (PFT). The imaging program was implemented using MATLAB®. Each lung area was detected semi-automatically on a coronal image (imaging level at the aortic valve) from the inspiratory phase to the expiratory phase. The Dice index of the manual measurements was calculated, with the relationship between lung area ratio and other clinical parameters, including PFTs then evaluated. The asynchronous movements of the diaphragm were also evaluated using a sagittal image.

Results

The Dice index for the lung region using the manual and semi-automatic extraction methods was high (Dice index = 0.97 ± 0.03). A significant correlation was observed between the time corrected lung area ratio and percentage of forced expiratory volume in 1 s (FEV₁%pred) and residual volume percentage (RV%pred) (r = −0.54, p = 0.01, r = 0.50, p = 0.03, respectively). The correlation coefficient between each point of the diaphragm in the group with visible see-saw like movements was significantly lower than that in the group without see-saw like movements (value = −0.36 vs 0.95, p = 0.001).

Conclusion

Semi-automated extraction of lung area from Cine MRI might be useful for detecting impaired respiratory kinetics in patients with COPD.

[Semi-automated Segmentation of Lungs Using the k-means Method in Cine MRI]

Dynamic magnetic resonance imaging (MRI) provides essential information on the respiratory kinetics in chronic obstructive pulmonary disease (COPD), such as impaired diaphragm and chest wall motions. The purpose of this study was to develop the semi-automated segmentation program of lungs using cine MRI. We enrolled five control participants and five patients with COPD who underwent cine MRI. The coronal balanced FFE images from each subject were used. The procedures were as follows: First, the maximum inspiratory image was selected from the time-sequential series, and the lung area was manually segmented, which was used for a mask image. Second, both mask image and cine image were accumulated to create a weighted cine image. Lung areas were segmented using the k-means method. Finally, lungs were detected as contiguous image regions with similar signal values using the flood-fill technique. We evaluated the correlation coefficients between the lung area segmented by the semi-automated method and those segmented by a pulmonologist. The correlation coefficients between the semi-automated method and the manual segmentations were excellent (r=0.99, p<0.001). The Dice index was also perfect (0.97). The best number of clusters in the k-means method was 8. These results suggested that the new segmentation method can appropriately extract lungs and help analyze respiratory dynamics in patients with COPD.

Interactive medical image segmentation via a point-based interaction

Due to low tissue contrast, irregular shape, and large location variance, segmenting the objects from different medical imaging modalities (e.g., CT, MR) is considered as an important yet challenging task. In this paper, a novel method is presented for interactive medical image segmentation with the following merits. (1) Its design is fundamentally different from previous pure patch-based and image-based segmentation methods. It is observed that during delineation, the physician repeatedly check the intensity from area inside-object to outside-object to determine the boundary, which indicates that comparison in an inside-out manner is extremely important. Thus, the method innovatively models the segmentation task as learning the representation of bi-directional sequential patches, starting from (or ending in) the given central point of the object. This can be realized by the proposed ConvRNN network embedded with a gated memory propagation unit. (2) Unlike previous interactive methods (requiring bounding box or seed points), the proposed method only asks the physician to merely click on the rough central point of the object before segmentation, which could simultaneously enhance the performance and reduce the segmentation time. (3) The method is utilized in a multi-level framework for better performance. It has been systematically evaluated in three different segmentation tasks, including CT kidney tumor, MR prostate, and PROMISE12 challenge, showing promising results compared with state-of-the-art methods.

Holistic segmentation of the lung in cine MRI

Duchenne muscular dystrophy (DMD) is a childhood-onset neuromuscular disease that results in the degeneration of muscle, starting in the extremities, before progressing to more vital areas, such as the lungs. Respiratory failure and pneumonia due to respiratory muscle weakness lead to hospitalization and early mortality. However, tracking the disease in this region can be difficult, as current methods are based on breathing tests and are incapable of distinguishing between muscle involvements. Cine MRI scans give insight into respiratory muscle movements, but the images suffer due to low spatial resolution and poor signal-to-noise ratio. Thus, a robust lung segmentation method is required for accurate analysis of the lung and respiratory muscle movement. We deployed a deep learning approach that utilizes sequence-specific prior information to assist the segmentation of lung in cine MRI. More specifically, we adopt a holistically nested network to conduct image-to-image holistic training and prediction. One frame of the cine MRI is used in the training and applied to the remainder of the sequence ([Formula: see text] frames). We applied this method to cine MRIs of the lung in the axial, sagittal, and coronal planes. Characteristic lung motion patterns during the breathing cycle were then derived from the segmentations and used for diagnosis. Our data set consisted of 31 young boys, age [Formula: see text] years, 15 of whom suffered from DMD. The remaining 16 subjects were age-matched healthy volunteers. For validation, slices from inspiratory and expiratory cycles were manually segmented and compared with results obtained from our method. The Dice similarity coefficient for the deep learning-based method was [Formula: see text] for the sagittal view, [Formula: see text] for the axial view, and [Formula: see text] for the coronal view. The holistic neural network approach was compared with an approach using Demon's registration and showed superior performance. These results suggest that the deep learning-based method reliably and accurately segments the lung across the breathing cycle.

Deformable Registration-Based Super-resolution for Isotropic Reconstruction of 4-D MRI Volumes

Multi-plane super-resolution (SR) has been widely employed for resolution improvement of MR images. However, this has mostly been limited to MRI acquisitions with rigid motion. In cases of non-rigid motion, volumes are usually pre-registered using deformable registration methods before SR reconstruction. The pre-registered images are then used as input for the SR reconstruction. Since deformable registration involves smoothening of the inputs, using pre-registered inputs could lead to loss in information in SR reconstructions. Additionally, any registration errors present in pre-registered inputs could propagate throughout SR reconstructions leading to error accumulation. To address these limitations, in this study, we propose a deformable registration-based super-resolution reconstruction (DIRSR) reconstruction, which handles deformable registration as part of super-resolution. This approach has been demonstrated using 12 synthetic 4-D MRI lung datasets created using single plane (coronal) datasets of six patients and multi-plane (coronal and axial) 4-D lung MRI dataset of one patient. From our evaluation, DIRSR reconstructions are sharper and well aligned compared to reconstructions using SR of pre-registered inputs and rigid-registration SR. MSE, SNR and SSIM evaluations also indicate better reconstruction quality from DIRSR compared to reconstructions from SR of pre-registered inputs (p-value less than 0.0001). In conclusion, we found superior isotropic reconstructions of 4-D MR datasets from DIRSR reconstructions, which could benefit volumetric MR analyses.

Interactive iterative relative fuzzy connectedness lung segmentation on thoracic 4D dynamic MR images

Lung delineation via dynamic 4D thoracic magnetic resonance imaging (MRI) is necessary for quantitative image analysis for studying pediatric respiratory diseases such as thoracic insufficiency syndrome (TIS). This task is very challenging because of the often-extreme malformations of the thorax in TIS, lack of signal from bone and connective tissues resulting in inadequate image quality, abnormal thoracic dynamics, and the inability of the patients to cooperate with the protocol needed to get good quality images. We propose an interactive fuzzy connectedness approach as a potential practical solution to this difficult problem. Manual segmentation is too labor intensive especially due to the 4D nature of the data and can lead to low repeatability of the segmentation results. Registration-based approaches are somewhat inefficient and may produce inaccurate results due to accumulated registration errors and inadequate boundary information. The proposed approach works in a manner resembling the Iterative Livewire tool but uses iterative relative fuzzy connectedness (IRFC) as the delineation engine. Seeds needed by IRFC are set manually and are propagated from slice-to-slice, decreasing the needed human labor, and then a fuzzy connectedness map is automatically calculated almost instantaneously. If the segmentation is acceptable, the user selects "next" slice. Otherwise, the seeds are refined and the process continues. Although human interaction is needed, an advantage of the method is the high level of efficient user-control on the process and non-necessity to refine the results. Dynamic MRI sequences from 5 pediatric TIS patients involving 39 3D spatial volumes are used to evaluate the proposed approach. The method is compared to two other IRFC strategies with a higher level of automation. The proposed method yields an overall true positive and false positive volume fraction of 0.91 and 0.03, respectively, and Hausdorff boundary distance of 2 mm.

Quantification of Diaphragm Mechanics in Pompe Disease Using Dynamic 3D MRI

BACKGROUND

Diaphragm weakness is the main reason for respiratory dysfunction in patients with Pompe disease, a progressive metabolic myopathy affecting respiratory and limb-girdle muscles. Since respiratory failure is the major cause of death among adult patients, early identification of respiratory muscle involvement is necessary to initiate treatment in time and possibly prevent irreversible damage. In this paper we investigate the suitability of dynamic MR imaging in combination with state-of-the-art image analysis methods to assess respiratory muscle weakness.

METHODS

The proposed methodology relies on image registration and lung surface extraction to quantify lung kinematics during breathing. This allows for the extraction of geometry and motion features of the lung that characterize the independent contribution of the diaphragm and the thoracic muscles to the respiratory cycle.

RESULTS

Results in 16 3D+t MRI scans (10 Pompe patients and 6 controls) of a slow expiratory maneuver show that kinematic analysis from dynamic 3D images reveals important additional information about diaphragm mechanics and respiratory muscle involvement when compared to conventional pulmonary function tests. Pompe patients with severely reduced pulmonary function showed severe diaphragm weakness presented by minimal motion of the diaphragm. In patients with moderately reduced pulmonary function, cranial displacement of posterior diaphragm parts was reduced and the diaphragm dome was oriented more horizontally at full inspiration compared to healthy controls.

CONCLUSION

Dynamic 3D MRI provides data for analyzing the contribution of both diaphragm and thoracic muscles independently. The proposed image analysis method has the potential to detect less severe diaphragm weakness and could thus be used to determine the optimal start of treatment in adult patients with Pompe disease in prospect of increased treatment response.

Minimally interactive segmentation of 4D dynamic upper airway MR images via fuzzy connectedness

PURPOSE

There are several disease conditions that lead to upper airway restrictive disorders. In the study of these conditions, it is important to take into account the dynamic nature of the upper airway. Currently, dynamic magnetic resonance imaging is the modality of choice for studying these diseases. Unfortunately, the contrast resolution obtainable in the images poses many challenges for an effective segmentation of the upper airway structures. No viable methods have been developed to date to solve this problem. In this paper, the authors demonstrate a practical solution by employing an iterative relative fuzzy connectedness delineation algorithm as a tool.

METHODS

3D dynamic images were collected at ten equally spaced instances over the respiratory cycle (i.e., 4D) in 20 female subjects with obstructive sleep apnea syndrome. The proposed segmentation approach consists of the following steps. First, image background nonuniformities are corrected which is then followed by a process to correct for the nonstandardness of MR image intensities. Next, standardized image intensity statistics are gathered for the nasopharynx and oropharynx portions of the upper airway as well as the surrounding soft tissue structures including air outside the body region, hard palate, soft palate, tongue, and other soft structures around the airway including tonsils (left and right) and adenoid. The affinity functions needed for fuzzy connectedness computation are derived based on these tissue intensity statistics. In the next step, seeds for fuzzy connectedness computation are specified for the airway and the background tissue components. Seed specification is needed in only the 3D image corresponding to the first time instance of the 4D volume; from this information, the 3D volume corresponding to the first time point is segmented. Seeds are automatically generated for the next time point from the segmentation of the 3D volume corresponding to the previous time point, and the process continues and runs without human interaction and completes in 10 s for segmenting the airway structure in the whole 4D volume.

RESULTS

Qualitative evaluations performed to examine smoothness and continuity of motions of the entire upper airway as well as its transverse sections at critical anatomic locations indicate that the segmentations are consistent. Quantitative evaluations of the separate 200 3D volumes and the 20 4D volumes yielded true positive and false positive volume fractions around 95% and 0.1%, respectively, and mean boundary placement errors under 0.5 mm. The method is robust to variations in the subjective action of seed specification. Compared with a segmentation approach based on a registration technique to propagate segmentations, the proposed method is more efficient, accurate, and less prone to error propagation from one respiratory time point to the next.

CONCLUSIONS

The proposed method is the first demonstration of a viable and practical approach for segmenting the upper airway structures in dynamic MR images. Compared to registration-based methods, it effectively reduces error propagation and consequently achieves not only more accurate segmentations but also more consistent motion representation in the segmentations. The method is practical, requiring minimal user interaction and computational time.

A hybrid approach for fusing 4D-MRI temporal information with 3D-CT for the study of lung and lung tumor motion

PURPOSE

Accurate visualization of lung motion is important in many clinical applications, such as radiotherapy of lung cancer. Advancement in imaging modalities [e.g., computed tomography (CT) and MRI] has allowed dynamic imaging of lung and lung tumor motion. However, each imaging modality has its advantages and disadvantages. The study presented in this paper aims at generating synthetic 4D-CT dataset for lung cancer patients by combining both continuous three-dimensional (3D) motion captured by 4D-MRI and the high spatial resolution captured by CT using the authors' proposed approach.

METHODS

A novel hybrid approach based on deformable image registration (DIR) and finite element method simulation was developed to fuse a static 3D-CT volume (acquired under breath-hold) and the 3D motion information extracted from 4D-MRI dataset, creating a synthetic 4D-CT dataset.

RESULTS

The study focuses on imaging of lung and lung tumor. Comparing the synthetic 4D-CT dataset with the acquired 4D-CT dataset of six lung cancer patients based on 420 landmarks, accurate results (average error <2 mm) were achieved using the authors' proposed approach. Their hybrid approach achieved a 40% error reduction (based on landmarks assessment) over using only DIR techniques.

CONCLUSIONS

The synthetic 4D-CT dataset generated has high spatial resolution, has excellent lung details, and is able to show movement of lung and lung tumor over multiple breathing cycles.