Computer-assisted Detection System Research Articles

Virtual colonoscopy: effect of computer-assisted detection (CAD) on radiographer performance

To investigate the effect of a virtual colonoscopy (VC) computed-assisted detection (CAD) system on polyp detection by trained radiographers. Four radiographers trained in VC interpretation and utilization of CAD systems read a total of 62 endoscopically validated VC examinations containing 150 polyps (size range 5-50mm) in four sessions, recording any polyps found and the examination interpretation time, first without and then with the addition of CAD as a "second reader". After a temporal separation of 6 weeks to reduce recall bias, VC examinations were re-read using "concurrent reader" CAD. Interpretation times, polyp detection, and number of false-positives were compared between the different reader paradigms using paired t and paired exact tests. Overall, use of "second reader" CAD significantly improved polyp detection by 12% (p<0.001, CI 6%,17%)) from 48 to 60%. There was no significant improvement using CAD as a concurrent reader (p=0.20; difference of 7%, CI -3%, 16%) and no significant overall difference in recorded false-positives with second reader or concurrent CAD paradigms compared with unassisted reading (p=0.25 and 0.65, respectively). The mean interpretation time was 21.7 min for unassisted reading, 29.6 (p<0.001) min for second reader and 19.1 min (p=0.12) for concurrent reading paradigms. CAD, when used as a second reader, can significantly improve radiographer reading performance with only a moderate increase in interpretation times.

Information‐theoretic CAD system in mammography: Entropy‐based indexing for computational efficiency and robust performance

We have previously presented a knowledge-based computer-assisted detection (KB-CADe) system for the detection of mammographic masses. The system is designed to compare a query mammographic region with mammographic templates of known ground truth. The templates are stored in an adaptive knowledge database. Image similarity is assessed with information theoretic measures (e.g., mutual information) derived directly from the image histograms. A previous study suggested that the diagnostic performance of the system steadily improves as the knowledge database is initially enriched with more templates. However, as the database increases in size, an exhaustive comparison of the query case with each stored template becomes computationally burdensome. Furthermore, blind storing of new templates may result in redundancies that do not necessarily improve diagnostic performance. To address these concerns we investigated an entropy-based indexing scheme for improving the speed of analysis and for satisfying database storage restrictions without compromising the overall diagnostic performance of our KB-CADe system. The indexing scheme was evaluated on two different datasets as (i) a search mechanism to sort through the knowledge database, and (ii) a selection mechanism to build a smaller, concise knowledge database that is easier to maintain but still effective. There were two important findings in the study. First, entropy-based indexing is an effective strategy to identify fast a subset of templates that are most relevant to a given query. Only this subset could be analyzed in more detail using mutual information for optimized decision making regarding the query. Second, a selective entropy-based deposit strategy may be preferable where only high entropy cases are maintained in the knowledge database. Overall, the proposed entropy-based indexing scheme was shown to reduce the computational cost of our KB-CADe system by 55% to 80% while maintaining the system's diagnostic performance.

GridCAD: Grid-based Computer-aided Detection System

Grid computing-the use of a distributed network of electronic resources to cooperatively perform subsets of computationally intensive tasks-may help improve the speed and accuracy of radiologic image interpretation by enabling collaborative computer-based and human readings. GridCAD, a software application developed by using the National Cancer Institute Cancer Biomedical Informatics Grid architecture, implements the fundamental elements of grid computing and demonstrates the potential benefits of grid technology for medical imaging. It allows users to query local and remote image databases, view images, and simultaneously run multiple computer-assisted detection (CAD) algorithms on the images selected. The prototype CAD systems that are incorporated in the software application are designed for the detection of lung nodules on thoracic computed tomographic images. GridCAD displays the original full-resolution images with an overlay of nodule candidates detected by the CAD algorithms, by human observers, or by a combination of both types of readers. With an underlying framework that is computer platform independent and scalable to the task, the software application can support local and long-distance collaboration in both research and clinical practice through the efficient, secure, and reliable sharing of resources for image data mining, analysis, and archiving.

GridIMAGE: A Novel Use of Grid Computing to Support Interactive Human and Computer-Assisted Detection Decision Support

This paper describes a Grid-aware image reviewing system (GridIMAGE) that allows practitioners to (a) select images from multiple geographically distributed digital imaging and communication in medicine (DICOM) servers, (b) send those images to a specified group of human readers and computer-assisted detection (CAD) algorithms, and (c) obtain and compare interpretations from human readers and CAD algorithms. The currently implemented system was developed using the National Cancer Institute caGrid infrastructure and is designed to support the identification of lung nodules on thoracic computed tomography. However, the infrastructure is general and can support any type of distributed review. caGrid data and analytical services are used to link DICOM image databases and CAD systems and to interact with human readers. Moreover, the service-oriented and distributed structure of the GridIMAGE framework enables a flexible system, which can be deployed in an institution (linking multiple DICOM servers and CAD algorithms) and in a Grid environment (linking the resources of collaborating research groups). GridIMAGE provides a framework that allows practitioners to obtain interpretations from one or more human readers or CAD algorithms. It also provides a mechanism to allow cooperative imaging groups to systematically perform image interpretation tasks associated with research protocols.

Korrelation zwischen histologischem Befund und einem Computer-assistierten Detektionssystem (CAD) für die Mammografie

PURPOSE: To assess the sensitivity and specificity of a wavelet transformation-based computer-assisted detection system (CAD) for mammography after film digitizing and data compression. MATERIALS AND METHODS: 128 one-sided mammograms of 128 patients with 135 histological proven lesions were evaluated by the CAD System (66 malignant lesions, 69 benign lesions). Two groups were enrolled and analysed: one group with calcifications only and one group with masses and calcifications. None of the lesions were palpable. After stereotactic wire placement the histological specimens were obtained by open breast biopsy. The mammograms were digitized and afterwards analysed by the CAD system integrated in the mammography system. RESULTS: CAD correctly detected 95.4 % of the malignant lesions with microcalcifications only and 81.8 % of the malignant lesions with mammographic masses and calcifications. Overall, 90.9 % of malignant lesions were detected. With respect to all benign lesions CAD correctly did not mark 50 % of lesions with microcalcifications only and 44 % of masses with calcifications. The specificity was 50.7 %. CONCLUSION: With regard to its high sensitivity the evaluated CAD system is useful as a supporting method in mammographies. Because of its low specificity every CAD mark must be cross-checked by a skilled mammography examiner.

Computer assisted detection software for CT colonography: effect of sphericity filter on performance characteristics for patients with and without fecal tagging

The aim of this study is to investigate the effect of changing sphericity filter values on performance of a computer assisted detection (CAD) system for CT colonography for data with and without fecal tagging. Colonography data from 138 patients with 317 validated polyps were divided into those with (86) and without (52) fecal tagging. Polyp coordinates were established by three observers and datasets analysed subsequently by a proprietary CAD system used at four discrete sphericity filter settings. Prompts were compared with the known coordinates in order to determine sensitivity and specificity. Sensitivity was highest at low sphericity; of 164 polyps 6 mm or more, 144 (87.8%) were detected at sphericity 0.3, and 132 (80.1%) at sphericity 0.9. Of 42 polyps measuring 10 mm or more, 40 (95.2%) were detected at sphericity 0.3, and 36 (85.7%) at sphericity 0.9. There was no significant difference in sensitivity for tagged and un-tagged data but specificity was reduced in tagged data at low sphericity and significantly reduced in untagged data at high sphericity. CAD had a sensitivity of 95.2% for polyps measuring 1 cm or more and 87.8% for polyps 6 mm or more when used at a sphericity setting of 0.3. Higher sphericity settings increased specificity while reducing sensitivity. The bowel preparation used significantly impacts on specificity.

Computer-assisted detection for CT colonography: external validation

To externally validate a computer-assisted detection (CAD) system for computed tomography (CT) colonography, using data from a single centre uninvolved with the software development. Twenty-five multi-detector CT colonography examinations of patients with validated polyps accumulated at a single centre were examined by two readers who used endoscopic and histopathological data to identify polyp coordinates. A CAD system that had been developed using data from elsewhere, and had not previously encountered the present data, was then applied to the data at sphericity filter settings of 0.75 and 0.50 and identified potential polyps. True-positive, false-negative, and false-positive counts were determined by comparison with the known polyp coordinates. Twenty-five patients had 57 polyps, median size 6mm (range 1-15mm). Per-patient sensitivity for the CAD system was 96% (24 of 25). The CAD system detected 44 (77%) polyps at sphericity setting 0.75 and 49 (86%) polyps at sphericity 0.50: the additional five polyps detected all measured 5mm or less. Sphericity of 0.75 resulted in a median of 10 (one to 34) easily dismissed false-positive prompts per patient and a median of 4 (zero to 15) that needed three-dimensional rendering before dismissal. This rose to 32 (16 to 99) and 11 (three to 35), respectively, at sphericity 0.5. A per-patient sensitivity of 96% was found for the CAD system (in patients with a median polyp diameter of 6mm) using external validation, a more stringent test than either internal cross-validation or temporal validation. Decreasing sphericity increases sensitivity for small polyps at the expense of decreased specificity.

Computer-assisted detection as a second reader in symptomatic Asian women with palpable breast cancer.

To determine the potential role of a computer-assisted detection (CAD) algorithm as a second reader for experienced and inexperienced radiologists in mammography reading in Asian women. Two-view mammograms performed in 124 consecutive patients who presented with palpable breast cancer masses were retrospectively evaluated by two experienced breast radiologists (7 and 10 years' experience). The original reports of the session radiologists with variable experience of reading mammograms (2 to more than 10 years) were also evaluated. The number of suspicious masses and microcalcification clusters detected in each patient by both groups of radiologists were recorded. The radiologists then re-evaluated the films with the CAD system as a second reader. Any improvement in the detectability of breast pathology by either the experienced radiologists and/or by the session radiologists was then assessed. A total of 127 breasts had biopsy-proven carcinoma; 74 breasts had mastectomy performed. All the imaging results were correlated with tru-cut biopsy or mastectomy histology. With CAD-aided interpretation, there were altogether 95 visible masses and 77 suspicious microcalcification clusters in 109 mammographically detectable cancers correlated with histology results. There was a 7.4% (7/95) and 10.4% (8/77) increase in the number of masses and microcalcification clusters detected, respectively, by the experienced radiologists after application of CAD, whereas the increase was 13.7% (13/95) and 27.3% (21/77) for detection of masses and microcalcifications by the session radiologists, respectively. In 9 patients, a secondary focus detected by CAD was confirmed by histology. Three patients had contralateral breast tumors, 1 had a satellite invasive tumor while 5 had ductal carcinoma in situ on the same breast. Based on the biopsies and 74 mastectomies, the true-positive and false-positive detection rate of CAD was 92.6% and 31.8% for detection of carcinomas. The true-positive and false-positive detection rates were 100% and 58.8% for microcalcification clusters. The current generation CAD algorithm helped to improve the detection rate of carcinomas, calcifications and multifocality in Asian breasts.

Computer-assisted detection of mammographic masses: a template matching scheme based on mutual information.

The purpose of this study was to develop a knowledge-based scheme for the detection of masses on digitized screening mammograms. The computer-assisted detection (CAD) scheme utilizes a knowledge databank of mammographic regions of interest (ROIs) with known ground truth. Each ROI in the databank serves as a template. The CAD system follows a template matching approach with mutual information as the similarity metric to determine if a query mammographic ROI depicts a true mass. Based on their information content, all similar ROIs in the databank are retrieved and rank-ordered. Then, a decision index is calculated based on the query's best matches. The decision index effectively combines the similarity indices and ground truth of the best-matched templates into a prediction regarding the presence of a mass in the query mammographic ROI. The system was developed and evaluated using a database of 1465 ROIs extracted from the Digital Database for Screening Mammography. There were 809 ROIs with confirmed masses (455 malignant and 354 benign) and 656 normal ROIs. CAD performance was assessed using a leave-one-out sampling scheme and Receiver Operating Characteristics analysis. Depending on the formulation of the decision index, CAD performance as high as A(zeta) = 0.87 +/- 0.01 was achieved. The CAD detection rate was consistent for both malignant and benign masses. In addition, the impact of certain implementation parameters on the detection accuracy and speed of the proposed CAD scheme was studied in more detail.

Clinical Performance of Computer-assisted Detection (CAD) System in Detecting Carcinoma in Breasts of Different Densities

OBJECTIVES: To determine the clinical performance of a computer-assisted detection (CAD) system in detecting carcinoma in breasts of different densities. MATERIALS AND METHODS: A total of 264 sets of bilateral screening mammograms taken in craniocaudal and medial–lateral oblique projections during the year 1997 were divided into four groups according to the BI-RADS density classification: fatty (pattern 1), scattered fibroglandular (pattern 2), heterogeneously dense (pattern 3) and extremely dense (pattern 4). Each group contained about 60% normal and 40% biopsy-proven cancer cases. Of the malignant cases, there were a mixture of mammographic findings including focal masses (<2.5 cm), asymmetrical density, architectural distortion or microcalcifications. Films with artefacts and obvious masses >2.5 cm were not included. The chosen cases were then digitized and analysed by the CAD system. Sensitivity was calculated as detection of cancer by at least one marker in at least one view. Specificity was calculated as the number of false-positive marks per image on normal cases. Statistical tests of significance were performed by using contingency tables and Chi square test. RESULTS: The CAD system detected 14 out of the total 15 cancer cases in totally fatty breasts with a sensitivity of 93.3% at a specificity of 1.3 false-positive marks per image. In breasts with scattered fibroglandular pattern, the sensitivity was 93.9% (31/33) and the specificity was 1.6 false-positive marks per image while in heterogeneously dense breasts, the sensitivity of the CAD system fell to 84.8% at a specificity of 1.6 false-positive marks per image. The sensitivity of the CAD system further dropped to 64.3% in markedly dense breasts while maintaining a specificity of 1.2 false-positive marks per image. The decrease in sensitivity in dense breast was found to be significant ( p = 0.046). CONCLUSION: The sensitivity of the CAD system deteriorated significantly as the density of the breast increased while the specificity of the system remained relatively constant. Ho, W. T. and Lam,P.W. (2003). Clinical Radiology 58, 133–136.

Soft-copy mammographic readings with different computer-assisted detection cuing environments: preliminary findings.

To assess the performance of radiologists in the detection of masses and microcalcification clusters on digitized mammograms by using different computer-assisted detection (CAD) cuing environments. Two hundred nine digitized mammograms depicting 57 verified masses and 38 microcalcification clusters in 85 positive and 35 negative cases were interpreted independently by seven radiologists using five display modes. Except for the first mode, for which no CAD results were provided, suspicious regions identified with a CAD scheme were cued in all the other modes by using a combination of two cuing sensitivities (90% and 50%) and two false-positive rates (0.5 and 2.0 per image). A receiver operating characteristic study was performed by using soft-copy images. CAD cuing at 90% sensitivity and a rate of 0.5 false-positive region per image improved observer performance levels significantly (P < .01). As accuracy of CAD cuing decreased so did observer performances (P < .01). Cuing specificity affected mass detection more significantly, while cuing sensitivity affected detection of microcalcification clusters more significantly (P < .01). Reduction of cuing sensitivity and specificity significantly increased false-negative rates in noncued areas (P < .05). Trends were consistent for all observers. CAD systems have the potential to significantly improve diagnostic performance in mammography. However, poorly performing schemes could adversely affect observer performance in both cued and noncued areas.

Coronary vasodilation with dihydropyridines--A pharmacokinetic study

In 26 patients with coronary artery disease, the mean diameters of angiographically 'normal' epicardial coronary arteries were assessed with the aid of a computer-assisted contour detection system (CAAS) before and up to 15 min after onset of a 4-min intravenous-infusion of 2 mg nifedipine (13 patients, group I) or 1 mg nisoldipine (13 patients, group II). Maximal coronary dilation amounted to 20 +/- 9% (4th min) in group I and to 18 +/- 9% (15th min) in group II. In addition, in group II changes of the minimal diameters of 9 coronary obstructions were measured; the maximum increase averaged 28 +/- 15% (7th min). In order to compare the pharmacokinetic properties of these compounds the dilation of the 'normal' coronary segments was correlated with the respective drug plasma levels; maximal plasma concentrations averaged 62 +/- 21 ng ml-1 (7th min) in group I and 17 +/- 7 ng ml-1 (4th min) in group II respectively. A positive, linear correlation between coronary dilation and plasma levels was only found with nifedipine (P less than 0.05); with nisoldipine, however, coronary dilation developed in form of a hysteresis curve, when plotted against plasma levels, probably due to the high receptor affinity of this substance. The prolonged efficacy of nisoldipine could be favourable in oral long-term treatment of patients with coronary artery disease.