Auto-evaluation of skull radiograph accuracy using unsupervised anomaly detection.

BACKGROUND: Radiography plays an important role in medical care, and accurate positioning is essential for providing optimal quality images. Radiographs with insufficient diagnostic value are rejected, and retakes are required. However, determining the suitability of retaking radiographs is a qualitative evaluation. OBJECTIVE: To evaluate skull radiograph accuracy automatically using an unsupervised learning-based autoencoder (AE) and a variational autoencoder (VAE). In this study, we eliminated visual qualitative evaluation and used unsupervised learning to identify skull radiography retakes from the quantitative evaluation. METHODS: Five skull phantoms were imaged on radiographs, and 1,680 images were acquired. These images correspond to two categories: normal images captured at appropriate positions and images captured at inappropriate positions. This study verified the discriminatory ability of skull radiographs using anomaly detection methods. RESULTS: The areas under the curves for AE and VAE were 0.7060 and 0.6707, respectively, in receiver operating characteristic analysis. Our proposed method showed a higher discrimination ability than those of previous studies which had an accuracy of 52%. CONCLUSIONS: Our findings suggest that the proposed method has high classification accuracy in determining the suitability of retaking skull radiographs. Automation of optimal image consideration, whether or not to retake radiographs, contributes to improving operational efficiency in busy X-ray imaging operations.

Automated angular measurement for puncture angle using a computer-aided method in ultrasound-guided peripheral insertion.

Ultrasound guidance has become the gold standard for obtaining vascular access. Angle information, which indicates the entry angle of the needle into the vein, is required to ensure puncture success. Although various image processing-based methods, such as deep learning, have recently been applied to improve needle visibility, these methods have limitations, in that the puncture angle to the target organ is not measured. We aim to detect the target vessel and puncture needle and to derive the puncture angle by combining deep learning and conventional image processing methods such as the Hough transform. Median cubital vein US images were obtained from 20 healthy volunteers, and images of simulated blood vessels and needles were obtained during the puncture of a simulated blood vessel in four phantoms. The U-Net architecture was used to segment images of blood vessels and needles, and various image processing methods were employed to automatically measure angles. The experimental results indicated that the mean dice coefficients of median cubital veins, simulated blood vessels, and needles were 0.826, 0.931, and 0.773, respectively. The quantitative results of angular measurement showed good agreement between the expert and automatic measurements of the puncture angle with 0.847 correlations. Our findings indicate that the proposed method achieves extremely high segmentation accuracy and automated angular measurements. The proposed method reduces the variability and time required in manual angle measurements and presents the possibility where the operator can concentrate on delicate techniques related to the direction of the needle.

A modified undecimated discrete wavelet transform based approach to mammographic image denoising.

In this work, the authors present an effective denoising method to attempt reducing the noise in mammographic images. The method is based on using hierarchical correlation of the coefficients of discrete stationary wavelet transforms. The features of the proposed technique include iterative use of undecimated multi-directional wavelet transforms at adjacent scales. To validate the proposed method, computer simulations were conducted, followed by its applications to clinical mammograms. Mutual information originating from information theory was used as an evaluation measure for selection of an optimal wavelet basis function. We examined the performance of the proposed method by comparing it with the conventional undecimated discrete wavelet transform (UDWT) method in terms of processing time-consuming and image quality. Our results showed that with the use of the proposed method the computation time can be reduced to approximately 1/10 of the conventional UDWT method consumed. The results of visual assessment indicated that the images processed with the proposed UDWT method showed statistically significant superior image quality over those processed with the conventional UDWT method. Our research results demonstrate the superiority and effectiveness of the proposed approach.

Quality control system for mammographic breast positioning using deep learning.

This study proposes a deep convolutional neural network (DCNN) classification for the quality control and validation of breast positioning criteria in mammography. A total of 1631 mediolateral oblique mammographic views were collected from an open database. We designed two main steps for mammographic verification: automated detection of the positioning part and classification of three scales that determine the positioning quality using DCNNs. After acquiring labeled mammograms with three scales visually evaluated based on guidelines, the first step was automatically detecting the region of interest of the subject part by image processing. The next step was classifying mammographic positioning accuracy into three scales using four representative DCNNs. The experimental results showed that the DCNN model achieved the best positioning classification accuracy of 0.7836 using VGG16 in the inframammary fold and a classification accuracy of 0.7278 using Xception in the nipple profile. Furthermore, using the softmax function, the breast positioning criteria could be evaluated quantitatively by presenting the predicted value, which is the probability of determining positioning accuracy. The proposed method can be quantitatively evaluated without the need for an individual qualitative evaluation and has the potential to improve the quality control and validation of breast positioning criteria in mammography.

Information entropy measure for evaluation of image quality.

This paper presents a simple and straightforward method for synthetically evaluating digital radiographic images by a single parameter in terms of transmitted information (TI). The features of our proposed method are (1) simplicity of computation, (2) simplicity of experimentation, and (3) combined assessment of image noise and resolution (blur). Two acrylic step wedges with 0-1-2-3-4-5 and 0-2-4-6-8-10 mm in thickness were used as phantoms for experiments. In the present study, three experiments were conducted. First, to investigate the relation between the value of TI and image noise, various radiation doses by changing exposure time were employed. Second, we examined the relation between the value of TI and image blurring by shifting the phantoms away from the center of the X-ray beam area toward the cathode end when imaging was performed. Third, we analyzed the combined effect of deteriorated blur and noise on the images by employing three smoothing filters. Experimental results show that the amount of TI is closely related to both image noise and image blurring. The results demonstrate the usefulness of our method for evaluation of physical image quality in medical imaging.

[Simulation study of radiographic image quality measurement based on transmitted information].

Although radiographic image quality is considered difficult to evaluate in a straightforward and systematic manner, it may be possible by using an index of transmitted information. As a preliminary study, relations between transmitted information and two image characteristics, namely, image noise and image blurring, were evaluated by simulation. The value of transmitted information was decreased if image noise and image blurring increased. The relationships were corroborated on an experimental basis. This paper suggests the possibility of a simple, straightforward method for synthetically evaluating radiographic images by a single parameter in terms of transmitted information.

Z-score-based semi-quantitative analysis of the volume of the temporal horn of the lateral ventricle on brain CT images.

The volume of the temporal horn of the lateral ventricle (THLV) on brain computed tomography (CT) images is important for neurologic diagnosis. Our purpose in this study was to develop a z-score-based semi-quantitative analysis for estimation of the THLV volume by using voxel-based morphometry. The THLV volume was estimated by use of a z-score mapping method that consisted of four main steps: anatomic standardization, construction of a normal reference database, calculation of the z score, and calculation of the mean z score in a volume of interest (VOI). A mean z score of the CT value obtained from a VOI around the THLV was used as an index for the THLV volume. CT scans from 50 subjects were evaluated. For evaluation of the accuracy of this method for estimating the THLV volume, the THLV volume was determined manually by neuroradiologists (serving as the reference volume). A mean z score was calculated from the VOI for each THLV of the 50 subjects by use of the proposed method. The accuracy of this method was evaluated by use of the relationship between the mean z score and the reference volume. The quadratic polynomial regression equation demonstrated a statistically significant correlation between the mean z score and the reference volume of the THLV (R (2) = 0.94; P < 0.0001). In 92 of 100 THLVs (92 %), the 95 % prediction interval of the regional mean z score captured the reference volume of the THLV. The z-score-based semi-quantitative analysis has the potential quantitatively to estimate the THLV volume on CT images.

An automated detection method for the MCA dot sign of acute stroke in unenhanced CT.

The hyperdense middle cerebral artery (MCA) dot sign representing a thromboembolus is one of the important computed tomography (CT) findings for acute stroke on unenhanced CT images. Our purpose in this study was to develop an automated method for detection of the MCA dot sign of acute stroke on unenhanced CT images. The algorithm of the method which we developed consisted of 5 major steps: extraction of the sylvian fissure region, initial identification of MCA dots based on the morphologic top-hat transformation, feature extraction of candidates, elimination of false positives (FPs) by use of a rule-based scheme, and classification of candidates using a support vector machine (SVM) classifier with four features. Our database comprised 297 CT images obtained from seven patients with the MCA dot sign. The performance of this scheme for classification of the MCA dot sign was evaluated by means of a leave-one-case out method. The performance of the classification by use of the SVM achieved a maximum sensitivity of 97.5% (39/40) at a FP rate of 1.28 per image. The sensitivity for detection of the MCA dot sign was 97.5% (39/40) with a FP rate of 0.5 per hemisphere. The method we developed has the potential to detect the MCA dot sign of acute stroke on unenhanced CT images.

Improving image quality in medical images using a combined method of undecimated wavelet transform and wavelet coefficient mapping.

We propose a method for improving image quality in medical images by using a wavelet-based approach. The proposed method integrates two components: image denoising and image enhancement. In the first component, a modified undecimated discrete wavelet transform is used to eliminate the noise. In the second component, a wavelet coefficient mapping function is applied to enhance the contrast of denoised images obtained from the first component. This methodology can be used not only as a means for improving visual quality of medical images but also as a preprocessing module for computer-aided detection/diagnosis systems to improve the performance of screening and detecting regions of interest in images. To confirm its superiority over existing state-of-the-art methods, the proposed method is experimentally evaluated via 30 mammograms and 20 chest radiographs. It is demonstrated that the proposed method can further improve the image quality of mammograms and chest radiographs, as compared to two other methods in the literature. These results reveal the effectiveness and superiority of the proposed method.

Comparison of a discrete wavelet transform method and a modified undecimated discrete wavelet transform method for denoising of mammograms.

The purpose of this study was to evaluate the performance of a conventional discrete wavelet transform (DWT) method and a modified undecimated discrete wavelet transform (M-UDWT) method applied to mammographic image denoising. Mutual information, mean square error, and signal to noise ratio were used as image quality measures of images processed by the two methods. We examined the performance of the two methods with visual perceptual evaluation. A two-tailed F test was used to measure statistical significance. The difference between the M-UDWT processed images and the conventional DWT-method processed images was statistically significant (P<0.01). The authors confirmed the superiority and effectiveness of the M-UDWT method. The results of this study suggest the M-UDWT method may provide better image quality as compared to the conventional DWT.