Joseph E. Burns

Distributed Human Intelligence for Colonic Polyp Classification in Computer-aided Detection for CT Colonography

PURPOSE To assess the diagnostic performance of distributed human intelligence for the classification of polyp candidates identified with computer-aided detection (CAD) for computed tomographic (CT) colonography. MATERIALS AND METHODS This study was approved by the institutional Office of Human Subjects Research. The requirement for informed consent was waived for this HIPAA-compliant study. CT images from 24 patients, each with at least one polyp of 6 mm or larger, were analyzed by using CAD software to identify 268 polyp candidates. Twenty knowledge workers (KWs) from a crowdsourcing platform labeled each polyp candidate as a true or false polyp. Two trials involving 228 KWs were conducted to assess reproducibility. Performance was assessed by comparing the area under the receiver operating characteristic curve (AUC) of KWs with the AUC of CAD for polyp classification. RESULTS The detection-level AUC for KWs was 0.845 ± 0.045 (standard error) in trial 1 and 0.855 ± 0.044 in trial 2. These were not significantly different from the AUC for CAD, which was 0.859 ± 0.043. When polyp candidates were stratified by difficulty, KWs performed better than CAD on easy detections; AUCs were 0.951 ± 0.032 in trial 1, 0.966 ± 0.027 in trial 2, and 0.877 ± 0.048 for CAD (P = .039 for trial 2). KWs who participated in both trials showed a significant improvement in performance going from trial 1 to trial 2; AUCs were 0.759 ± 0.052 in trial 1 and 0.839 ± 0.046 in trial 2 (P = .041). CONCLUSION The performance of distributed human intelligence is not significantly different from that of CAD for colonic polyp classification.

Detection of vertebral body fractures based on cortical shell unwrapping.

Assessment of trauma patients with multiple injuries can be one of the most clinically challenging situations dealt with by the radiologist. We propose a fully automated method to detect acute vertebral body fractures on trauma CT studies. The spine is first segmented and partitioned into vertebrae. Then the cortical shell of the vertebral body is extracted using deformable dual-surface models. The extracted cortical shell is unwrapped onto a 2D map effectively converting a complex 3D fracture detection problem into a pattern recognition problem of fracture lines on a 2D plane. Twenty-eight features are computed for each fracture line and sent to a committee of support vector machines for classification. The system was tested on 18 trauma CT datasets and achieved 95.3% sensitivity and 1.7 false positives per case by leave-one-out cross validation.

Pitfalls That May Mimic Injuries of the Triangular Fibrocartilage and Proximal Intrinsic Wrist Ligaments at MR Imaging

Diagnosis of injuries to the ligamentous structures of the wrist can be a challenge, particularly when there is involvement of the small, complex structures of the proximal wrist. Recent advances in magnetic resonance (MR) imaging, especially in spatial and contrast resolution, have facilitated more precise visualization of these structures. However, there are a number of pitfalls that may cause difficulty in diagnosis of injuries to the triangular fibrocartilage complex (TFCC), lunotriquetral ligament, and scapholunate ligament. Use of inappropriate MR imaging sequences and MR imaging artifacts may decrease the accuracy of diagnosis of injuries to the TFCC and wrist ligaments, whereas variant anatomy of the proximal wrist structures may mimic disease of the TFCC and wrist ligaments. Knowledge of the detailed anatomy of the wrist, as well as variant patterns of structure morphology and signal intensity, can help differentiate actual disease from normal or variant appearances at assessment with MR imaging.

Detection of Sclerotic Spine Metastases via Random Aggregation of Deep Convolutional Neural Network Classifications

Automated detection of sclerotic metastases (bone lesions) in Computed Tomography (CT) images has potential to be an important tool in clinical practice and research. State-of-the-art methods show performance of 79 % sensitivity or true-positive (TP) rate, at 10 false-positives (FP) per volume. We design a two-tiered coarse-to-fine cascade framework to first operate a highly sensitive candidate generation system at a maximum sensitivity of \(\sim \)92 % but with high FP level (\(\sim \)50 per patient). Regions of interest (ROI) for lesion candidates are generated in this step and function as input for the second tier. In the second tier we generate \(N\) 2D views, via scale, random translations, and rotations with respect to each ROI centroid coordinates. These random views are used to train a deep Convolutional Neural Network (CNN) classifier. In testing, the CNN is employed to assign individual probabilities for a new set of \(N\) random views that are averaged at each ROI to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. We validate the approach on CT images of 59 patients (49 with sclerotic metastases and 10 normal controls). The proposed method reduces the number of FP/vol. from 4 to 1.2, 7 to 3, and 12 to 9.5 when comparing a sensitivity rates of 60, 70, and 80 % respectively in testing. The Area-Under-the-Curve (AUC) is 0.834. The results show marked improvement upon previous work.

Ly-alpha emission from disk absorption systems at high redshift - Star formation in young galaxy disks

Narrow-band imaging observations are reported which were made in an attempt to detect Ly-alpha emission from high-redshift candidate galaxy disk systems discovered as high column density absorbers of background QSOs. For four systems with z = 2.3-2.8, no emission is detected to a limit of about 10 to the -16th ergs/sq cm/s, corresponding to luminosity limits of about 10 exp 42-43 ergs/s for the material producing the absorption. The inferred Ly-alpha luminosities lie one to two orders of magnitude below estimates of the Ly-alpha luminosities for active star-forming epochs of many prescriptions for galaxy formation and also considerably below measured Ly-alpha luminosities for other candidate young galaxies detected in radio surveys. A limiting star-formation rate in these systems of about 2-7 solar masses/yr is set; the limit may be about 10 times larger with small but observationally allowable amounts of dust. 76 refs.

A multi-center milestone study of clinical vertebral CT segmentation☆

A multiple center milestone study of clinical vertebra segmentation is presented in this paper. Vertebra segmentation is a fundamental step for spinal image analysis and intervention. The first half of the study was conducted in the spine segmentation challenge in 2014 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) Workshop on Computational Spine Imaging (CSI 2014). The objective was to evaluate the performance of several state-of-the-art vertebra segmentation algorithms on computed tomography (CT) scans using ten training and five testing dataset, all healthy cases; the second half of the study was conducted after the challenge, where additional 5 abnormal cases are used for testing to evaluate the performance under abnormal cases. Dice coefficients and absolute surface distances were used as evaluation metrics. Segmentation of each vertebra as a single geometric unit, as well as separate segmentation of vertebra substructures, was evaluated. Five teams participated in the comparative study. The top performers in the study achieved Dice coefficient of 0.93 in the upper thoracic, 0.95 in the lower thoracic and 0.96 in the lumbar spine for healthy cases, and 0.88 in the upper thoracic, 0.89 in the lower thoracic and 0.92 in the lumbar spine for osteoporotic and fractured cases. The strengths and weaknesses of each method as well as future suggestion for improvement are discussed. This is the first multi-center comparative study for vertebra segmentation methods, which will provide an up-to-date performance milestone for the fast growing spinal image analysis and intervention.

Automated Detection of Sclerotic Metastases in the Thoracolumbar Spine at CT

PURPOSE To design and validate a computer system for automated detection and quantitative characterization of sclerotic metastases of the thoracolumbar spine on computed tomography (CT) images. MATERIALS AND METHODS This retrospective study was approved by the institutional review board and was HIPAA compliant; informed consent was waived. The data set consisted of CT examinations in 49 patients (14 female, 35 male patients; mean age, 57.0 years; range, 12-77 years), demonstrating a total of 532 sclerotic lesions of the spine of greater than 0.3 cm(3) in volume, and in 10 control case patients (four women, six men; mean age, 55.2 years; range, 19-70 years) without spinal lesions. CT examinations were divided into training and test sets, and images were analyzed according to prototypical fully-automated computer-aided detection (CAD) software. Free-response receiver operating characteristic analysis was performed. RESULTS Lesion detection sensitivity on images in the training set was 90%, relative to reference-standard marked lesions (95% confidence interval [CI]: 83%, 97%), at a false-positive rate (FPR) of 10.8 per patient (95% CI: 6.6, 15.0). For images in the testing set, sensitivity was 79% (95% CI: 74%, 84%), with an FPR of 10.9 per patient (95% CI: 8.5, 13.3). False-negative findings were most commonly (37 [40%] of 93) a result of endplate proximity, with 32 (34% of 93) caused by low CT attenuation. Marginal sclerosis caused by degenerative change (174 [28.1%] of 620 actual detections) was the most common cause of false-positive detections, followed by partial volume averaging with vertebral endplates (173 [27.9%] of 620) and pedicle cortex parallel to the axial imaging plane (121 [19.5%] 620). CONCLUSION This CAD system successfully identified and segmented sclerotic lesions in the thoracolumbar spine.

Automated Detection, Localization, and Classification of Traumatic Vertebral Body Fractures in the Thoracic and Lumbar Spine at CT

PURPOSE To design and validate a fully automated computer system for the detection and anatomic localization of traumatic thoracic and lumbar vertebral body fractures at computed tomography (CT). MATERIALS AND METHODS This retrospective study was HIPAA compliant. Institutional review board approval was obtained, and informed consent was waived. CT examinations in 104 patients (mean age, 34.4 years; range, 14-88 years; 32 women, 72 men), consisting of 94 examinations with positive findings for fractures (59 with vertebral body fractures) and 10 control examinations (without vertebral fractures), were performed. There were 141 thoracic and lumbar vertebral body fractures in the case set. The locations of fractures were marked and classified by a radiologist according to Denis column involvement. The CT data set was divided into training and testing subsets (37 and 67 subsets, respectively) for analysis by means of prototype software for fully automated spinal segmentation and fracture detection. Free-response receiver operating characteristic analysis was performed. RESULTS Training set sensitivity for detection and localization of fractures within each vertebra was 0.82 (28 of 34 findings; 95% confidence interval [CI]: 0.68, 0.90), with a false-positive rate of 2.5 findings per patient. The sensitivity for fracture localization to the correct vertebra was 0.88 (23 of 26 findings; 95% CI: 0.72, 0.96), with a false-positive rate of 1.3. Testing set sensitivity for the detection and localization of fractures within each vertebra was 0.81 (87 of 107 findings; 95% CI: 0.75, 0.87), with a false-positive rate of 2.7. The sensitivity for fracture localization to the correct vertebra was 0.92 (55 of 60 findings; 95% CI: 0.79, 0.94), with a false-positive rate of 1.6. The most common cause of false-positive findings was nutrient foramina (106 of 272 findings [39%]). CONCLUSION The fully automated computer system detects and anatomically localizes vertebral body fractures in the thoracic and lumbar spine on CT images with a high sensitivity and a low false-positive rate.

Seeing Is Believing: Video Classification for Computed Tomographic Colonography Using Multiple-Instance Learning

In this paper, we present development and testing results for a novel colonic polyp classification method for use as part of a computed tomographic colonography (CTC) computer-aided detection (CAD) system. Inspired by the interpretative methodology of radiologists using 3-D fly-through mode in CTC reading, we have developed an algorithm which utilizes sequences of images (referred to here as videos) for classification of CAD marks. For each CAD mark, we created a video composed of a series of intraluminal, volume-rendered images visualizing the detection from multiple viewpoints. We then framed the video classification question as a multiple-instance learning (MIL) problem. Since a positive (negative) bag may contain negative (positive) instances, which in our case depends on the viewing angles and camera distance to the target, we developed a novel MIL paradigm to accommodate this class of problems. We solved the new MIL problem by maximizing a L2-norm soft margin using semidefinite programming, which can optimize relevant parameters automatically. We tested our method by analyzing a CTC data set obtained from 50 patients from three medical centers. Our proposed method showed significantly better performance compared with several traditional MIL methods.

Magnetic resonance imaging of triangular fibrocartilage

Due to their small size and complex structure, diagnosing injury of the proximal wrist ligamentous structures can be challenging. The triangular fibrocartilage complex (TFCC) is an example of one such structure, for which lesions may be missed unless high‐resolution magnetic resonance imaging (MRI) obtained via a standard matrix with a small field of view or high‐resolution imaging matrix (small spatial scale matrix elements/large matrix size) is utilized. While there have been recent advances in increasing MRI spatial resolution, attempts at improved visualization by isolated increase in the spatial resolution will be ineffective if the signal‐to‐noise ratio (SNR) of the images obtained is low. Additionally, high contrast resolution is important to facilitate a more precise visualization of these structures and their pathology. Thus, a balance of the three important imaging factor qualifications of high spatial resolution, high SNR, and high contrast resolution must be struck for optimized TFCC and wrist imaging. The goal of this article, then, is to elucidate the theory and techniques of effective high‐resolution imaging of the proximal ligamentous structures of the wrist, balancing SNR and high contrast resolution constraints, and focusing on imaging of the TFCC as a prototypical example. J. Magn. Reson. Imaging 2012;35:764‐778.