Your browser doesn't support javascript.
Mostrar: 20 | 50 | 100
Resultados 1 - 4 de 4
Mais filtros

Base de dados
Intervalo de ano de publicação
Artigo em Inglês | MEDLINE | ID: mdl-31634851


Quality control / assessment of ultrasound (US) images is an essential step in clinical diagnosis. This process is usually done manually, suffering from some drawbacks, such as dependence on operator's experience and extensive labors, as well as high inter- and intra-observer variation. Automatic quality assessment of US images is therefore highly desirable. Fetal US cardiac four-chamber plane (CFP) is one of the most commonly used cardiac views, which was used in the diagnosis of heart anomalies in the early 1980s. In this paper, we propose a generic deep learning framework for automatic quality control of fetal US CFPs. The proposed framework consists of three networks: (1) a basic CNN (B-CNN), roughly classifying four-chamber views from the raw data; (2) a deeper CNN (D-CNN), determining the gain and zoom of the target images in a multi-task learning manner; and (3) the aggregated residual visual block net (ARVBNet), detecting the key anatomical structures on a plane. Based on the output of the three networks, overall quantitative score of each CFP is obtained, so as to achieve fully automatic quality control. Experiments on a fetal US dataset demonstrated our proposed method achieved a highest mean average precision (mAP) of 93.52% at a fast speed of 101 frames per second (FPS). In order to demonstrate the adaptability and generalization capacity, the proposed detection network (i.e., ARVBNet) has also been validated on the PASCAL VOC dataset, obtaining a highest mAP of 81.2% when input size is approximately 300×300.

Med Image Anal ; 58: 101548, 2019 12.
Artigo em Inglês | MEDLINE | ID: mdl-31525671


It is essential to measure anatomical parameters in prenatal ultrasound images for the growth and development of the fetus, which is highly relied on obtaining a standard plane. However, the acquisition of a standard plane is, in turn, highly subjective and depends on the clinical experience of sonographers. In order to deal with this challenge, we propose a new multi-task learning framework using a faster regional convolutional neural network (MF R-CNN) architecture for standard plane detection and quality assessment. MF R-CNN can identify the critical anatomical structure of the fetal head and analyze whether the magnification of the ultrasound image is appropriate, and then performs quality assessment of ultrasound images based on clinical protocols. Specifically, the first five convolution blocks of the MF R-CNN learn the features shared within the input data, which can be associated with the detection and classification tasks, and then extend to the task-specific output streams. In training, in order to speed up the different convergence of different tasks, we devise a section train method based on transfer learning. In addition, our proposed method also uses prior clinical and statistical knowledge to reduce the false detection rate. By identifying the key anatomical structure and magnification of the ultrasound image, we score the ultrasonic plane of fetal head to judge whether it is a standard image or not. Experimental results on our own-collected dataset show that our method can accurately make a quality assessment of an ultrasound plane within half a second. Our method achieves promising performance compared with state-of-the-art methods, which can improve the examination effectiveness and alleviate the measurement error caused by improper ultrasound scanning.

Cardiovasc Pathol ; 39: 38-50, 2019.
Artigo em Inglês | MEDLINE | ID: mdl-30623879


The traditional classification of congenital aortic arch abnormalities was described by James Stewart and colleagues in 1964. Since that time, advances in diagnostic imaging technology have led to better delineation of the vasculature anatomy and the identification of previously unrecognized and unclassified anomalies. In this manuscript, we review the existing literature and propose a series of modifications to the original Stewart classification of congenital aortic arch abnormalities to incorporate this new knowledge. In brief, we propose the following modifications: (1) In Group I, we further divide subgroup B into left arch atretic and right arch atretic; (2) In Group II, we add three more subgroups, including aberrant right innominate artery, "isolated" right innominate artery (RIA), "isolated" right carotid artery with aberrant right subclavian artery; (3) In Groups I, II, and III, we add a subgroup of absence of both ductus arteriosus; and (4) In Group IV, we add three subgroups, including circumflex retro-esophageal aorta arch, persistent V aortic arch, and anomalous origin of pulmonary artery from ascending aorta.

Aorta Torácica/anormalidades , Cardiopatias Congênitas/classificação , Terminologia como Assunto , Malformações Vasculares/classificação , Aorta Torácica/diagnóstico por imagem , Tomada de Decisão Clínica , Cardiopatias Congênitas/diagnóstico por imagem , Cardiopatias Congênitas/terapia , Humanos , Valor Preditivo dos Testes , Prognóstico , Malformações Vasculares/diagnóstico por imagem , Malformações Vasculares/terapia
Prenat Diagn ; 36(2): 117-26, 2016 Feb.
Artigo em Inglês | MEDLINE | ID: mdl-26573084


OBJECTIVE: We aim to determine the accuracy of first-trimester ultrasonography in detecting fetal limb abnormalities. METHODS: This is a retrospective study of all women undergoing fetal nuchal translucency (NT) assessment and detailed fetal anatomic survey in the first trimester at a single tertiary-care referral center in China. Fetal anatomy scans were repeated in the second trimester. Detection of fetal limb abnormalities was compared between first and second trimester anatomy scans and confirmed at delivery or at autopsy. RESULTS: Analyzed were 9438 fetuses from 9197 women (241 twin pairs). The incidence of fetal limb abnormalities was 0.38% (36/9438). Of these, 28 (77.8%) were diagnosed prenatally: 23 (63.9%) on first trimester scan and 5 (13.9%) on second trimester scan. Limb reduction defects (usually transverse limb deficiencies) were the most common limb defects identified in the first trimester (n = 12), followed by clubfoot (n = 4), skeletal dysplasia (n = 3), sirenomelia (n = 1), limb dysplasia (n = 1), malposition (n = 1), and syndactyly (n = 1). Nine fetuses with isolated limb abnormalities had normal NT, while 74.1% (20/27) of limb abnormalities that were associated with other abnormalities had increased NT. CONCLUSIONS: This study demonstrates that the majority of limb abnormalities detected prenatally [23/28 (82%)] can be identified in the first trimester, especially major limb defects; however, our numbers are small and still need larger cases for further investigation.

Deformidades Congênitas dos Membros/diagnóstico por imagem , Primeiro Trimestre da Gravidez , Ultrassonografia Pré-Natal/métodos , Adolescente , Adulto , Doenças do Desenvolvimento Ósseo/diagnóstico por imagem , China , Pé Torto Equinovaro/diagnóstico por imagem , Ectromelia/diagnóstico por imagem , Feminino , Humanos , Pessoa de Meia-Idade , Medição da Translucência Nucal , Gravidez , Estudos Retrospectivos , Sensibilidade e Especificidade , Sindactilia , Centros de Atenção Terciária , Adulto Jovem