Imaging evaluation of a proposed 3D generative model for MRI to CT translation in the lumbar spine.

BACKGROUND CONTEXT: A computed tomography (CT) and magnetic resonance imaging (MRI) are used routinely in the radiologic evaluation and surgical planning of patients with lumbar spine pathology, with the modalities being complimentary. We have developed a deep learning algorithm which can produce 3D lumbar spine CT images from MRI data alone. This has the potential to reduce radiation to the patient as well as burden on the health care system. PURPOSE: The purpose of this study is to evaluate the accuracy of the synthetic lumbar spine CT images produced using our deep learning model. STUDY DESIGN: A training set of 400 unpaired CTs and 400 unpaired MRI scans of the lumbar spine was used to train a supervised 3D cycle-Gan model. Evaluators performed a set of clinically relevant measurements on 20 matched synthetic CTs and true CTs. These measurements were then compared to assess the accuracy of the synthetic CTs. PATIENT SAMPLE: The evaluation data set consisted of 20 patients who had CT and MRI scans performed within a 30-day period of each other. All patient data was deidentified. Notable exclusions included artefact from patient motion, metallic implants or any intervention performed in the 30 day intervening period. OUTCOME MEASURES: The outcome measured was the mean difference in measurements performed by the group of evaluators between real CT and synthetic CTs in terms of absolute and relative error. METHODS: Data from the 20 MRI scans was supplied to our deep learning model which produced 20 "synthetic CT" scans. This formed the evaluation data set. Four clinical evaluators consisting of neurosurgeons and radiologists performed a set of 24 clinically relevant measurements on matched synthetic CT and true CTs in 20 patients. A test set of measurements were performed prior to commencing data collection to identify any significant interobserver variation in measurement technique. RESULTS: The measurements performed in the sagittal plane were all within 10% relative error with the majority within 5% relative error. The pedicle measurements performed in the axial plane were considerably less accurate with a relative error of up to 34%. CONCLUSIONS: The computer generated synthetic CTs demonstrated a high level of accuracy for the measurements performed in-plane to the original MRIs used for synthesis. The measurements performed on the axial reconstructed images were less accurate, attributable to the images being synthesized from nonvolumetric routine sagittal T1-weighted MRI sequences. It is hypothesized that if axial sequences or volumetric data were input into the algorithm these measurements would have improved accuracy.

Segmentation of cervical nuclei using SLIC and pairwise regional contrast.

A framework to detect and segment nuclei from cervical cytology images is proposed in this study. Poor contrast, spurious edges, degree of overlap, and intensity inhomogeneity make the nuclei segmentation task more complex in overlapping cell images. The proposed technique segments cervical nuclei by merging over-segmented SLIC superpixel regions using a novel region merging criteria based on pairwise regional contrast and image gradient contour evaluations. The framework was evaluated using the first overlapping cervical cytology image segmentation challenge - ISBI 2014 dataset. The result shows that the proposed framework outperforms the state-of-the-art algorithms in nucleus detection and segmentation accuracies.

Circular shape constrained fuzzy clustering (CiscFC) for nucleus segmentation in Pap smear images.

Accurate detection and segmentation of cell nucleus is the precursor step towards computer aided analysis of Pap smear images. This is a challenging and complex task due to degree of overlap, inconsistent staining and poor contrast. In this paper, a novel nucleus segmentation method is proposed by incorporating a circular shape function in fuzzy clustering. The proposed method was evaluated quantitatively and qualitatively using the Overlapping Cervical Cytology Image Segmentation Challenge - ISBI 2014 challenge dataset comprised of 945 overlapping Pap smear images. It achieved superior performance in terms of Dice similarity coefficient of 0.938, pixel-based recall 0.939 and object based precision 0.968. The results were compared with the standard fuzzy c-means (FCM) clustering, ISBI 2014 challenge submissions and recent state-of-the-art methods. The outcome shows that the new approach can produce more accurate nucleus boundaries while keeping high level of precision and recall.

Automated analysis of breast parenchymal patterns in whole breast ultrasound images: preliminary experience.

PURPOSE: A computerized classification scheme to recognize breast parenchymal patterns in whole breast ultrasound (US) images was developed. A preliminary evaluation of the system performance was performed. METHODS: Breast parenchymal patterns were classified into three categories: mottled pattern (MP), intermediate pattern (IP), and atrophic pattern (AP). Each classification was defined as proposed by an experienced physician. A total of 281 image features were extracted from a volume of interest which was automatically segmented. Canonical discriminant analysis with stepwise feature selection was employed for the classification of the parenchymal patterns. RESULTS: The classification scheme accuracy was computed to be 83.3% (10/12 cases) in MP cases, 91.7% (22/24 cases) in IP cases, 92.9% (13/14 cases) in AP cases, and 90.0% (45/50 cases) in all the cases. CONCLUSIONS: The feasibility of an automated ultrasonography classifier for parenchymal patterns was demonstrated with promising results in whole breast US images.

Automated segmentation of mammary gland regions in non-contrast X-ray CT images.

The identification of mammary gland regions is a necessary processing step during the anatomical structure recognition of human body and can be expected to provide useful information for breast tumor diagnosis. This paper proposes a fully automated scheme for segmenting the mammary gland regions in non-contrast torso CT images. This scheme calculates the probability of each voxel belonging to the mammary gland or chest muscle in CT images as the reference of the segmentation, and decides the mammary gland regions based on CT number automatically. The probability is estimated from the location of the mammary glands and chest muscles in CT images. The location is investigated from a knowledge base that stores pre-recognized anatomical structures using a number of different CT scans. We applied this scheme to 66 patient cases (female, age: 20-80) and evaluated the accuracy by using the Jaccard similarity coefficient (JSC) between the segmented results and two gold standards that were generated manually by 2 medical experts independently for each CT case. The result showed that the mean value of the JSC score was 0.83 with the standard deviation of 0.09 for 66 CT cases. The proposed scheme was applied to investigate the breast density distributions in normal mammary gland regions so as to demonstrate the effect and usefulness of the proposed scheme.

Computer-aided diagnosis: the emerging of three CAD systems induced by Japanese health care needs.

The aim of this paper is to describe three emerging computer-aided diagnosis (CAD) systems induced by Japanese health care needs. CAD has been developing fast in the last two decades. The idea of using a computer to help in medical image diagnosis is not new. Some pioneer studies are dated back to the 1960s. In 1998, the first U.S. FDA (Food and Drug Administration) approved commercial CAD system, a film-digitized mammography system, was launched by R2 Technologies, Inc. The success was quickly repeated by a number of companies. The approval of Medicare CAD reimbursement in the U.S. in 2001 further boosted the industry. Today, CAD has its significance in the economy of the medical industry. FDA approved CAD products in the field of breast imaging (mammography, ultrasonography and breast MRI) and chest imaging (radiography and CT) can be seen. In Japan, as part of the "Knowledge Cluster Initiative" of the government, three computer-aided diagnosis (CAD) projects are hosted at the Gifu University since 2004. These projects are regarding the development of CAD systems for the early detection of (1) cerebrovascular diseases using brain MRI and MRA images by detecting lacunar infarcts, unruptured aneurysms, and arterial occlusions; (2) ocular diseases such as glaucoma, diabetic retinopathy, and hypertensive retinopathy using retinal fundus images; and (3) breast cancers using ultrasound 3-D volumetric whole breast data by detecting the breast masses. The projects are entering their final development stage. Preliminary results are presented in this paper. Clinical examinations will be started soon, and commercialized CAD systems for the above subjects will appear by the completion of this project.