In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients—manually annotated by up to four raters—and to 65 comparable scans generated using tumor image simulation software Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%–85%), illustrating the difficulty of this task We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource

/pdf/the-multimodal-brain-tumor-image-segmentation-benchmark-2gfraj98dk.pdf

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

The Visual Display of Quantitative Information

With an estimated 160,000 deaths in 2018, lung cancer is the most common cause of cancer death in the United States1. Lung cancer screening using low-dose computed tomography has been shown to reduce mortality by 20–43% and is now included in US screening guidelines1–6. Existing challenges include inter-grader variability and high false-positive and false-negative rates7–10. We propose a deep learning algorithm that uses a patient’s current and prior computed tomography volumes to predict the risk of lung cancer. Our model achieves a state-of-the-art performance (94.4% area under the curve) on 6,716 National Lung Cancer Screening Trial cases, and performs similarly on an independent clinical validation set of 1,139 cases. We conducted two reader studies. When prior computed tomography imaging was not available, our model outperformed all six radiologists with absolute reductions of 11% in false positives and 5% in false negatives. Where prior computed tomography imaging was available, the model performance was on-par with the same radiologists. This creates an opportunity to optimize the screening process via computer assistance and automation. While the vast majority of patients remain unscreened, we show the potential for deep learning models to increase the accuracy, consistency and adoption of lung cancer screening worldwide. A convolutional neural network performs automated prediction of malignancy risk of pulmonary nodules in chest CT scan volumes and improves accuracy of lung cancer screening.

End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography

Image enhancement plays an important role in image processing and analysis. Among various enhancement algorithms, Retinex-based algorithms can efficiently enhance details and have been widely adopted. Since Retinex-based algorithms regard illumination removal as a default preference and fail to limit the range of reflectance, the naturalness of non-uniform illumination images cannot be effectively preserved. However, naturalness is essential for image enhancement to achieve pleasing perceptual quality. In order to preserve naturalness while enhancing details, we propose an enhancement algorithm for non-uniform illumination images. In general, this paper makes the following three major contributions. First, a lightness-order-error measure is proposed to access naturalness preservation objectively. Second, a bright-pass filter is proposed to decompose an image into reflectance and illumination, which, respectively, determine the details and the naturalness of the image. Third, we propose a bi-log transformation, which is utilized to map the illumination to make a balance between details and naturalness. Experimental results demonstrate that the proposed algorithm can not only enhance the details but also preserve the naturalness for non-uniform illumination images.

Naturalness Preserved Enhancement Algorithm for Non-Uniform Illumination Images

Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge.

An imaging system ( 10 ) analyzes airways of a patient. The system ( 10 ) includes a hardware phantom ( 50 ) including tubes ( 54 ) representative of airways. The tubes ( 54 ) include different lumen sizes and/or wall thicknesses. The system further includes an imaging scanner ( 12 ) for scanning a region of interest (ROI), including the airways, and the hardware phantom ( 50 ) to create raw image data. At least one processor ( 32 ) is programmed to at least one of: (1) correct measurements of walls of the airways based on measurements of lumen size and/or wall thickness of the tubes ( 54 ) and known lumen size and/or wall thickness of the tubes ( 54 ); and (2) generate an image of the ROI in which color and/or opacity of the airways are based on a comparison of images or maps of the tubes ( 54 ) and images or maps of the airways. A corresponding method is also provided.

Chronic obstructive pulmonary disease (copd) phantom for computed tomography (ct) and methods of using the same

The bronchial tree is of direct clinical importance in the context of respective diseases, such as chronic obstructive
pulmonary disease (COPD). It furthermore constitutes a reference structure for object localization in the lungs and it
finally provides access to lung tissue in, e.g., bronchoscope based procedures for diagnosis and therapy. This paper
presents a comprehensive anatomical model for the bronchial tree, including statistics of position, relative and absolute
orientation, length, and radius of 34 bronchial segments, going beyond previously published results. The model has been
built from 16 manually annotated CT scans, covering several branching variants. The model is represented as a
centerline/tree structure but can also be converted in a surface representation. Possible model applications are either to
anatomically label extracted bronchial trees or to improve the tree extraction itself by identifying missing segments or
sub-trees, e.g., if located beyond a bronchial stenosis. Bronchial tree labeling is achieved using a naive Bayesian
classifier based on the segment properties contained in the model in combination with tree matching. The tree matching
step makes use of branching variations covered by the model. An evaluation of the model has been performed in a leaveone-
out manner. In total, 87% of the branches resulting from preceding airway tree segmentation could be correctly
labeled. The individualized model enables the detection of missing branches, allowing a targeted search, e.g., a local rerun
of the tree-segmentation segmentation.

Anatomical modeling of the bronchial tree

A graph based approach to automated shutter detection in digital X-ray images

A system and a method are provided for enabling a user to interactively navigate through a set of slice images, the set of slice images jointly representing an image volume showing an anatomical structure of a patient A user may be enabled to switch from a static viewing mode to a navigation mode based on navigation commands received from a user input device operable by the user A display processor may be configured for, in the static viewing mode, generating an output image comprising one slice image of the set of slice images The display processor may be configured for, in the navigation mode, replacing the said one slice image in the output image by a volume rendering of a slab of the image volume, the slab comprising more than one slice image The system and method thus selectively switch to volume rendering, namely during navigation, whereas in a static (ie, non-navigation) viewing mode, a slice image is shown Advantageously, the user may thus follow structures more accurately when navigate through a volume image, thereby more quickly and accurately identifying slice images of interest

Medical image navigation system

Dynamic Contrast Enhanced Breast MR Imaging (DCE BMRI) has emerged as a modality for breast cancer diagnosis. In
this modality a temporal sequence of volume images of the breasts is acquired, where a contrast agent is injected after
acquisition of the first 3D image. Since the introduction of the modality, research has been directed at the development
of computer-aided support for the diagnostic workup. This includes automatic segmentation of mass-like lesions, lesion
characterization, and lesion classification. Robustness, user-independence, and reproducibility of the results of
computerized methods are essential for such methods to be acceptable for clinical application.
A previously proposed and evaluated computerized lesion segmentation method has been further analyzed in this study.
The segmentation method uses as input a subtraction image (post-contrast - pre-contrast) and a user defined region of
interest (ROI). Previous evaluation studies investigated the robustness of the segmentation against variations in the user
selected ROI. Robustness of the method against variations in the image data itself has so far not been investigated. To fill
this gap is the purpose of this study.
In this study, the segmentation algorithm was applied to a series of subtraction images built from the pre-contrast volume
and all available post-contrast image volumes, successively. This provides set of typically 4-5 delineations per lesion,
each based on a different phase of the dynamic sequence.
Analysis of the apparent lesion volumes derived from these delineations and comparison to manual delineations showed
that computerized segmentation is more robust and reproducible than manual segmentation, even if computer
segmentations are computed on subtraction images derived from different dynamic phases of the DCE MRI study, while
all manual segmentations of a lesion are derived from one and the same dynamic phase of the study.
Furthermore, it could be shown that the rate of apparent change of lesion volume over the course of a DCE MRI study is
significantly dependent on the lesion type (benign vs. malignant).

Rafael Wiemker

Papers

Chronic obstructive pulmonary disease (copd) phantom for computed tomography (ct) and methods of using the same

Anatomical modeling of the bronchial tree

A graph based approach to automated shutter detection in digital X-ray images

Medical image navigation system

Temporal Variations in Apparent Breast Lesion Volume in Dynamic Contrast Enhanced Breast MR Imaging