Search

BACKGROUND: Dividing liver organs or lesions depicting on computed tomography (CT) images could be applied to help tumor staging and treatment. However, most existing image segmentation technologies use manual or semi-automatic analysis, making the analysis process costly and time-consuming. OBJECTIVE: This research aims to develop and apply a deep learning network architecture to segment liver tumors automatically after fine tuning parameters. METHODS AND MATERIALS: The medical imaging is obtained from the International Symposium on Biomedical Imaging (ISBI), which includes 3D abdominal CT scans of 131 patients diagnosed with liver tumors. From these CT scans, there are 7,190 2D CT images along with the labeled binary images. The labeled binary images are regarded as gold standard for evaluation of the segmented results by FCN (Fully Convolutional Network). The backbones of FCN are extracted from Xception, InceptionresNetv2, MobileNetv2, ResNet18, ResNet50 in this study. Meanwhile, the parameters including optimizers (SGDM and ADAM), size of epoch, and size of batch are investigated. CT images are randomly divided into training and testing sets using a ratio of 9:1. Several evaluation indices including Global Accuracy, Mean Accuracy, Mean IoU (Intersection over Union), Weighted IoU and Mean BF Score are applied to evaluate tumor segmentation results in the testing images. RESULTS: The Global Accuracy, Mean Accuracy, Mean IoU, Weighted IoU, and Mean BF Scores are 0.999, 0.969, 0.954, 0.998, 0.962 using ResNet50 in FCN with optimizer SGDM, batch size 12, and epoch 9. It is important to fine tuning the parameters in FCN model. Top 20 FNC models enable to achieve higher tumor segmentation accuracy with Mean IoU over 0.900. The occurred frequency of InceptionresNetv2, MobileNetv2, ResNet18, ResNet50, and Xception are 9, 6, 3, 5, and 2 times. Therefore, the InceptionresNetv2 has higher performance than others. CONCLUSIONS: This study develop and test an automated liver tumor segmentation model based on FCN. Study results demonstrate that many deep learning models including InceptionresNetv2, MobileNetv2, ResNet18, ResNet50, and Xception have high potential to segment liver tumors from CT images with accuracy exceeding 90%. However, it is still difficult to accurately segment tiny and small size tumors by FCN models.

According to the Health Promotion Administration in the Ministry of Health and Welfare statistics in Taiwan, over ten thousand women have breast cancer every year. Mammography is widely used to detect breast cancer. However, it is limited by the operator’s technique, the cooperation of the subjects, and the subjective interpretation by the physician. It results in inconsistent identification. Therefore, this study explores the use of a deep neural network algorithm for the classification of mammography images. In the experimental design, a retrospective study was used to collect imaging data from actual clinical cases. The mammography images were collected and classified according to the breast image reporting and data-analyzing system (BI-RADS). In terms of model building, a fully convolutional dense connection network (FC-DCN) is used for the network backbone. All the images were obtained through image preprocessing, a data augmentation method, and transfer learning technology to build a mammography image classification model. The research results show the model’s accuracy, sensitivity, and specificity were 86.37%, 100%, and 72.73%, respectively. Based on the FC-DCN model framework, it can effectively reduce the number of training parameters and successfully obtain a reasonable image classification model for mammography.

OBJECTIVE: This study aims to analyze and compare the diagnostic effectiveness of 320-row multi-detector computed tomography for coronary artery angiography (MDCTA) in subjects with and without sublingual vasodilator (nitroglycerin). MATERIALS AND METHODS: From September 2015 to September 2016, 70 individuals without history of major cardiovascular diseases who underwent MDCTA for health examination were retrospectively categorized into sublingual nitroglycerin (NTG) and non-NTG groups. Medical history, CT dose index (CTDI), and multi-slice CT images were compared between two groups. A diameter of coronary artery (DA, mm) was computed and analyzed. RESULTS: A total of 41 males and 29 females (mean age: 55.43±8.84 years, range: 34– 76) were reviewed. Normal and abnormal MDCTA findings were noted in 54 and 16 participants, respectively, with the detection rate of coronary artery disease being 23%. There was no significant difference in inter-observer variability of coronary CTA image quality and diagnosis between the NTG and non-NTG groups among three experienced radiologists. Although the percentage dilatation of left anterior descending branch (LAD), right coronary artery (RCA) and left circumflex branch (LCX) following in the NTG group were 12.4%, 12.8% and 25.3%, respectively (p

Anesthesia assessment is most important during surgery. Anesthesiologists use electrocardiogram (ECG) signals to assess the patient’s condition and give appropriate medications. However, it is not easy to interpret the ECG signals. Even physicians with more than 10 years of clinical experience may still misjudge. Therefore, this study uses convolutional neural networks to classify ECG image types to assist in anesthesia assessment. The research uses Internet of Things (IoT) technology to develop ECG signal measurement prototypes. At the same time, it classifies signal types through deep neural networks, divided into QRS widening, sinus rhythm, ST depression, and ST elevation. Three models, ResNet, AlexNet, and SqueezeNet, are developed with 50% of the training set and test set. Finally, the accuracy and kappa statistics of ResNet, AlexNet, and SqueezeNet in ECG waveform classification were (0.97, 0.96), (0.96, 0.95), and (0.75, 0.67), respectively. This research shows that it is feasible to measure ECG in real time through IoT and then distinguish four types through deep neural network models. In the future, more types of ECG images will be added, which can improve the real-time classification practicality of the deep model.

Single photon emission computed tomography (SPECT) has been employed to detect Parkinson&rsquo, s disease (PD). However, analysis of the SPECT PD images was mostly based on the region of interest (ROI) approach. Due to limited size of the ROI, especially in the multi-stage classification of PD, this study utilizes deep learning methods to establish a multiple stages classification model of PD. In the retrospective study, the 99mTc-TRODAT-1 was used for brain SPECT imaging. A total of 202 cases were collected, and five slices were selected for analysis from each subject. The total number of images was thus 1010. According to the Hoehn and Yahr Scale standards, all the cases were divided into healthy, early, middle, late four stages, and HYS I~V six stages. Deep learning is compared with five convolutional neural networks (CNNs). The input images included grayscale and pseudo color of two types. The training and validation sets were 70% and 30%. The accuracy, recall, precision, F-score, and Kappa values were used to evaluate the models&rsquo, performance. The best accuracy of the models based on grayscale and color images in four and six stages were 0.83 (AlexNet), 0.85 (VGG), 0.78 (DenseNet) and 0.78 (DenseNet).

Background: Ultrasound imaging has become one of the most widely utilized adjunct tools in breast cancer screening due to its advantages. The computer-aided detection of breast ultrasound is rapid development via significant features extracted from images. Objectives: The main aim was to identify features of breast ultrasound image that can facilitate reasonable classification of ultrasound images between malignant and benign lesions. Patients and Methods: This research was a retrospective study in which 85 cases (35 malignant [positive group] and 50 benign [negative group] with diagnostic reports) with ultrasound images were collected. The B-mode ultrasound images have manually selected regions of interest (ROI) for estimated features of an image. Then, a fractal dimensional (FD) image was generated from the original ROI by using the box-counting method. Both FD and ROI images were extracted features, including mean, standard deviation, skewness, and kurtosis. These extracted features were tested as significant by t-test, receiver operating characteristic (ROC) analysis and Kappa coefficient. Results: The statistical analysis revealed that the mean texture of images performed the best in differentiating benign versus malignant tumors. As determined by the ROC analysis, the appropriate qualitative values for the mean and the LR model were 0.85 and 0.5, respectively. The sensitivity, specificity, accuracy, positive predicted value (PPV), negative predicted value (NPV), and Kappa for the mean was 0.77, 0.84, 0.81, 0.77, 0.84, and 0.61, respectively. Conclusion: The presented method was efficient in classifying malignant and benign tumors using image textures. Future studies on breast ultrasound texture analysis could focus on investigations of edge detection, texture estimation, classification models, and image features.

The B-mode ultrasound usually contains scattering speckle noise which reduces the detailed resolution of the target and is regarded as an intrinsic noise that interferes with diagnostic precision. The aim of this study was to classify hepatic steatosis through applying attenuation correction with a phantom to reduce speckle noise in liver ultrasound tomography in patients. This retrospective study applied three randomized groups signifying different liver statuses. A total of 114 patients’ effective liver ultrasound images—30 normal, 44 fatty, and 40 cancerous—were included. The proposed depth attenuation correction method was first applied to images. Three regions of interest were manually drawn on the images. Next, five feature values for the regions of interest were calculated. Finally, the hybrid method of logistic regression and support vector machine was employed to classify the ultrasound images with 10-fold cross-validation. The accuracy, kappa statistic, and mean absolute error of the proposed hybrid method were 87.5%, 0.812, and 0.119, respectively, which were higher than those of the logistic regression method—75.0%, 0.548, and 0.280—or those of the support vector machine method—75.7%, 0.637, and 0.293—respectively. Therefore, the hybrid method has been proven to be more accurate and have better performance and less error than either single method. The hybrid method provided acceptable accuracy of classification in three liver ultrasound image groups after depth attenuation correction. In the future, the deep learning approaches may be considered for the application in classifying liver ultrasound images.

The neuroimaging techniques such as dopaminergic imaging using Single Photon Emission Computed Tomography (SPECT) with 99mTc-TRODAT-1 have been employed to detect the stages of Parkinson&rsquo, s disease (PD). In this retrospective study, a total of 202 99mTc-TRODAT-1 SPECT imaging were collected. All of the PD patient cases were separated into mild (HYS Stage 1 to Stage 3) and severe (HYS Stage 4 and Stage 5) PD, according to the Hoehn and Yahr Scale (HYS) standard. A three-dimensional method was used to estimate six features of activity distribution and striatal activity volume in the images. These features were skewness, kurtosis, Cyhelsky&rsquo, s skewness coefficient, Pearson&rsquo, s median skewness, dopamine transporter activity volume, and dopamine transporter activity maximum. Finally, the data were modeled using logistic regression (LR) and support vector machine (SVM) for PD classification. The results showed that SVM classifier method produced a higher accuracy than LR. The sensitivity, specificity, PPV, NPV, accuracy, and AUC with SVM method were 0.82, 1.00, 0.84, 0.67, 0.83, and 0.85, respectively. Additionally, the Kappa value was shown to reach 0.68. This claimed that the SVM-based model could provide further reference for PD stage classification in medical diagnosis. In the future, more healthy cases will be expected to clarify the false positive rate in this classification model.

BACKGROUND: Coronary artery disease (CAD) remains the leading cause of death worldwide. Currently, cardiac multidetector computed tomography (MDCT) is widely used to diagnose CAD. The purpose in this study is to identify informative and useful predictors from left ventricular (LV) in the early CAD patients using cardiac MDCT images. MATERIALS AND METHODS: Study groups comprised 42 subjects who underwent a screening health examination, including laboratory testing and cardiac angiography by 64-slice MDCT angiography. Two geometrical characteristics and one image density were defined as shape, size and stiffness on MDCT image. The t-test, logistic regression, and receiver operating characteristic curve were applied to assess and identify the significant predictors. The Kappa statistics was used to exam the agreements with physician’s judgments (i.e., Golden of True, GOT). RESULTS: The proposed three characteristics of LV MDCT images are important predictors and risk factors for the early CAD patients. These predictors present over 80% of AUC and higher odds ratio. The Kappa statistics was 0.68 for the combinations of shape and stiffness into logistic regression. CONCLUSIONS: The shape, size and stiffness of the left ventricular on MDCT can be used to be the effective indicators in the early CAD patients. Besides, the combinations of shape and stiffness into logistic regression could provide substantial agreement with physician’s judgments.

The attenuation in ultrasound is one of primary factor reducing quality of image. The heterogeneity intensity of ultrasound images was made by scatter and attenuation. In this study, the quantitative analysis of liver ultrasound images was performed by three extracted features with and without depth corrections. A tissue phantom and 136 empirical images were involved to evaluate the reliability and performance with three extracted features. The average intensity of ROI with depth correction had better reliability and performance with AUC 0.689 and specificity, sensitivity, and accuracy are 67%, 60%, and 62% with respectively.