The recent advancement in deep learning-based approaches vastly outperforms the traditional image descriptors. Deep learning models, such as residual networks (ResNet), are well known for finding salient features. Although effective, high-level description often has a high dimensionality that increases computational overhead. The autoencoders find the useful approximation of the input data without losing critical information. Considering this, we propose a content-based image retrieval system for natural color images using a deep stacked sparse autoencoder (DSSA). The DSSA model learns latent features in an unsupervised way from the high-level description obtained using ResNet. The DSSA model achieves a nearly 50% reduction in size compared with the full-length features for the simple distance-based retrieval approach while increasing accuracy. The image retrieval efficacy of the learned latent features is also evaluated for two classifier-based methods using a Softmax classifier. Further, this study investigates the impact of unsupervised feature learning on retrieval using three benchmark natural color image databases of varying complexities, viz., Corel-1K, Corel-10K, and Canadian Institute for Advanced Research (CIFAR)-10. The latent features learned by the DSSA model with the fuzzy class membership-based retrieval method achieve promising improvements and yield a highly competitive retrieval performance with the large-size CIFAR-10 database.
We describe the improvements of the content-based image retrieval (CBIR) system using a fuzzy class membership for the natural-color images. The fuzzy class membership-based retrieval (CMR) framework has shown promising improvements on texture databases by exploiting confidence in classification using a multilayer perceptron (MLP). CMR is known to improve the average precision of retrieval along with modest variance, and the framework is not restricted to any particular feature set. However, their efficacy is not known for natural colored images. In the proposed approach, we have added a new classifier, radial basis function network, in place of MLP in the CMR framework. We show a way to adapt a new classifier in the fuzzy CMR framework. Comparison with state-of-the-art CBIR systems shows that the proposed modifications have an edge over its competition in terms of precision for four popular image databases: viz. Corel-1k, Corel-5k, Corel-10k, and CIFAR-10.
Lung field segmentation is a prerequisite step for the development of a computer-aided diagnosis system for interstitial lung diseases observed in chest HRCT images. Conventional methods of lung field segmentation rely on a large gray value contrast between lung fields and surrounding tissues. These methods fail on lung HRCT images with dense and diffused pathology. An efficient prepro- cessing could improve the accuracy of segmentation of pathological lung field in HRCT images. In this paper, a convolution neural network is used for localization of lung fields in HRCT images. The proposed method provides an optimal bounding box enclosing the lung fields irrespective of the presence of diffuse pathology. The performance of the proposed algorithm is validated on 330 lung HRCT images obtained from MedGift database on ZF and VGG networks. The model achieves a mean average precision of 0.94 with ZF net and a slightly better performance giving a mean average precision of 0.95 in case of VGG net.
Mammography is the most efficient modality for detection of breast cancer at early stage. Microcalcifications are tiny bright spots in mammograms and can often get missed by the radiologist during diagnosis. The presence of microcalcification clusters in mammograms can act as an early sign of breast cancer. This paper presents a completely automated computer-aided detection (CAD) system for detection of microcalcification clusters in mammograms. Unsharp masking is used as a preprocessing step which enhances the contrast between microcalcifications and the background. The preprocessed image is thresholded and various shape and intensity based features are extracted. Support vector machine (SVM) classifier is used to reduce the false positives while preserving the true microcalcification clusters. The proposed technique is applied on two different databases i.e DDSM and private database. The proposed technique shows good sensitivity with moderate false positives (FPs) per image on both databases.
In this paper, retrieval accuracy of different types of pulmonary nodules is studied. The trainee radiologists could enrich their knowledge using the visual information of the retrieved nodules. In the proposed retrieval system, the pulmonary nodules are segmented using a semi-automated technique. Several 3D features are explored to improve the performance of the proposed retrieval system. A set of relevant shape and texture features is determined for efficient representation of the nodules in the feature space. The proposed CBIR system is validated on a data set of 542 nodules of Lung Image Database Consortium and Image Database Resource Initiative (LIDC/IDRI). The nodules with composite rank of malignancy “1”, “2” are considered as benign and “4”, “5” are considered as malignant. Considering top five retrieved images, the precision of the proposed retrieval system are 84.76%, 80.75%, and 80.34% for well-circumscribed, juxta-pleural, and juxtavascular nodules, respectively.
This manuscript presents an analytical treatment on the feasibility of multi-scale Gabor filter bank response for non-invasive oral cancer pre-screening and detection in the long infrared spectrum. Incapability of present healthcare technology to detect oral cancer in budding stage manifests in high mortality rate. The paper contributes a step towards automation in non-invasive computer-aided oral cancer detection using an amalgamation of image processing and machine intelligence paradigms. Previous works have shown the discriminative difference of facial temperature distribution between a normal subject and a patient. The proposed work, for the first time, exploits this difference further by representing the facial Region of Interest(ROI) using multiscale rotation invariant Gabor filter bank responses followed by classification using Radial Basis Function(RBF) kernelized Support Vector Machine(SVM). The proposed study reveals an initial increase in classification accuracy with incrementing image scales followed by degradation of performance; an indication that addition of more and more finer scales tend to embed noisy information instead of discriminative texture patterns. Moreover, the performance is consistently better for filter responses from profile faces compared to frontal faces.This is primarily attributed to the ineptness of Gabor kernels to analyze low spatial frequency components over a small facial surface area. On our dataset comprising of 81 malignant, 59 pre-cancerous, and 63 normal subjects, we achieve state-of-the-art accuracy of 85.16% for normal v/s precancerous and 84.72% for normal v/s malignant classification. This sets a benchmark for further investigation of multiscale feature extraction paradigms in IR spectrum for oral cancer detection.
In this paper, we investigate the effect of the error criteria used during a training phase of the artificial neural network (ANN) on the accuracy of the classifier for classification of lung tissues affected with Interstitial Lung Diseases (ILD). Mean square error (MSE) and the cross-entropy (CE) criteria are chosen being most popular choice in state-of-the-art implementations. The classification experiment performed on the six interstitial lung disease (ILD) patterns viz. Consolidation, Emphysema, Ground Glass Opacity, Micronodules, Fibrosis and Healthy from MedGIFT database. The texture features from an arbitrary region of interest (AROI) are extracted using Gabor filter. Two different neural networks are trained with the scaled conjugate gradient back propagation algorithm with MSE and CE error criteria function respectively for weight updation. Performance is evaluated in terms of average accuracy of these classifiers using 4 fold cross-validation. Each network is trained for five times for each fold with randomly initialized weight vectors and accuracies are computed. Significant improvement in classification accuracy is observed when ANN is trained by using CE (67.27%) as error function compared to MSE (63.60%). Moreover, standard deviation of the classification accuracy for the network trained with CE (6.69) error criteria is found less as compared to network trained with MSE (10.32) criteria.
Automated face detection is the pivotal step in computer vision aided facial medical diagnosis and biometrics.
This paper presents an automatic, subject adaptive framework for accurate face detection in the long infrared
spectrum on our database for oral cancer detection consisting of malignant, precancerous and normal subjects
of varied age group. Previous works on oral cancer detection using Digital Infrared Thermal Imaging(DITI)
reveals that patients and normal subjects differ significantly in their facial thermal distribution. Therefore, it is
a challenging task to formulate a completely adaptive framework to veraciously localize face from such a subject
specific modality. Our model consists of first extracting the most probable facial regions by minimum error
thresholding followed by ingenious adaptive methods to leverage the horizontal and vertical projections of the
segmented thermal image. Additionally, the model incorporates our domain knowledge of exploiting temperature
difference between strategic locations of the face. To our best knowledge, this is the pioneering work on detecting
faces in thermal facial images comprising both patients and normal subjects. Previous works on face detection
have not specifically targeted automated medical diagnosis; face bounding box returned by those algorithms are
thus loose and not apt for further medical automation. Our algorithm significantly outperforms contemporary
face detection algorithms in terms of commonly used metrics for evaluating face detection accuracy. Since our
method has been tested on challenging dataset consisting of both patients and normal subjects of diverse age
groups, it can be seamlessly adapted in any DITI guided facial healthcare or biometric applications.
Histopathology is considered the gold standard for oral cancer detection. But a major fraction of patient pop- ulation is incapable of accessing such healthcare facilities due to poverty. Moreover, such analysis may report false negatives when test tissue is not collected from exact cancerous location. The proposed work introduces a pioneering computer aided paradigm of fast, non-invasive and non-ionizing modality for oral cancer detection us- ing Digital Infrared Thermal Imaging (DITI). Due to aberrant metabolic activities in carcinogenic facial regions, heat signatures of patients are different from that of normal subjects. The proposed work utilizes asymmetry of temperature distribution of facial regions as principle cue for cancer detection. Three views of a subject, viz. front, left and right are acquired using long infrared (7:5 - 13μm) camera for analysing distribution of temperature. We study asymmetry of facial temperature distribution between: a) left and right profile faces and b) left and right half of frontal face. Comparison of temperature distribution suggests that patients manifest greater asymmetry compared to normal subjects. For classification, we initially use k-means and fuzzy k-means for unsupervised clustering followed by cluster class prototype assignment based on majority voting. Average classification accuracy of 91:5% and 92:8% are achieved by k-mean and fuzzy k-mean framework for frontal face. The corresponding metrics for profile face are 93:4% and 95%. Combining features of frontal and profile faces, average accuracies are increased to 96:2% and 97:6% respectively for k-means and fuzzy k-means framework.
Differentiation of malignant and benign pulmonary nodules is important for prognosis of lung cancer. In this paper, benign and malignant nodules are classified using support vector machine. Several shape-based and texture-based features are used to represent the pulmonary nodules in the feature space. A semi-automated technique is used for nodule segmentation. Relevant features are selected for efficient representation of nodules in the feature space. The proposed scheme and the competing technique are evaluated on a data set of 542 nodules of Lung Image Database Consortium and Image Database Resource Initiative. The nodules with composite rank of malignancy “1",”2" are considered as benign and “4",”5" are considered as malignant. Area under the receiver operating characteristics curve is 0:9465 for the proposed method. The proposed method outperforms the competing technique.
Pulmonary nodules are a potential manifestation of lung cancer, and their early detection can remarkably enhance the survival rate of patients. This paper presents an automated pulmonary nodule detection algorithm for lung CT images. The algorithm utilizes a two-stage approach comprising nodule candidate detection followed by reduction of false positives. The nodule candidate detection involves thresholding, followed by morphological opening. The geometrical features at this stage are selected from properties of nodule size and compactness, and lead to reduced number of false positives. An SVM classifier is used with a radial basis function kernel. The data imbalance, due to uneven distribution of nodules and non-nodules as a result of the candidate detection stage, is proposed to be addressed by oversampling of minority class using Synthetic Minority Over-sampling Technique (SMOTE), and over-imposition of its misclassification penalty. Experiments were performed on 97 CT scans of a publically-available (LIDC-IDRI) database. Performance is evaluated in terms of sensitivity and false positives per scan (FP/scan). Results indicate noteworthy performance of the proposed approach (nodule detection sensitivity after 4-fold cross-validation is 92.91% with 3 FP/scan). Comparative analysis also reflects a comparable and often better performance of the proposed setup over some of the existing techniques.
Interstitial lung disease (ILD) is complicated group of pulmonary disorders. High Resolution Computed Tomography (HRCT) considered to be best imaging technique for analysis of different pulmonary disorders. HRCT findings can be categorised in several patterns viz. Consolidation, Emphysema, Ground Glass Opacity, Nodular, Normal etc. based on their texture like appearance. Clinician often find it difficult to diagnosis these pattern because of their complex nature. In such scenario computer-aided diagnosis system could help clinician to identify patterns. Several approaches had been proposed for classification of ILD patterns. This includes computation of textural feature and training /testing of classifier such as artificial neural network (ANN), support vector machine (SVM) etc. In this paper, wavelet features are calculated from two different ILD database, publically available MedGIFT ILD database and private ILD database, followed by performance evaluation of ANN and SVM classifiers in terms of average accuracy. It is found that average classification accuracy by SVM is greater than ANN where trained and tested on same database. Investigation continued further to test variation in accuracy of classifier when training and testing is performed with alternate database and training and testing of classifier with database formed by merging samples from same class from two individual databases. The average classification accuracy drops when two independent databases used for training and testing respectively. There is significant improvement in average accuracy when classifiers are trained and tested with merged database. It infers dependency of classification accuracy on training data. It is observed that SVM outperforms ANN when same database is used for training and testing.
Automated segmentation of pathological bearing region is the first step towards the development of lung CAD. Most of the work reported in the literature related to automated analysis of lung tissue aims towards classification of fixed sized block into one of the classes. This block level classification of lung tissues in the image never results in accurate or smooth boundaries between different regions. In this work, effort is taken to investigate the performance of three automated image segmentation algorithms those results in smooth boundaries among lung tissue patterns commonly encountered in HRCT images of the thorax. A public database that consists of HRCT images taken from patients affected with Interstitial Lung Diseases (ILDs) is used for the evaluation. The algorithms considered are Markov Random Field (MRF), Gaussian Mixture Model (GMM) and Mean Shift (MS). 2-fold cross validation approach is followed for the selection of the best parameter value for individual algorithm as well as to evaluate the performance of all the algorithms. Mean shift algorithm is observed as the best performer in terms of Jaccard Index, Modified Hausdorff Distance, accuracy, Dice Similarity Coefficient and execution speed.
Diabetic retinopathy is a condition of the eye of diabetic patients where the retina is damaged because of long-term diabetes. The condition deteriorates towards irreversible blindness in extreme cases of diabetic retinopathy. Hence, early detection of diabetic retinopathy is important to prevent blindness. Regular screening of fundus images of diabetic patients could be helpful in preventing blindness caused by diabetic retinopathy. In this paper, we propose techniques for staging of diabetic retinopathy in fundus images using several shape and texture features computed from detected microaneurysms, exudates, and hemorrhages. The classification accuracy is reported in terms of the area (Az) under the receiver operating characteristic curve using 200 fundus images from the MESSIDOR database. The value of Az for classifying normal images versus mild, moderate, and severe nonproliferative diabetic retinopathy (NPDR) is 0:9106. The value of Az for classification of mild NPDR versus moderate and severe NPDR is 0:8372. The Az value for classification of moderate NPDR and severe NPDR is 0:9750.
This paper evaluates the performance of recently proposed rotation invariant texture feature extraction method for the classi¯cation and retrieval of lung tissues a®ected with Interstitial Lung Diseases (ILDs). The method makes use of principle texture direction as the reference direction and extracts texture features using Discrete Wavelet Transform (DWT). A private database containing high resolution computed tomography (HRCT) images belonging to ¯ve category of lung tissue is used for the experiment. The experimental result shows that the texture appearances of lung tissues are anisotropic in nature and hence rotation invariant features achieve better retrieval as well as classi¯cation accuracy.
KEYWORDS: Databases, Image retrieval, Cardiovascular magnetic resonance imaging, Medical imaging, Medical research, Neural networks, Imaging systems, Electronics, Telecommunications, Communication engineering
Several single valued measures have been proposed by researchers for the quantitative performance evaluation of medical image retrieval systems. Precision and recall are the most common evaluation measures used by researchers. Amongst graphical measures proposed, precision vs. recall graph is the most common evaluation measure. Precision vs. recall graph evaluates di®erent systems by varying the operating points (number of top retrieval considered). However, in real life the operating point for di®erent applications are known. Therefore, it is essential to evaluate di®erent retrieval systems at a particular operating point set by the user. None of the graphical metric provides the variation of performance of query images over the entire database at a particular operating point. This paper proposes a graphical metric called Complementary Cumulative Precision Distribution (CCPD) that evaluates di®erent systems at a particular operating point considering each images in the database for query. The strength of the metric is its ability to represent all these measures pictorially. The proposed metric (CCPD) pictorially represents the di®erent possible values of precision and the fraction of query images at those precision values considering number of top retrievals constant. Di®erent scalar measures are derived from the proposed graphical metric (CCPD) for e®ective evaluation of retrieval systems. It is also observed that the proposed metric can be used as a tie breaker when the performance of di®erent methods are very close to each other in terms of average precision.
Lung field segmentation is a prerequisite for development of automated computer aided diagnosis system from chest computed tomography (CT) scans. Intensity based algorithm such as mean shift (MS) segmentation on CT images for delineation of lung field is reported as the best technique in terms of accuracy and speed in the literature. However, in presence of high dense abnormalities, accurate and automated delineation of lung field becomes difficult. So an improved lung field segmentation using mean shift clustering followed by geometric property based techniques such as lung region of interest (ROI) created from symmetric centroid map of two normal subjects, false positives (FP) reduction module (using eccentricity, solidity, area, centroid features) and false negatives (FN) reduction module (using overlap feature between clusters from MS label map and convex hull of costal lung) is proposed. The performance of the proposed algorithm is validated on images obtained from Lung Image Database Consortium (LIDC) - Image Database Resource Initiative (IDRI) public database of 17 subjects containing nodular patterns and from local database of 26 subjects containing interstitial lung disease (ILD) patterns. The proposed algorithm has achieved mean Modified Hausdorff Distance (MHD) in mm of 1.47 ± 4.31, Dice Similarity Coefficient (DSC) of 0.9854 ± 0.0288, sensitivity of 0.9771 ± 0.0433, specificity of 0.9991 ± 0.0014 for 133 normal images from 32 subjects and MHD in mm of 6.23 ± 9.00, DSC of 0.8954 ± 0.1498, sensitivity of 0.8468 ± 0.1908, specificity of 0.9969 ± 0.0061 for 296 abnormal images from 43 subjects.
The nipple is an important landmark in mammograms. Detection of the nipple is useful for alignment and registration of mammograms in computer-aided diagnosis of breast cancer. In this paper, a novel approach is proposed for automatic detection of the nipple based on the oriented patterns of the breast tissues present in mammograms. The Radon transform is applied to the oriented patterns obtained by a bank of Gabor filters to detect the linear structures related to the tissue patterns. The detected linear structures are then used to locate the nipple position using the characteristics of convergence of the tissue patterns towards the nipple. The performance of the method was evaluated with 200 scanned-film images from the mini-MIAS database and 150 digital radiography (DR) images from a local database. Average errors of 5:84 mm and 6:36 mm were obtained with respect to the reference nipple location marked by a radiologist for the mini-MIAS and the DR images, respectively.
Content Based Image Retrieval (CBIR) system could exploit the wealth of High-Resolution Computed Tomography (HRCT) data stored in the archive by finding similar images to assist radiologists for self learning and differential diagnosis of Interstitial Lung Diseases (ILDs). HRCT findings of ILDs are classified into several categories (e.g. consolidation, emphysema, ground glass, nodular etc.) based on their texture like appearances. Therefore, analysis of ILDs is considered as a texture analysis problem. Many approaches have been proposed for CBIR of lung images using texture as primitive visual content. This paper presents a new approach to CBIR for ILDs. The proposed approach makes use of a trained neural network (NN) to find the output class label of query image. The degree of confidence of the NN classifier is analyzed using Naive Bayes classifier that dynamically takes a decision on the size of the search space to be used for retrieval. The proposed approach is compared with three simple distance based and one classifier based texture retrieval approaches. Experimental results show that the proposed technique achieved highest average percentage precision of 92.60% with lowest standard deviation of 20.82%.
In this paper a differential geometry based method is proposed for calculating surface speculation of solitary pulmonary nodule (SPN) in 3D from lung CT images. Spiculation present in SPN is an important shape feature to assist radiologist for measurement of malignancy. Performance of Computer Aided Diagnostic (CAD) system depends on the accurate estimation of feature like spiculation. In the proposed method, the peak of the spicules is identified using the property of Gaussian and mean curvature calculated at each surface point on segmented SPN. Once the peak point for a particular SPN is identified, the nearest valley points for the corresponding peak point are determined. The area of cross-section of the best fitted plane passing through the valley points is the base of that spicule. The solid angle subtended by the base of spicule at peak point and the distance of peak point from nodule base are taken as the measures of spiculation. The speculation index (SI) for a particular SPN is the weighted combination of all the spicules present in that SPN. The proposed method is validated on 95 SPN from Imaging Database Resources Initiative (IDRI) public database. It has achieved 87.4% accuracy in calculating quantified spiculation index compared to the spiculation index provided by radiologists in IDRI database.
In this paper we have investigated a new approach for texture features extraction using co-occurrence matrix from volumetric lung CT image. Traditionally texture analysis is performed in 2D and is suitable for images collected from 2D imaging modality. The use of 3D imaging modalities provide the scope of texture analysis from 3D object and 3D texture feature are more realistic to represent 3D object. In this work, Haralick's texture features are extended in 3D and computed from volumetric data considering 26 neighbors. The optimal texture features to characterize the internal structure of Solitary Pulmonary Nodules (SPN) are selected based on area under curve (AUC) values of ROC curve and p values from 2-tailed Student's t-test. The selected texture feature in 3D to represent SPN can be used in efficient Computer Aided Diagnostic (CAD) design plays an important role in fast and accurate lung cancer screening. The reduced number of input features to the CAD system will decrease the computational time and classification errors caused by irrelevant features. In the present work, SPN are classified from Ground Glass Nodule (GGN) using Artificial Neural Network (ANN) classifier considering top five 3D texture features and top five 2D texture features separately. The classification is performed on 92 SPN and 25 GGN from Imaging Database Resources Initiative (IDRI) public database and classification accuracy using 3D texture features and 2D texture features provide 97.17% and 89.1% respectively.
Architectural distortion is an important sign of early breast cancer. Due to its subtlety, it is often missed during screening. We propose a method to detect architectural distortion in prior mammograms of interval-cancer cases based on statistical measures of oriented patterns. Oriented patterns were analyzed in the present work because regions with architectural distortion contain a large number of tissue structures spread over a wide angular range. Two new types of cooccurrence matrices were derived to estimate the joint occurrence of the angles of oriented structures. Statistical features were computed from each of the angle cooccurrence matrices to discriminate sites of architectural distortion from falsely detected regions in normal parts of mammograms. A total of 4,224 regions of interest (ROIs) were automatically obtained from 106 prior mammograms of 56 interval-cancer cases and 52 mammograms of 13 normal cases with the application of Gabor filters and phase portrait analysis. For each ROI, Haralick's 14 features were computed using the angle cooccurrence matrices. The best result obtained in terms of the area under the receiver operating characteristic (ROC) curve with the leave-one-patient-out method was 0.76; the free-response ROC curve indicated a sensitivity of 80% at 4.2 false positives per patient.
The computer aided diagnostic (CAD) system has been developed to assist radiologist for early
detection and analysis of lung nodules. For pulmonary nodule detection, image preprocessing is
required to remove the anatomical structure of lung parenchyma and to enhance the visibility of
pulmonary nodules. In this paper a hybrid preprocessing technique using geometry based diffusion
and selective enhancement filtering have been proposed. This technique provides a unified preprocessing
framework for solid nodule as well as ground glass opacity (GGO) nodules. Geometry
based diffusion is applied to smooth the images by preserving the boundary. In order to improve
the sensitivity of pulmonary nodule detection, selective enhancement filter is used to highlight blob
like structure. But selective enhancement filter sometimes enhances the structures like blood vessel
and airways other than nodule and results in large number of false positive. In first step, geometry
based diffusion (GBD) is applied for reduction of false positive and in second step, selective
enhancement filtering is used for further reduction of false negative. Geometry based diffusion and
selective enhancement filtering has been used as preprocessing step separately but their combined
effect was not investigated earlier. This hybrid preprocessing approach is suitable for accurate calculation
of voxel based features. The proposed method has been validated on one public database
named Lung Image Database Consortium (LIDC) containing 50 nodules (30 solid and 20 GGO
nodule) from 30 subjects and one private database containing 40 nodules (25 solid and 15 GGO
nodule) from 30 subjects.
We present a method using statistical measures of the orientation of texture to characterize and detect architectural
distortion in prior mammograms of interval-cancer cases. Based on the orientation field, obtained by
the application of a bank of Gabor filters to mammographic images, two types of co-occurrence matrices were
derived to estimate the joint occurrence of the angles of oriented structures. For each of the matrices, Haralick's
14 texture features were computed. From a total of 106 prior mammograms of 56 interval-cancer cases and
52 mammograms of 13 normal cases, 4,224 regions of interest (ROIs) were automatically obtained by applying
Gabor filters and phase portrait analysis. For each ROI, statistical features were computed using the angle
co-occurrence matrices. The performance of the features in the detection of architectural distortion was analyzed
and compared with that of Haralick's features computed using the gray-level co-occurrence matrices of
the ROIs. Using logistic regression for feature selection, an artificial neural network for classification, and the
leave-one-image-out approach for cross-validation, the best result achieved was 0.77 in terms of the area under
the receiver operating characteristic (ROC) curve. Analysis of the free-response ROC curve yielded a sensitivity
of 80% at 5.4 false positives per image.
The existing image compression standards like JPEG and JPEG 2000, compress the whole image as a single frame. This makes the system simple but inefficient. The problem is acute for applications where lossless compression is mandatory viz. medical image compression. If the spatial characteristics of the image are considered, it can give rise to a more efficient coding scheme. For example, CT reconstructed images have uniform background outside the field of view (FOV). Even the portion within the FOV can be divided as anatomically relevant and irrelevant parts. They have distinctly different statistics. Hence coding them separately will result in more efficient compression. Segmentation is done based on thresholding and shape information is stored using 8-connected differential chain code. Simple 1-D DPCM is used as the prediction scheme. The experiments show that the 1st order entropies of images fall by more than 11% when each segment is coded separately. For simplicity and speed of decoding Huffman code is chosen for entropy coding. Segment based coding will have an overhead of one table per segment but the overhead is minimal. Lossless compression of image based on segmentation resulted in reduction of bit rate by 7%-9% compared to lossless compression of whole image as a single frame by the same prediction coder. Segmentation based scheme also has the advantage of natural ROI based progressive decoding. If it is allowed to delete the diagnostically irrelevant portions, the bit budget can go down as much as 40%. This concept can be extended to other modalities.
An important telemedicine application is the perusal of CT scans (digital format) from a central server housed in a healthcare enterprise across a bandwidth constrained network by radiologists situated at remote locations for medical diagnostic purposes. It is generally expected that a viewing station respond to an image request by displaying the image within 1-2 seconds. Owing to limited bandwidth, it may not be possible to deliver the complete image in such a short period of time with traditional techniques. In this paper, we investigate progressive image delivery solutions by using JPEG 2000. An estimate of the time taken in different network bandwidths is performed to compare their relative merits. We further make use of the fact that most medical images are 12-16 bits, but would ultimately be converted to an 8-bit image via windowing for display on the monitor. We propose a windowing progressive RoI technique to exploit this and investigate JPEG 2000 RoI based compression after applying a favorite or a default window setting on the original image. Subsequent requests for different RoIs and window settings would then be processed at the server. For the windowing progressive RoI mode, we report a 50% reduction in transmission time.
In this paper, we propose a block-based conditional entropy coding
scheme for medical image compression using the 2-D integer Haar
wavelet transform. The main motivation to pursue conditional
entropy coding is that the first-order conditional entropy is
always theoretically lesser than the first and second-order
entropies. We propose a sub-optimal scan order and an optimum
block size to perform conditional entropy coding for various
modalities. We also propose that a similar scheme can be used to
obtain a sub-optimal scan order and an optimum block size for
other wavelets. The proposed approach is motivated by a desire to
perform better than JPEG2000 in terms of compression ratio. We
hint towards developing a block-based conditional entropy coder,
which has the potential to perform better than JPEG2000. Though we
don't indicate a method to achieve the first-order conditional
entropy coder, the use of conditional adaptive arithmetic coder
would achieve arbitrarily close to the theoretical conditional
entropy. All the results in this paper are based on the medical
image data set of various bit-depths and various modalities.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.