An interactive medical image segmentation system based on the optimal management of regions of interest using topological medical knowledge

Clinical validation of 3D virtual modelling for laparoscopic complete mesocolic excision with central vascular ligation for proximal colon cancer

INTRODUCTION

Laparoscopic Complete Mesocolic Excision (CME) with Central Vascular Ligation (CVL) in colon cancer surgery has not been broadly adopted in part because of safety concerns. Pre-operative 3-D virtual modelling (3DVM) may help but needs validation.

METHODS

3DVM were routinely constructed from CT mesenteric angiograms (CTMA) using a commercial service (Visible Patient, Strasbourg, France) for consecutive patients during our CMECVL learning curve over three years. 3DVMs were independently checked versus CTMA and operative findings. CMECVL outcomes were compared versus other patients undergoing standard mesocolic excision (SME) surgery laparoscopically in the same hospital as control. Stakeholders were studied regarding 3DVM use and usefulness (including detail retention) versus CTMA and a physical 3D-printed model.

RESULTS

26 patients underwent 3DVM with intraoperative display during laparoscopic CMECVL within existing workflows. 3DVM accuracy was 96 % re arteriovenous variations at patient level versus CTMA/intraoperative findings including accessory middle colic artery identification in three patients. Twenty-two laparoscopic CMECVL with 3DVM cases were compared with 49 SME controls (age 69 ± 10 vs 70.9 ± 11 years, 55 % vs 53 % males). There were no intraoperative complications with CMECVL and similar 30-day postoperative morbidity (30 % vs 29 %), hospital stay (9 ± 3 vs 12 ± 13 days), 30-day readmission (6 % vs 4 %) and reoperation (0 % vs 4 %) rates. Intraoperative times were longer (215.7 ± 43.9 vs 156.9 ± 52.9 min, p=<0.01) but decreased significantly over time. 3DVM surveys (n = 98, 20 surgeons, 48 medical students, 30 patients/patient relatives) and comparative study revealed majority endorsement (90 %) and favour (87 %).

CONCLUSION

3DVM use was positively validated for laparoscopic CMECVL and valued by clinicians, students, and patients alike.

A Graph Based Image Interpretation Method Using A Priori Qualitative Inclusion and Photometric Relationships

This paper presents a method for recovering and identifying image regions from an initial oversegmentation using qualitative knowledge of its content. Compared to recent works favoring spatial information and quantitative techniques, our approach focuses on simple a priori qualitative inclusion and photometric relationships such as "region A is included in region B", "the intensity of region A is lower than the one of region B" or "regions A and B depict similar intensities" (photometric uncertainty). The proposed method is based on a two steps' inexact graph matching approach. The first step searches for the best subgraph isomorphism candidate between expected regions and a subset of regions resulting from the initial oversegmentation. Then, remaining segmented regions are progressively merged with appropriate already matched regions, while preserving the coherence with a priori declared relationships. Strengths and weaknesses of the method are studied on various images (grayscale and color), with various intial oversegmentation algorithms (k-means, meanshift, quickshift). Results show the potential of the method to recover, in a reasonable runtime, expected regions, a priori described in a qualitative manner. For further evaluation and comparison purposes, a Python opensource package implementing the method is provided, together with the specifically built experimental database.

Approach for sequential image interpretation using a priori binary perceptual topological and photometric knowledge and k-means-based segmentation

The proposed approach exploits a priori known qualitative inclusion and photometric relationships between image regions, represented by oriented graphs. Our work assumes a sequential image segmentation procedure where regions are progressively segmented and recognized by associating them with corresponding nodes in graphs related to the prior knowledge. The main contribution concerns the parameterization of the k-means clustering algorithm, to be used during the segmentation procedure, and the graph-matching-based identification of resulting clusters, corresponding to regions declared in graphs. The parameterization of k-means is based on known relationships as well as on regions that have been segmented and recognized at previous steps. Parameters are the region of interest within which k-means clustering is constrained, the number of clusters, and seeding constraints. Photometric relationships built from resulting clusters are matched with a priori known relationships to identify each cluster, this being formulated as an exact graph-matching problem. The potential of this approach is studied in four use cases involving real gray-scale and color images with dedicated sequential analysis procedures. Processing results are compared with those obtained without the proposed parameterization of k-means, as well as with some other clustering approaches. Results show the relevance of our approach, in particular in terms of segmentation accuracy, computation time, and seeding reliability.

Supervised recursive segmentation of volumetric CT images for 3D reconstruction of lung and vessel tree

Three dimensional reconstruction of lung and vessel tree has great significance to 3D observation and quantitative analysis for lung diseases. This paper presents non-sheltered 3D models of lung and vessel tree based on a supervised semi-3D lung tissues segmentation method. A recursive strategy based on geometric active contour is proposed instead of the "coarse-to-fine" framework in existing literature to extract lung tissues from the volumetric CT slices. In this model, the segmentation of the current slice is supervised by the result of the previous one slice due to the slight changes between adjacent slice of lung tissues. Through this mechanism, lung tissues in all the slices are segmented fast and accurately. The serious problems of left and right lungs fusion, caused by partial volume effects, and segmentation of pleural nodules can be settled meanwhile during the semi-3D process. The proposed scheme is evaluated by fifteen scans, from eight healthy participants and seven participants suffering from early-stage lung tumors. The results validate the good performance of the proposed method compared with the "coarse-to-fine" framework. The segmented datasets are utilized to reconstruct the non-sheltered 3D models of lung and vessel tree.

Design and development of a robotized system coupled to µCT imaging for intratumoral drug evaluation in a HCC mouse model

Hepatocellular carcinoma (HCC) is one of the most common cancer related deaths worldwide. One of the main challenges in cancer treatment is drug delivery to target cancer cells specifically. Preclinical evaluation of intratumoral drugs in orthotopic liver cancer mouse models is difficult, as percutaneous injection hardly can be precisely performed manually. In the present study we have characterized a hepatoma model developing a single tumor nodule by implantation of Hep55.1C cells in the liver of syngeneic C57BL/6J mice. Tumor evolution was followed up by µCT imaging, and at the histological and molecular levels. This orthotopic, poorly differentiated mouse HCC model expressing fibrosis, inflammation and cancer markers was used to assess the efficacy of drugs. We took advantage of the high precision of a previously developed robotized system for automated, image-guided intratumoral needle insertion, to administer every week in the tumor of the Hep55.1C mouse model. A significant tumor growth inhibition was observed using our robotized system, whereas manual intraperitoneal administration had no effect, by comparison to untreated control mice.

The therapeutic efficacy of I131-PSCA-mAb in orthotopic mouse models of prostate cancer

Background

Prostate stem cell antigen (PSCA) is upregulated in prostate cancer tissues. Here we aimed to study the therapeutic efficacy of a monoclonal antibody of PSCA-labeled I¹³¹ (I¹³¹-PSCA-mAb) in orthotopic mouse models of prostate cancer.

Methods

The proliferation, apoptosis and invasion abilities of PC-3 and LNCaP cells treated with I¹³¹-PSCA-mAb were measured by methyl thiazolyl tetrazolium assay, flow cytometry and transwell culture, respectively. The human prostate cancer models were established by orthotopic implantation of PC-3 and LNCaP cells in nude mice. I¹³¹-PSCA-mAb distribution and tumor cell apoptosis in the tumor-bearing nude mice were measured.

Results

The inhibitory and apoptosis rates of PC-3 and LNCaP cells treated with I¹³¹-PSCA-mAb reached a maximum of 84%, 80% and 50%, 46%, respectively, which were obviously higher than in the cells treated with I¹³¹-IgG or PSCA-mAb. The invaded number of PC-3 and LNCaP cells treated with I¹³¹-PSCA-mAbe was significantly reduced (P < 0.01) compared with the control group. The ratios of I¹³¹-PSCA-mAb in tumor to intramuscular I¹³¹-PSCA-mAb (T/NT) in tumor-bearing nude mice were increased with time and reached the highest level after 8 h. T/NT stayed above 3.0 after 12 h, and the tumor could still be developed after 24 h. The number of apoptotic cells in tumor tissue of nude mice treated with I¹³¹-PSCA-mAb was larger than that in the control group.

Conclusion

I¹³¹-PSCA-mAb has the potential to become a new targeted therapy drug for the treatment of prostate cancer.

Object information based interactive segmentation for fatty tissue extraction

Lymph nodes are very important factors for diagnosing gastric cancer in clinical use, and are usually distributed within the fatty tissue around the stomach. When extracting fatty tissues whose structures and textures are complicated, automatic extraction is still a challenging task, while manual extraction is time-consuming. Consequently, semi-automatic extraction, which allows introducing interactive operations, appears to be more realistic. Currently, most interactive methods need to indicate the position and main features in both the object and background. However, it is easier for radiologists to only mark object information. Due to this issue, a new Object Information based Interactive Segmentation (OIIS) method is proposed in this paper. Different from the most existing methods, OIIS just needs to input the object information, while the background information is not required. Experimental results and comparative studies show that OIIS is effective for fatty tissue extraction.

Functional activity maps based on significance measures and Independent Component Analysis

The use of functional imaging has been proven very helpful for the process of diagnosis of neurodegenerative diseases, such as Alzheimer's Disease (AD). In many cases, the analysis of these images is performed by manual reorientation and visual interpretation. Therefore, new statistical techniques to perform a more quantitative analysis are needed. In this work, a new statistical approximation to the analysis of functional images, based on significance measures and Independent Component Analysis (ICA) is presented. After the images preprocessing, voxels that allow better separation of the two classes are extracted, using significance measures such as the Mann-Whitney-Wilcoxon U-Test (MWW) and Relative Entropy (RE). After this feature selection step, the voxels vector is modelled by means of ICA, extracting a few independent components which will be used as an input to the classifier. Naive Bayes and Support Vector Machine (SVM) classifiers are used in this work. The proposed system has been applied to two different databases. A 96-subjects Single Photon Emission Computed Tomography (SPECT) database from the "Virgen de las Nieves" Hospital in Granada, Spain, and a 196-subjects Positron Emission Tomography (PET) database from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Values of accuracy up to 96.9% and 91.3% for SPECT and PET databases are achieved by the proposed system, which has yielded many benefits over methods proposed on recent works.

A novel tool for segmenting 3D medical images based on generalized cylinders and active surfaces

Three-dimensional (3D) medical imaging has been incorporated in routine clinical practice, since the required infrastructure has become increasingly affordable. New algorithms and applications are needed to serve the additional image processing and analysis functions in 3D space. In this work we propose a system for semi-automatic modeling and segmentation of elongated salient and anatomical objects in 3D medical images. The proposed methodology is based on a novel mathematical formalization of a well-known class of geometric primitives, namely generalized cylinders (GCs), which exhibits advantages over the existing parametric definition. Since the anatomical objects have to be modeled by their intersection with the transverse image planes, the proposed methodology includes also a new seeded region growing (SRG) segmentation algorithm for ellipse detection in 2D images, based on a priori shape knowledge. Finally, the resulting GC model is used to initialize an active surface (AS) segmentation method, in order to accurately delineate the required object. In this work we present the proposed algorithms in detail, along with the evaluation of the accuracy of the model-based segmentation by experts. Results show that elongated objects like the aorta and the trachea may be segmented with sensitivity between 90% and 95%. The proposed SRG-ellipse detector requires minimal user-initialization and its executions requires only few seconds for each image slice on an average laptop. The evolution of the AS requires less than one second per iteration for a typical CT image. Comparisons are provided with state of the art semi-automatic medical image processing software, which validate the merit of the proposed work.

Active ellipse model and automatic chamber detection in apical views of echocardiography images

In this article, an automatic method for detection of all chambers in apical two- and four-chamber views is proposed. The method is based on four evolving ellipses with their sizes and alignments (centre point) gradually changing through iterations until they reach to the point that approximates the chamber boundaries. The interaction between the internal, external and inter-elliptic forces controls the simultaneous evolution of ellipses. Since no prior assumption of the approximate location is required with our approach, the specialists are not required to locate the centre points of chambers in apical images, making the overall segmentation fully automated. Moreover, the resultant ellipse inside a chamber could be used as the initial contour in segmentation techniques such as active contour models, where the initial contour has a significant role for higher accuracy and faster convergence. The simplicity of equations developed in our approach make for a computationally faster algorithm, compared with former approaches that utilize morphologic operators. Our evolving ellipse does not go beyond the gaps, a problem that normally exists within boundaries in echo images, making our overall segmentation process more robust against the gaps. To evaluate the proposed method, a subset of 80 images is selected and three observers are requested to manually draw best ellipses inside the images and compare them with our results. The obtained dice coefficient results (87.62 ± 4.53% for observer-1, 83.18 ± 6.20% for observer-2, 86.02 ± 5.16% for observer-3) indicate that the proposed method has a useful performance.

Fully automatic liver segmentation through graph-cut technique

The accurate knowledge of the liver structure including blood vessels topography, liver surface and lesion localizations is usually required in treatments like liver ablations and radiotherapy. In this paper, we propose an approach for automatic segmentation of liver complex geometries. It consists of applying a graph-cut method initialized by an adaptive threshold. The algorithm has been tested on 10 datasets (CT and MR). A parametric comparison with the results obtained by previous algorithms based on active contour is also carried out and discussed. Main limitations of active contour approaches result to be overcome and segmentation is improved. Feasibility to routinely use graph-cut approach for automatic liver segmentation is also demonstrated.

A new fully automatic and robust algorithm for fast segmentation of liver tissue and tumors from CT scans

Accurate knowledge of the liver structure, including liver surface and lesion localization, is usually required in treatments such as liver tumor ablations and/or radiotherapy. This paper presents a new method and corresponding algorithm for fast segmentation of the liver and its internal lesions from CT scans. No interaction between the user and analysis system is required for initialization since the algorithm is fully automatic. A statistical model-based approach was created to distinguish hepatic tissue from other abdominal organs. It was combined to an active contour technique using gradient vector flow in order to obtain a smoother and more natural liver surface segmentation. Thereafter, automatic classification was performed to isolate hepatic lesions from liver parenchyma. Twenty-one datasets, presenting different anatomical and pathological situations, have been processed and analyzed. Special focus has been driven to the resulting processing time together with quality assessment. Our method allowed robust and efficient liver and lesion segmentations very close to the ground truth, in a relatively short processing time (average of 11.4 s for a 512 x 512-pixel slice). A volume overlap of 94.2% and an accuracy of 3.7 mm were achieved for liver surface segmentation. Sensitivity and specificity for tumor lesion detection were 82.6% and 87.5%, respectively.

A new fully automatic and robust algorithm for fast segmentation of liver tissue and tumors from CT scans

Accurate knowledge of the liver structure, including liver surface and lesion localization, is usually required in treatments such as liver tumor ablations and/or radiotherapy. This paper presents a new method and corresponding algorithm for fast segmentation of the liver and its internal lesions from CT scans. No interaction between the user and analysis system is required for initialization since the algorithm is fully automatic. A statistical model-based approach was created to distinguish hepatic tissue from other abdominal organs. It was combined to an active contour technique using gradient vector flow in order to obtain a smoother and more natural liver surface segmentation. Thereafter, automatic classification was performed to isolate hepatic lesions from liver parenchyma. Twenty-one datasets, presenting different anatomical and pathological situations, have been processed and analyzed. Special focus has been driven to the resulting processing time together with quality assessment. Our method allowed robust and efficient liver and lesion segmentations very close to the ground truth, in a relatively short processing time (average of 11.4 s for a 512 x 512-pixel slice). A volume overlap of 94.2% and an accuracy of 3.7 mm were achieved for liver surface segmentation. Sensitivity and specificity for tumor lesion detection were 82.6% and 87.5%, respectively.

Fully automated image-guided needle insertion: application to small animal biopsies

The study of biological process evolution in small animals requires time-consuming and expansive analyses of a large population of animals. Serial analyses of the same animal is potentially a great alternative. However non-invasive procedures must be set up, to retrieve valuable tissue samples from precisely defined areas in living animals. Taking advantage of the high resolution level of in vivo molecular imaging, we defined a procedure to perform image-guided needle insertion and automated biopsy using a micro CT-scan, a robot and a vision system. Workspace limitations in the scanner require the animal to be removed and laid in front of the robot. A vision system composed of a grid projector and a camera is used to register the designed animal-bed with to respect to the robot and to calibrate automatically the needle position and orientation. Automated biopsy is then synchronised with respiration and performed with a pneumatic translation device, at high velocity, to minimize organ deformation. We have experimentally tested our biopsy system with different needles.

A modular and evolutive component oriented software architecture for patient modeling

This paper deals with the design aspect of a software aiming at modeling the anatomical and pathological structures of patients from medical images, for diagnosis purposes. In terms of functionalities, it allows to combine image processing algorithms, and to visualize and manipulate 3D models and images. The proposed software uses specific extensible and reusable components and a system managing their combination, thanks to a formal XML-based description of their interfaces. This architecture facilitates the dynamic integration of new functionalities, in particular in terms of image processing algorithms. We describe the structural and behavioral aspects of the proposed reusable component-based architecture. We also discuss the potential of this work for developing other softwares in the field of computer aided surgery.