Automatic feature learning using multichannel ROI based on deep structured algorithms for computerized lung cancer diagnosis

This study aimed to analyze the ability of extracting automatically generated features using deep structured algorithms in lung nodule CT image diagnosis, and compare its performance with traditional computer aided diagnosis (CADx) systems using hand-crafted features. All of the 1018 cases were acquired from Lung Image Database Consortium (LIDC) public lung cancer database. The nodules were segmented according to four radiologists' markings, and 13,668 samples were generated by rotating every slice of nodule images. Three multichannel ROI based deep structured algorithms were designed and implemented in this study: convolutional neural network (CNN), deep belief network (DBN), and stacked denoising autoencoder (SDAE). For the comparison purpose, we also implemented a CADx system using hand-crafted features including density features, texture features and morphological features. The performance of every scheme was evaluated by using a 10-fold cross-validation method and an assessment index of the area under the receiver operating characteristic curve (AUC). The observed highest area under the curve (AUC) was 0.899±0.018 achieved by CNN, which was significantly higher than traditional CADx with the AUC=0.848±0.026. The results from DBN was also slightly higher than CADx, while SDAE was slightly lower. By visualizing the automatic generated features, we found some meaningful detectors like curvy stroke detectors from deep structured schemes. The study results showed the deep structured algorithms with automatically generated features can achieve desirable performance in lung nodule diagnosis. With well-tuned parameters and large enough dataset, the deep learning algorithms can have better performance than current popular CADx. We believe the deep learning algorithms with similar data preprocessing procedure can be used in other medical image analysis areas as well.

Computerized breast cancer analysis system using three stage semi-supervised learning method

BACKGROUND AND OBJECTIVE

A large number of labeled medical image data is usually a requirement to train a well-performed computer-aided detection (CAD) system. But the process of data labeling is time consuming, and potential ethical and logistical problems may also present complications. As a result, incorporating unlabeled data into CAD system can be a feasible way to combat these obstacles.

METHODS

In this study we developed a three stage semi-supervised learning (SSL) scheme that combines a small amount of labeled data and larger amount of unlabeled data. The scheme was modified on our existing CAD system using the following three stages: data weighing, feature selection, and newly proposed dividing co-training data labeling algorithm. Global density asymmetry features were incorporated to the feature pool to reduce the false positive rate. Area under the curve (AUC) and accuracy were computed using 10 fold cross validation method to evaluate the performance of our CAD system. The image dataset includes mammograms from 400 women who underwent routine screening examinations, and each pair contains either two cranio-caudal (CC) or two mediolateral-oblique (MLO) view mammograms from the right and the left breasts. From these mammograms 512 regions were extracted and used in this study, and among them 90 regions were treated as labeled while the rest were treated as unlabeled.

RESULTS

Using our proposed scheme, the highest AUC observed in our research was 0.841, which included the 90 labeled data and all the unlabeled data. It was 7.4% higher than using labeled data only. With the increasing amount of labeled data, AUC difference between using mixed data and using labeled data only reached its peak when the amount of labeled data was around 60.

CONCLUSIONS

This study demonstrated that our proposed three stage semi-supervised learning can improve the CAD performance by incorporating unlabeled data. Using unlabeled data is promising in computerized cancer research and may have a significant impact for future CAD system applications.

A novel approach to nodule feature optimization on thin section thoracic CT

RATIONALE AND OBJECTIVES

An analysis for the optimum selection of image features in feature domain to represent lung nodules was performed, with implementation into a classification module of a computer-aided diagnosis system.

MATERIALS AND METHODS

Forty-two regions of interest obtained from 38 cases with effective diameters of 3 to 8.5 mm were used. On the basis of image characteristics and dimensionality, 11 features were computed. Nonparametric correlation coefficients, multiple regression analysis, and principal-component analysis were used to map the relation between the represented features from four radiologists and the computed features. An artificial neural network was used for the classification of benign and malignant nodules to test the hypothesis obtained from the mapping analysis.

RESULTS

Correlation coefficients ranging from 0.2693 to 0.5178 were obtained between the radiologists' annotations and the computed features. Of the 11 features used, three were found to be redundant when both nodule and non-nodule cases were used, and five were found redundant when nodule or non-nodule cases were used. Combination of analysis from correlation coefficients, regression analysis, principal-component analysis, and the artificial neural network resulted in the selection of optimum features to achieve F-test values of 0.821 and 0.643 for malignant and benign nodules, respectively.

CONCLUSION

This study demonstrates that for the optimum selection of features, each feature should be analyzed individually and collectively to evaluate the impact on the computer-aided diagnosis system on the basis of its class representation. This methodology will ultimately aid in improving the generalization capability of a classification module for early lung cancer diagnosis.

Automated breast mass detection in 3D reconstructed tomosynthesis volumes: a featureless approach

The purpose of this study was to propose and implement a computer aided detection (CADe) tool for breast tomosynthesis. This task was accomplished in two stages-a highly sensitive mass detector followed by a false positive (FP) reduction stage. Breast tomosynthesis data from 100 human subject cases were used, of which 25 subjects had one or more mass lesions and the rest were normal. For stage 1, filter parameters were optimized via a grid search. The CADe identified suspicious locations were reconstructed to yield 3D CADe volumes of interest. The first stage yielded a maximum sensitivity of 93% with 7.7 FPs/breast volume. Unlike traditional CADe algorithms in which the second stage FP reduction is done via feature extraction and analysis, instead information theory principles were used with mutual information as a similarity metric. Three schemes were proposed, all using leave-one-case-out cross validation sampling. The three schemes, A, B, and C, differed in the composition of their knowledge base of regions of interest (ROIs). Scheme A's knowledge base was comprised of all the mass and FP ROIs generated by the first stage of the algorithm. Scheme B had a knowledge base that contained information from mass ROIs and randomly extracted normal ROIs. Scheme C had information from three sources of information-masses, FPs, and normal ROIs. Also, performance was assessed as a function of the composition of the knowledge base in terms of the number of FP or normal ROIs needed by the system to reach optimal performance. The results indicated that the knowledge base needed no more than 20 times as many FPs and 30 times as many normal ROIs as masses to attain maximal performance. The best overall system performance was 85% sensitivity with 2.4 FPs per breast volume for scheme A, 3.6 FPs per breast volume for scheme B, and 3 FPs per breast volume for scheme C.

Eigendetection of masses considering false positive reduction and breast density information

The purpose of this article is to present a novel algorithm for the detection of masses in mammographic computer-aided diagnosis systems. Four key points provide the novelty of our approach: (1) the use of eigenanalysis for describing variation in mass shape and size; (2) a Bayesian detection methodology providing a mathematical sound framework, flexible enough to include additional information; (3) the use of a two-dimensional principal components analysis approach to facilitate false positive reduction; and (4) the incorporation of breast density information, a parameter correlated with the performance of most mass detection algorithms and which is not considered in existing approaches. To study the performance of the system two experiments were carried out. The first is related to the ability of the system to detect masses, and thus, free-response receiver operating characteristic analysis was used, showing that the method is able to give high accuracy at a high specificity (80% detection at 1.40 false positives per image). Second, the ability of the system to highlight the pixels belonging to a mass is studied using receiver operating characteristic analysis, resulting in A(z) = 0.89 +/- 0.04. In addition, the robustness of the approach is demonstrated in an experiment where we used the Digital Database for Screening Mammography database for training and the Mammographic Image Analysis Society database for testing the algorithm.

Computer-aided mass detection based on ipsilateral multiview mammograms

RATIONALE AND OBJECTIVES

Recent reports on advances in computer-aided detection (CAD) indicate that current schemes miss early-stage breast cancers and result in a relatively large false-positive detection rate in order to achieve a high sensitivity rate for mass detection. This paper is inspired by the interpretation procedure from mammographers. The abnormal diagnosis can be derived from multiple views but is not available through single-view image analysis.

MATERIALS AND METHODS

A new multiview CAD system for early-stage breast cancer detection, which is based on modifying the optimized CAD algorithms from our prior single-view CAD system for constructing an adaptive ipsilateral multiview concurrent CAD system, is presented in this paper. The selection and design for the training and testing ipsilateral multiview mammogram databases are described here.

RESULTS

The performance evaluation of the developed ipsilateral multiview CAD system using free-response receiver operating characteristic analysis and computerized receiver operating characteristic experiments are presented. The results indicated that the proposed multiview CAD system is significantly superior to the single-view CAD systems based on statistically standard P-values.

CONCLUSION

This paper addresses a very important and timely project. It is related to two main problems regarding the development of breast cancer detection and diagnosis: early-stage detection and diagnosis of breast cancer with digital mammogram, and overall improvement of CAD system performance for clinical implementation. In order to improve the efficacy, accuracy, and efficiency of the current CAD scheme, an entirely new class of CAD method is required. This paper is unique in that a comprehensive and state-of-the-art approach is proposed for the CAD scheme of digital mammography. From the design aspect of the CAD scheme, the proposed ipsilateral multiview CAD method is innovative and quite different from current single-view CAD methods.

Characterization of mammographic masses using a gradient-based segmentation algorithm and a neural classifier

Computerized methods have recently shown a great potential in providing radiologists with a second opinion about the visual diagnosis of the malignancy of mammographic masses. The computer-aided diagnosis (CAD) system we developed for the mass characterization is mainly based on a segmentation algorithm and on the neural classification of several features computed on the segmented mass. Mass-segmentation plays a key role in most computerized systems. Our technique is a gradient-based one, showing the main characteristic that no free parameters have been evaluated on the data set used in this analysis, thus it can directly be applied to data sets acquired in different conditions without any ad hoc modification. A data set of 226 masses (109 malignant and 117 benign) has been used in this study. The segmentation algorithm works with a comparable efficiency both on malignant and benign masses. Sixteen features based on shape, size and intensity of the segmented masses are extracted and analyzed by a multi-layered perceptron neural network trained with the error back-propagation algorithm. The capability of the system in discriminating malignant from benign masses has been evaluated in terms of the receiver-operating characteristic (ROC) analysis. A feature selection procedure has been carried out on the basis of the feature discriminating power and of the linear correlations interplaying among them. The comparison of the areas under the ROC curves obtained by varying the number of features to be classified has shown that 12 selected features out of the 16 computed ones are powerful enough to achieve the best classifier performances. The radiologist assigned the segmented masses to three different categories: correctly-, acceptably- and non-acceptably-segmented masses. We initially estimated the area under ROC curve only on the first category of segmented masses (the 88.5% of the data set), then extending the classification to the second subclass (reaching the 97.8% of the data set) and finally to the whole data set, obtaining A(z)=0.805+/-0.030, 0.787+/-0.024 and 0.780+/-0.023, respectively.

Breast MRI lesion classification: improved performance of human readers with a backpropagation neural network computer-aided diagnosis (CAD) system

PURPOSE

To develop and test a computer-aided diagnosis (CAD) system to improve the performance of radiologists in classifying lesions on breast MRI (BMRI).

MATERIALS AND METHODS

A CAD system was developed that uses a semiautomated segmentation method. After segmentation, 42 features based on lesion shape, texture, and enhancement kinetics were computed, and the 13 best features were selected and used as inputs to a backpropagation neural network (BNN). The BNN was trained and tested using the leave-one-out method on 80 BMRI lesions (37 benign, 43 malignant). Lesion histopathology was used as the reference standard. Five human readers classified the 80 lesions first without and then with CAD assistance. The performance of the computer classifier and the human readers was assessed using receiver operating characteristic curves; the performance of the human readers was also evaluated using multireader multicase (MRMC) analysis.

RESULTS

The performance of the human readers significantly improved when aided by the CAD system (P < 0.05). MRMC analysis showed that human reader performance with and without CAD system assistance can be generalized to the population of cases (P < 0.001).

CONCLUSION

A CAD system based on lesion morphology and enhancement kinetics can improve the performance of human readers in classifying lesions on breast MRI.

An incremental neural network for tissue segmentation in ultrasound images

This paper presents an incremental neural network (INeN) for the segmentation of tissues in ultrasound images. The performances of the INeN and the Kohonen network are investigated for ultrasound image segmentation. The elements of the feature vectors are individually formed by using discrete Fourier transform (DFT) and discrete cosine transform (DCT). The training set formed from blocks of 4x4 pixels (regions of interest, ROIs) on five different tissues designated by an expert is used for the training of the Kohonen network. The training set of the INeN is formed from randomly selected ROIs of 4x4 pixels in the image. Performances of both 2D-DFT and 2D-DCT are comparatively examined for the segmentation of ultrasound images.

Morphological concentric layer analysis for automated detection of suspicious masses in screening mammograms

Computer assisted detection systems (CAD) in mammography incorporate two critical stages: (i) prescreening to localize suspicious regions and (ii) detailed analysis of the regions for false positive reduction. In this work, we present a new technique for automatic detection of suspicious masses for prescreening mammograms. The hypothesis of the proposed technique is that malignant masses manifestate as superimposed concentric layers. Morphological characterization of these layers can form the foundation of an automated scheme for detection of suspicious masses. The study was based on fifty nine screening mammograms from the digital database of screening mammography (DDSM). Overall, the proposed scheme performs at 85.7% sensitivity with an average of 0.53 false positives per image. The scheme targets malignant masses and, thus it can provide a second opinion to radiologists without sending benign masses to unnecessary biopsy.

Characterization of difference of Gaussian filters in the detection of mammographic regions

In this article, we present a characterization of the effect of difference of Gaussians (DoG) filters in the detection of mammographic regions. DoG filters have been used previously in mammographic mass computer-aided detection (CAD) systems. As DoG filters are constructed from the subtraction of two bivariate Gaussian distributions, they require the specification of three parameters: the size of the filter template and the standard deviations of the constituent Gaussians. The influence of these three parameters in the detection of mammographic masses has not been characterized. In this work, we aim to determine how the parameters affect (1) the physical descriptors of the detected regions, (2) the true and false positive rates, and (3) the classification performance of the individual descriptors. To this end, 30 DoG filters are created from the combination of three template sizes and four values for each of the Gaussians' standard deviations. The filters are used to detect regions in a study database of 181 craniocaudal-view mammograms extracted from the Digital Database for Screening Mammography. To describe the physical characteristics of the identified regions, morphological and textural features are extracted from each of the detected regions. Differences in the mean values of the features caused by altering the DoG parameters are examined through statistical and empirical comparisons. The parameters' effects on the true and false positive rate are determined by examining the mean malignant sensitivities and false positives per image (FPpI). Finally, the effect on the classification performance is described by examining the variation in FPpI at the point where 81% of the malignant masses in the study database are detected. Overall, the findings of the study indicate that increasing the standard deviations of the Gaussians used to construct a DoG filter results in a dramatic decrease in the number of regions identified at the expense of missing a small number of malignancies. The sharp reduction in the number of identified regions allowed the identification of textural differences between large and small mammographic regions. We find that the classification performances of the features that achieve the lowest average FPpI are influenced by all three of the parameters.

Dual system approach to computer-aided detection of breast masses on mammograms

In this study, our purpose was to improve the performance of our mass detection system by using a new dual system approach which combines a computer-added detection (CAD) system optimized with "average" masses with another CAD system optimized with "subtle" masses. The two single CAD systems have similar image processing steps, which include prescreening, object segmentation, morphological and texture feature extraction, and false positive (FP) reduction by rule-based and linear discriminant analysis (LDA) classifiers. A feed-forward backpropagation artificial neural network was trained to merge the scores from the LDA classifiers in the two single CAD systems and differentiate true masses from normal tissue. For an unknown test mammogram, the two single CAD systems are applied to the image in parallel to detect suspicious objects. A total of three data sets were used for training and testing the systems. The first data set of 230 current mammograms, referred to as the average mass set, was collected from 115 patients. We also collected 264 mammograms, referred to as the subtle mass set, which were one to two years prior to the current exam from these patients. Both the average and the subtle mass sets were partitioned into two independent data sets in a cross validation training and testing scheme. A third data set containing 65 cases with 260 normal mammograms was used to estimate the FP marker rates during testing. When the single CAD system trained on the average mass set was applied to the test set with average masses, the FP marker rates were 2.2, 1.8, and 1.5 per image at the case-based sensitivities of 90%, 85%, and 80%, respectively. With the dual CAD system, the FP marker rates were reduced to 1.2, 0.9, and 0.7 per image, respectively, at the same case-based sensitivities. Statistically significant (p < 0.05) improvements on the free response receiver operating characteristic curves were observed when the dual system and the single system were compared using the test sets with either average masses or subtle masses.

Textural assessment in digital mammograms

This work focuses on testing textural and morphological parameters to assess characteristic features of digital mammograms. The selected images were radiological studies from the Institute Nacional de Cancerologia in Mexico City, evaluated as BI-RADS 4 or 5, meaning "probably malign" or "malign" findings, respectively. All patients were subjected to a biopsy procedure after the image was taken. The study group consisted of patients diagnosed with cancer, while the control group included those without cancer. We propose to analyze textural roughness by the mean height-width ratio of extrema (MHWRE) and morphological features by circularity. Results show good differentiation (correct diagnosis) for 46% of the images, bad differentiation (wrong diagnosis) for 25%, and undetermined diagnosis for 29% of the cases.

CAD-aided mammogram training

RATIONALE AND OBJECTIVES

Although computer-aided detection (CAD) improves the diagnosis rate of early breast cancer, it has not been well integrated into radiology residency and technician training program. Moreover, CAD performance studies ignore the reader's training and experience with CAD. The purpose of this study was to investigate whether CAD training via a cognitive-perceptual based hypermedia program has effects on the performance studies of mammogram reading.

MATERIALS AND METHODS

Three observers read a pretest set of 80 breast cancer cases (43 negative, 23 benign, and 14 malignant cancer cases). During 4 weeks' training, the observers used a hypermedia instructional program in CAD-aided mammography interpretation. The program includes modules of CAD attention-focusing schemes, CAD procedural knowledge, and case-based simulations in mammography interpretation in consensus with CAD. By the end of the fourth week of the training, they reviewed a posttest set of cases. Data were analyzed with multireader, multicase receiver operating characteristic methods.

RESULTS

Three readers performed better in mammogram reading after training in CAD knowledge than they did before CAD training. CAD training and experience improved the performance of CAD-aided mammography interpretation.

CONCLUSION

A statistically significant difference was found in each observer's performance in CAD-aided mammogram reading before and after the training. CAD training will influence the perception, recognition, and interpretation of early breast cancer and CAD performance studies. Furthermore, the young generation of radiologic professionals can have more training in various attention-focusing features, declarative knowledge, procedural knowledge, and conditional knowledge of CAD and incorporate them into their knowledge base and strategic processing for the purpose of improving the accuracy of mammography interpretation performance.

Evaluating the effect of a wavelet enhancement method in characterization of simulated lesions embedded in dense breast parenchyma

Presence of dense parenchyma in mammographic images masks lesions resulting in either missed detections or mischaracterizations, thus decreasing mammographic sensitivity and specificity. The aim of this study is evaluating the effect of a wavelet enhancement method on dense parenchyma for a lesion contour characterization task, using simulated lesions. The method is recently introduced, based on a two-stage process, locally adaptive denoising by soft-thresholding and enhancement by linear stretching. Sixty simulated low-contrast lesions of known image characteristics were generated and embedded in dense breast areas of normal mammographic images selected from the DDSM database. Evaluation was carried out by an observer performance comparative study between the processed and initial images. The task for four radiologists was to classify each simulated lesion with respect to contour sharpness/unsharpness. ROC analysis was performed. Combining radiologists' responses, values of the area under ROC curve (Az) were 0.93 (95% CI 0.89, 0.96) and 0.81 (CI 0.75, 0.86) for processed and initial images, respectively. This difference in Az values was statistically significant (Student's t-test, P<0.05), indicating the effectiveness of the enhancement method. The specific wavelet enhancement method should be tested for lesion contour characterization tasks in softcopy-based mammographic display environment using naturally occurring pathological lesions and normal cases.

Ipsilateral-mammogram computer-aided detection of breast cancer

In this paper, an ipsilateral multi-view computer-aided detection (CAD) scheme is presented for mass detection in digital mammograms by exploiting correlative information of suspicious lesions between mammograms of the same breast. After nonlinear tree-structured filtering for image noise suppression, two wavelet-based methods, directional wavelet transform and tree-structured wavelet transform for image enhancement, and adaptive fuzzy C-means algorithm for segmentation are employed on each mammograms of the same breast, respectively, concurrent analysis is developed for iterative analysis of ipsilateral multi-view mammograms by inter-projective feature matching analysis. A supervised artificial neural network is developed as a classifier, in which the back-propagation algorithm combined with Kalman filtering is used as training algorithm, and free-response receiver operating characteristic analysis is used to test the performance of the developed unilateral CAD system. Performance comparison has been conducted between the final ipsilateral multi-view CAD system and our previously developed single-mammogram-based CAD system. The study results demonstrate the advantages of ipsilateral multi-view CAD method combined with concurrent analysis over current single-view CAD system on false positive reduction.

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.

Computer aided diagnosis of breast cancer in digitized mammograms

A computer aided neural network classification of regions of suspicion (ROS) on digitized mammograms is presented in this paper which employs features extracted by a new technique based on independent component analysis. Our approach is concentrated in finding a set of independent source regions that generate the observed mammograms. The coefficients of the linear transformation of the source regions are used as features that describe effectively any normal and abnormal region in digital mammograms as well as benign and malignant ROS in the latter case. Extensive experiments in the MIAS Database have shown a recognition accuracy of 88.23% in the detection of all kinds of abnormalities and 79.31% in the task of distinguishing between benign and malignant regions, outperforming in both cases standard textural features, widely used for cancer detection in mammograms.

Wavelets in temporal and spatial processing of biomedical images

We review some of the most recent advances in the area of wavelet applications in medical imaging. We first review key concepts in the processing of medical images with wavelet transforms and multiscale analysis, including time-frequency tiling, overcomplete representations, higher dimensional bases, symmetry, boundary effects, translational invariance, orientation selectivity, and best-basis selection. We next describe some applications in magnetic resonance imaging, including activation detection and denoising of functional magnetic resonance imaging and encoding schemes. We then present an overview in the area of ultrasound, including computational anatomy with three-dimensional cardiac ultrasound. Next, wavelets in tomography are reviewed, including their relationship to the radon transform and applications in position emission tomography imaging. Finally, wavelet applications in digital mammography are reviewed, including computer-assisted diagnostic systems that support the detection and classification of small masses and methods of contrast enhancement.

Digital mammography: wavelet transform and Kalman-filtering neural network in mass segmentation and detection

RATIONALE AND OBJECTIVES

The authors developed a new adaptive module to improve their computer-assisted diagnostic (CAD) method for mass segmentation and classification. The goal was an adaptive module that used a novel four-channel wavelet transform with neural network rather than a two-channel wavelet transform with manual subimage selection. The four-channel wavelet transform is used for image decomposition and reconstruction, and a novel Kalman-filtering neural network is used for adaptive subimage selection.

MATERIALS AND METHODS

The adaptive CAD module was compared with the nonadaptive module by comparing receiver operating characteristic curves for the whole CAD system. An image database containing 800 regions of interest enclosing all mass types and normal tissues was used for the relative comparison of system performance, with electronic ground truth established in advance.

RESULTS

The receiver operating characteristic curves yield Az values of 0.93 and 0.86 with and without the adaptive module respectively, suggesting that overall CAD performance is improved with the adaptive module.

CONCLUSION

The results of this study confirm the importance of using a new class of adaptive CAD methods that allow a more generalized application for larger image databases or images generated from different sensors or by means of direct x-ray detection, as required for clinical trials.

Digital mammography: comparison of adaptive and nonadaptive CAD methods for mass detection

RATIONALE AND OBJECTIVES

The authors compared the performance of adaptive and nonadaptive computer-aided diagnostic (CAD) methods for breast mass detection with digital mammography.

MATERIALS AND METHODS

Both adaptive and nonadaptive modular CAD methods employed recent advances in multiresolution and mutiorientation wavelet transforms for improved feature extraction. The nonadaptive method uses fixed parameters for the image preprocessing modules. The adaptive method, a new class of algorithms, adapts to image content by selecting parameters for the image preprocessing modules within a parameter range. Comparison of the two methods was performed for each individual CAD module with a region-of-interest (ROI) database containing all mass types and normal tissue.

RESULTS

Receiver operating characteristic (ROC) analysis clearly demonstrated an improvement in performance for the three adaptive modules and a significant overall difference between the two methods. The average ROC area index (Az) values were 0.86 and 0.95 for the nonadaptive and adaptive methods, respectively. The corresponding P value is .0145. For a previously reported database of full mammographic images containing 50 abnormal cases with all mass types and 50 normal images, the adaptive CAD method had a sensitivity of 96% (1.71 false-positive results per image) compared with 89% (1.91 false-positive results per image) for the nonadaptive CAD method.

CONCLUSION

The adaptive CAD method demonstrated better performance. A study is in progress to determine the generalizability of the adaptive CAD method by applying it to larger retrospective image databases with different film digitizers.