The l 1-norm regularization has attracted attention for image reconstruction in computed tomography. The l 0-norm of the gradients of an image provides a measure of the sparsity of gradients of the image. In this paper, we present a new combined l 1-norm and l 0-norm regularization model for image reconstruction from limited projection data in computed tomography. We also propose an algorithm in the algebraic framework to solve the optimization effectively using the nonmonotone alternating direction algorithm with hard thresholding method. Numerical experiments indicate that this new algorithm makes much improvement by involving l 0-norm regularization.
The key component in deep learning research is the availability of training data sets. With a limited number of publicly available COVID-19 chest X-ray images, the generalization and robustness of deep learning models to detect COVID-19 cases developed based on these images are questionable. We aimed to use thousands of readily available chest radiograph images with clinical findings associated with COVID-19 as a training data set, mutually exclusive from the images with confirmed COVID-19 cases, which will be used as the testing data set. We used a deep learning model based on the ResNet-101 convolutional neural network architecture, which was pretrained to recognize objects from a million of images and then retrained to detect abnormality in chest X-ray images. The performance of the model in terms of area under the receiver operating curve, sensitivity, specificity, and accuracy was 0.82, 77.3%, 71.8%, and 71.9%, respectively. The strength of this study lies in the use of labels that have a strong clinical association with COVID-19 cases and the use of mutually exclusive publicly available data for training, validation, and testing.
In single photon emission computed tomography (SPECT) imaging, the choice of a suitable filter and its parameters for noise reduction purposes is a big challenge. Adverse effects on image quality arise if an improper filter is selected. Filtered back projection (FBP) is the most popular technique for image reconstruction in SPECT. With this technique, different types of reconstruction filters are used, such as the Butterworth and the Hamming. In this study, the effects on the quality of reconstructed images of the Butterworth filter were compared with the ones of the Hamming filter. A Philips ADAC forte gamma camera was used. A low-energy, high-resolution collimator was installed on the gamma camera. SPECT data were acquired by scanning a phantom with an insert composed of hot and cold regions. A Technetium-99m radioactive solution was homogenously mixed into the phantom. Furthermore, a symmetrical energy window (20%) centered at 140 keV was adjusted. Images were reconstructed by the FBP method. Various cutoff frequency values, namely, 0.35, 0.40, 0.45, and 0.50 cycles/cm, were selected for both filters, whereas for the Butterworth filter, the order was set at 7. Images of hot and cold regions were analyzed in terms of detectability, contrast, and signal-to-noise ratio (SNR). The findings of our study indicate that the Butterworth filter was able to expose more hot and cold regions in reconstructed images. In addition, higher contrast values were recorded, as compared to the Hamming filter. However, with the Butterworth filter, the decrease in SNR for both types of regions with the increase in cutoff frequency as compared to the Hamming filter was obtained. Overall, the Butterworth filter under investigation provided superior results than the Hamming filter. Effects of both filters on the quality of hot and cold region images varied with the change in cutoff frequency.
Bone age assessment (BAA) is an essential topic in the clinical practice of evaluating the biological maturity of children. Because the manual method is time-consuming and prone to observer variability, it is attractive to develop computer-aided and automated methods for BAA. In this paper, we present a fully automatic BAA method. To eliminate noise in a raw X-ray image, we start with using U-Net to precisely segment hand mask image from a raw X-ray image. Even though U-Net can perform the segmentation with high precision, it needs a bigger annotated dataset. To alleviate the annotation burden, we propose to use deep active learning (AL) to select unlabeled data samples with sufficient information intentionally. These samples are given to Oracle for annotation. After that, they are then used for subsequential training. In the beginning, only 300 data are manually annotated and then the improved U-Net within the AL framework can robustly segment all the 12611 images in RSNA dataset. The AL segmentation model achieved a Dice score at 0.95 in the annotated testing set. To optimize the learning process, we employ six off-the-shell deep Convolutional Neural Networks (CNNs) with pretrained weights on ImageNet. We use them to extract features of preprocessed hand images with a transfer learning technique. In the end, a variety of ensemble regression algorithms are applied to perform BAA. Besides, we choose a specific CNN to extract features and explain why we select that CNN. Experimental results show that the proposed approach achieved discrepancy between manual and predicted bone age of about 6.96 and 7.35 months for male and female cohorts, respectively, on the RSNA dataset. These accuracies are comparable to state-of-the-art performance.
The purpose of this study is to determine if microvascular tortuosity can be used as an imaging biomarker for the presence of tumor-associated angiogenesis and if imaging this biomarker can be used as a specific and sensitive method of locating solid tumors. Acoustic angiography, an ultrasound-based microvascular imaging technology, was used to visualize angiogenesis development of a spontaneous mouse model of breast cancer (n = 48). A reader study was used to assess visual discrimination between image types, and quantitative methods utilized metrics of tortuosity and spatial clustering for tumor detection. The reader study resulted in an area under the curve of 0.8, while the clustering approach resulted in the best classification with an area under the curve of 0.95. Both the qualitative and quantitative methods produced a correlation between sensitivity and tumor diameter. Imaging of vascular geometry with acoustic angiography provides a robust method for discriminating between tumor and healthy tissue in a mouse model of breast cancer. Multiple methods of analysis have been presented for a wide range of tumor sizes. Application of these techniques to clinical imaging could improve breast cancer diagnosis, as well as improve specificity in assessing cancer in other tissues. The clustering approach may be beneficial for other types of morphological analysis beyond vascular ultrasound images.
Brain tumors are a major health problem that affect the lives of many people. These tumors are classified as benign or cancerous. The latter can be fatal if not properly diagnosed and treated. Therefore, the diagnosis of brain tumors at the early stages of their development can significantly improve the chances of patient's full recovery after treatment. In addition to laboratory analyses, clinicians and surgeons extract information from medical images, recorded by various systems such as magnetic resonance imaging (MRI), X-ray, and computed tomography (CT). The extracted information is used to identify the essential characteristics of brain tumors (location, size, and type) in order to achieve an accurate diagnosis to determine the most appropriate treatment protocol. In this paper, we present an automated machine vision technique for the detection and localization of brain tumors in MRI images at their very early stages using a combination of k-means clustering, patch-based image processing, object counting, and tumor evaluation. The technique was tested on twenty real MRI images and was found to be capable of detecting multiple tumors in MRI images regardless of their intensity level variations, size, and location including those with very small sizes. In addition to its use for diagnosis, the technique can be integrated into automated treatment instruments and robotic surgery systems.