Search

A deep learning-based automatic bone age identification system (ABAIs) was introduced in medical imaging. This ABAIs enhanced accurate, consistent, and timely clinical diagnostics and enlightened research fields of deep learning and artificial intelligence (AI) in medical imaging.The goal of this study was to use the Deep Neural Network (DNN) model to assess bone age in months based on a database of pediatric left-hand radiographs.The Inception Resnet V2 model with a Global Average Pooling layer to connect to a single fully connected layer with one neuron using the Rectified Linear Unit (ReLU) activation function consisted of the DNN model for bone age assessment (BAA) in this study. The medical data in each case contained posterior view of X-ray image of left hand, information of age, gender and weight, and clinical skeletal bone assessment.A database consisting of 8,061 hand radiographs with their gender and age (0-18 years) as the reference standard was used. The DNN model's accuracies on the testing set were 77.4%, 95.3%, 99.1% and 99.7% within 0.5, 1, 1.5 and 2 years of the ground truth respectively. The MAE for the study subjects was 0.33 and 0.25 year for male and female models, respectively.In this study, Inception Resnet V2 model was used for automatic interpretation of bone age. The convolutional neural network based on feature extraction has good performance in the bone age regression model, and further improves the accuracy and efficiency of image-based bone age evaluation. This system helps to greatly reduce the burden on clinical personnel.

Breast cancer is one of the most harmful diseases for women with the highest morbidity. An efficient way to decrease its mortality is to diagnose cancer earlier by screening. Clinically, the best approach of screening for Asian women is ultrasound images combined with biopsies. However, biopsy is invasive and it gets incomprehensive information of the lesion. The aim of this study is to build a model for automatic detection, segmentation, and classification of breast lesions with ultrasound images. Based on deep learning, a technique using Mask regions with convolutional neural network was developed for lesion detection and differentiation between benign and malignant. The mean average precision was 0.75 for the detection and segmentation. The overall accuracy of benign/malignant classification was 85%. The proposed method provides a comprehensive and noninvasive way to detect and classify breast lesions.

Stroke is a deadly condition. It is a leading cause of death in the world. Even with today's technologies, stroke detection and prevention is still a difficult and inaccurate task. In order to improve stroke detection, more knowledge about the physiological systems that stroke affects must be better understood. Much has been done in past studies already on how stroke affects each subsystem, such as blood pressure, cerebral blood flow, and heart rate. However, not much is known about the independence and relations between the cardiovascular subsystems and other physiological mechanisms. The purpose of this study is to approach this problem by using the nonlinear coupling analysis technique to study the independence of blood pressure, cerebral blood flow, and heart rate based on chaotic measures, in order to determine the relationships between these parameters. The linear coupling relation between heart rate and blood pressure is analyzed in time and frequency domain for evaluating baroreflex. 10 healthy controls and 10 stroke outpatients were enrolled in this study. The result indicates that stroke outpatients have a less independent chaoticness of the three subsystems. Cardiac-baroreceptor sensitivity (BRS) decreased significantly in stroke patients (p

Parkinson’s disease cases have been on the rise in the recent years, which promoted several different researches into the disorder. However, there hasn’t much research been done in the non-motor aspects of the disease. This study aims to improve the understanding of one of the non-motor symptoms of Parkinson’s disease. Specifically, this research aims to further understand cerebral autoregulation in patients with Parkinson’s disease. In order to achieve this aim, 25 subjects were recruited, with 11 healthy controls and 14 patients with Parkinson’s disease. The continuous blood pressure and continuous cerebral blood flow velocity of all subjects were recorded and processed while the subjects were at rest, tilt-up, and during hyperventilation. Linear signal and system analysis techniques were applied such as the power spectral density analysis and cross-correlation function analysis. Results showed that patients with Parkinson’s disease did not show a significant difference from the control group while at rest and after tilt-up. However, there was a significant difference between the groups during hyperventilation. The results obtained in this study suggested that the metabolic regulatory pathway for cerebral autoregulation is impaired in patients with Parkinson’s disease.

Independent component analysis (ICA) has been proven to be a powerful tool for removing artifacts from electroencephalogram (EEG) recordings in the form of blind source separation (BSS). Independent components (ICs) come from undesired sources that are mixed with the useful signal, and the assessment of such ICs allows them to be detected. But the unwanted ICs also can contain some useful information. To overcome this problem, wavelet-enhanced ICA (wICA) can be used, and this method applies a wavelet threshold for each wavelet coefficient to suppress abnormal deformation in each wavelet coefficient. Using the wICA algorithm to suppress artifacts provides an EEG signal with less distortion in the amplitude and in the phase of the cerebral part of the EEG, and the cerebral part of the EEG can be estimated and obtained very similar to control conditions. However, the EEG signals are affected by various artifact components, and those that have the greatest influence are electromyography (EMG) and electrooculography (EOG). These artifacts may appear simultaneously, randomly or interruptedly, so a fixed threshold level is not really appropriate. We proposed a system including wICA integrated with an adaptive filter model, and this combination system can provide the best prediction of the impacts of artifacts to set up a threshold value that is adaptive and suitable. Our experimental results showed that are approach provided better rejection of artifacts than the wICA system.