MF2ResU-Net: a multi-feature fusion deep learning architecture for retinal blood vessel segmentation / 数字中医药（英文）

Objective@#For computer-aided Chinese medical diagnosis and aiming at the problem of insufficient segmentation, a novel multi-level method based on the multi-scale fusion residual neural network (MF2ResU-Net) model is proposed.@*Methods@#To obtain refined features of retinal blood vessels, three cascade connected U-Net networks are employed. To deal with the problem of difference between the parts of encoder and decoder, in MF2ResU-Net, shortcut connections are used to combine the encoder and decoder layers in the blocks. To refine the feature of segmentation, atrous spatial pyramid pooling (ASPP) is embedded to achieve multi-scale features for the final segmentation networks.@*Results@#The MF2ResU-Net was superior to the existing methods on the criteria of sensitivity (Sen), specificity (Spe), accuracy (ACC), and area under curve (AUC), the values of which are 0.8013 and 0.8102, 0.9842 and 0.9809, 0.9700 and 0.9776, and 0.9797 and 0.9837, respectively for DRIVE and CHASE DB1. The results of experiments demonstrated the effectiveness and robustness of the model in the segmentation of complex curvature and small blood vessels.@*Conclusion@#Based on residual connections and multi-feature fusion, the proposed method can obtain accurate segmentation of retinal blood vessels by refining the segmentation features, which can provide another diagnosis method for computer-aided Chinese medical diagnosis.

Breast Cancer Computer-aided Diagnosis System from Digital Mammograms

Recently, breast cancer is one of the most popular cancers that women could suffer from. The gravity and seriousness of breast cancer can be evidenced by the fact that the mortality rates associated with it are the second highest after lung cancer. For the treatment of breast cancer, Mammography has emerged as the one whose modality when it comes to the defection of this cancer is most effective despite the challenges posed by dense breast parenchyma. In thisregard, computer-aided diagnosis (CADe) leverages the mammography systems’ output to facilitate the radiologist’s decision. It can be defined as a system that makes a similar diagnosis to the one done by a radiologist who relies for his/her interpretationon the suggestions generated by a computer after it analyzed a set of patient radiological images when making. Against this backdrop, the current paper examines different ways of utilizing known image processing and techniques of machine learning detection of breast cancer using CAD –more specifically, using mammogram images. This, in turn, helps pathologist in their decision-making process. For effective implementation of this methodology, CADe system was developed and tested on the public and freely available mammographic databases named MIAS database. CADe system is developed to differentiate between normal and abnormal tissues, and it assists radiologists to avoid missing breast abnormalities. The performance of all classifiers is the best by using thesequential forward selection (SFS) method. Also, we can conclude that the quantization grey level of (gray-level co-occurrence matrices) GLCM is a very significant factor to get robust high order features where the results are better with L equal to the size of ROI. Using an enormous number of several features assist the CADe system to be strong enough to distinguish between the different tissues

Multi-channel conditional generative adversarial networks retinal vessel segmentation algorithm / 中华实验眼科杂志

Objective To propose a model for accurately segmenting blood vessels in medical fundus images. Methods The algorithm of deep learning was used for the task of automatic segmentation of blood vessels in retinal fundus images in this paper. An improved vascular segmentation algorithm was proposed. For the different types of blood vessels in the fundus image, a multi-scale network structure was designed to extract features of both main blood vessels and vessel branches at the same time. Results The segmentation model proposed could achieve good results on all kinds of blood vessels even if they have low contrast and few obvious characteristics. The automatic vessel segmentation of retinal fundus images was implemented, and the performance of the model was evaluated through multiple evaluation indexes which are widely used in the field of medical image segmentation in the test stage. A specificity of 0. 9829,an F1 score of 0. 7944,a G-mean of 0. 8748,an Matthews correlation coefficient(MCC) of 0. 7764 and a specificity of 0. 9782 were obtained on the DRIVE dataset. An F1 score of 0. 7735 and an MCC of 0. 7573 were obtained on the STARE data set. Conclusions The proposed method has a great improvement over the segmentation algorithm of the same task. Furthermore,the results generated by our model can achieve comparable effect with the segmentation of human doctor.

A nucleus location method based on distance estimation / 生物医学工程学杂志

To locate the nuclei in hematoxylin-eosin (HE) stained section images more simply, efficiently and accurately, a new method based on distance estimation is proposed in this paper, which shows a new mind on locating the nuclei from a clump image. Different from the mainstream methods, proposed method avoids the operations of searching the combined singles. It can directly locate the nuclei in a full image. Furthermore, when the distance estimation built on the matrix sequence of distance rough estimating (MSDRE) is combined with the fact that a center of a convex region must have the farthest distance to the boundary, it can fix the positions of nuclei quickly and precisely. In addition, a high accuracy and efficiency are achieved by this method in experiments, with the precision of 95.26% and efficiency of 1.54 second per thousand nuclei, which are better than the mainstream methods in recognizing nucleus clump samples. Proposed method increases the efficiency of nuclear location while maintaining the location's accuracy. This can be helpful for the automatic analysis system of HE images by improving the real-time performance and promoting the application of related researches.

Using Parallel Convolutional Neural Networks for Treatment Position Recognition in X-ray Images / 中国医疗器械杂志

Treatment position recognition in medical images is a key technique in medical image processing. Due to the excellent performance of convolutional neural networks on features extraction and classification, an architecture of parallel convolutional neural networks is proposed to recognize treatment positions in X-ray images, which uses convolution kernels of different sizes to extract local features of different sizes in these images. The experimental analysis shows that parallel convolution neural networks, which can extract representative image features with more dimensions, are competent to classify and recognize treatment positions in medical images.

Improving teaching effects of medical image processing by research / 医疗卫生装备

Objective To improve the teaching effects of medical image processing.Methods Teaching content was reformed considering the problems encountered during clinical practice and research. Modern teaching methods such as problem-based learning and flipped classroom were adopted. In addition, professors from world famous Universities were invited to teach lessons.A website was also built for this course.Moreover,the students were encouraged to participate in after-class research.Finally,a new practice-based teaching mode with characteristic content and modern methods was developed.By adopting information technology,all-round students could be gifted with international vision.Results The students'interest was stimulated and their problem solving skills were improved.Conclusion The teaching effects of medical image processing can be largely improved by research,so that the students can not only master the course content well but also develop skills to solve practical problems.

Algoritmo para la identificación de nódulos pulmonares solitarios en imágenes de tomografía de tórax / Algorithm for identification of solitary pulmonary nodules in chest tomography scan images

La identificación del cáncer de pulmón en fases iniciales ha sido en los últimos años una tarea priorizada de la comunidad científica. Esta enfermedad representa la primera causa de muerte en el varón y la tercera después del cáncer de colon y mama en la mujer. La realización de estudios imagenológicos contribuye a la detección temprana de esta enfermedad. El elevado volumen de imágenes generado por los equipos médicos provoca la revisión de mucha información para emitir un diagnóstico médico. Con frecuencia se requiere la valoración de varios especialistas para llegar a un diagnóstico acertado, retardando el proceso de atención al paciente. En la presente investigación se exponen los resultados obtenidos al desarrollar un algoritmo utilizando métodos de procesamiento de imágenes, para la identificación de nódulos pulmonares solitarios. La utilización de sistemas que dirigen la atención de los especialistas a regiones candidatas en la imagen, proporcionando una segunda opinión en la interpretación de los resultados, pudiera mejorar la consistencia y agilizar el proceso de diagnóstico. Los resultados arrojados por el algoritmo desarrollado fueron contrastados con las anotaciones realizadas en imágenes publicadas en The Lung Image Database Consortium Image Collection (LIDC-IDRI) y se obtuvo un 77.78 por ciento de acierto en la detección de nódulos pulmonares solitarios(AU)

The identification of lung cancer at early stages has been in recent years a prioritized task for the scientific community. This disease is the leading cause of death in men and the third after the colon and breast cancer in women. Performing imaging studies contributes to the early detection of this disease. The high volume of images generated by medical equipment leads to reviewing much information to issue a medical diagnosis. Often are required the assessment of several specialists to reach an accurate diagnosis, slowing the process of patient care. In the present investigation are exposed the results obtained to develop an algorithm using image processing methods for the identification of solitary pulmonary nodules. The use of systems that direct the attention of specialists to candidate regions in the image, providing a second opinion in the interpretation of results could improve consistency and agility in the diagnostic process. The results obtained by the developed algorithm were compared with annotations in images published in The Lung Image Database Consortium Image Collection (LIDC-IDRI) and was obtained 77.78 percent accuracy in the detection of solitary pulmonary nodules(AU)

Application of medical image processing, transmission and on-line learning system in medical imaging teaching / 中华医学教育探索杂志

Using the Medical Imaging Experimental Teaching Demonstration Center of Nanjing Medical University and related web site as a platform, we have established a case library and online learning system which fused medical imaging, pathology and anatomy images according to the human organ system. The system has achieved a“medical imaging”teaching real-time, diversified, systematic and clinical simulation technology. Medical students can perform image processing, simulate writing reports and discuss online using this system. Teachers can use the system to carry out teaching reform-problem-based learning (PBL), and the integration of teaching with horgan systems and case based curriculum.

Processamento de estruturas tridimensionais de Medicina Nuclear na modalidade PET / Processing of three-dimensional structures of Nuclear Medicine in PET modality

A Medicina Nuclear, como especialidade de obtenção de imagens médicas é um dos principais procedimentos utilizados hoje nos centros de saúde, tendo como grande vantagem a capacidade de analisar o comportamento metabólico do paciente. Este projeto está baseado em imagens médicas obtidas através da modalidade PET (Positron Emission Tomography). Para isso, foi desenvolvida uma estrutura de processamento de imagens tridimensionais PET, constituída por etapas sucessivas que se iniciam com a obtenção das imagens padrões (gold standard), sendo utilizados para este fim volumes simulados do Ventrículo Esquerdo do Coração criadas como parte do projeto, assim como phantoms gerados com o software NCAT-4D. A seguir, nos volumes simulados é introduzido ruído Poisson que é o ruído característico das imagens PET. Na sequência é executada uma etapa de pré-processamento, utilizando alguns filtros 3D tais como o filtro da mediana, o filtro da Gaussiana ponderada e o filtro Anscombe/Wiener. Posteriormente é aplicada a etapa de segmentação, processo baseado na teoria de Conectividade Fuzzy sendo implementadas quatro diferentes abordagens 3D: Algoritmo Genérico, LIFO, kTetaFOEMS e Pesos Dinâmicos. Finalmente, um procedimento de avaliação conformado por três parâmetros (Verdadeiro Positivo, Falso Positivo e Máxima Distância) foi utilizado para mensurar o nível de eficiência e precisão do processo. Constatou-se que o par Filtro - Segmentador constituído pelo filtro Anscombe/Wiener junto com o segmentador Fuzzy baseado em Pesos Dinâmicos proporcionou os melhores resultados, com taxas de VP e FP na ordem de 98,49 ± 0,27% e 2,19 ± 0,19%, respectivamente, para o caso do volume do Ventrículo Esquerdo simulado. Com o conjunto de escolhas feitas ao longo da estrutura de processamento, encerrou-se o projeto analisando um número reduzido de volumes pertencentes a um exame PET real, obtendo-se a quantificação dos volumes.

The Nuclear medicine, as a specialty to obtain medical images is very important, and it has became one of the main procedures utilized in Health Care Centers to analyze the metabolic behavior of the patient. This project was based on medical images obtained by the PET modality (Positron Emission Tomography). Thus, we developed a framework for processing Nuclear Medicine three-dimensional images of the PET modality, which is composed of consecutive steps that start with the generation of standard images (gold standard) by using simulated images of the Left Ventricular Heart, such as phantoms obtained from the NCAT-4D software. Then, Poisson quantum noise was introduced into the whole volume to simulate the characteristic noises in PET images. Subsequently, the pre-processing step was executed by using specific 3D filters, such as the median filter, the weighted Gaussian filter, and the Anscombe/Wiener filter. Then the segmentation process, which is based on the Fuzzy Connectedness theory, was implemented. For that purpose four different 3D approaches were implemented: Generic, LIFO, kTetaFOEMS, and Dynamic Weight algorithm. Finally, an assessment procedure was used as a measurement tool to quantify three parameters (True Positive, False Positive and Maximum Distance) that determined the level of efficiency and precision of our process. It was found that the pair filter - segmenter formed by the Anscombe/Wiener filter together with the Fuzzy segmenter based on Dynamic Weights provided the best results, with VP and FP rates of 98.49 ± 0.27% and 2.19 ± 0.19%, respectively, for the simulation of the Left Ventricular volume. Along with the set of choices made during the processing structure, the project was finished with the analysis of a small number of volumes that belonged to a real PET test, thus the quantification of the volumes was obtained.

Application of wavelet transformation in medical imaging processing / 中国医学影像技术

Medical imaging plays an important role in modern medical diagnosis. Wavelet transformation is widely applied in medical imaging processing, including the fields of edge-detection, noise-reducing, image-enhancement, image-compression and image-integration. The theories and application of Wavelet transformation were systematically reviewed in this paper.

Característica de Euler-Poincaré aplicada para identificar baixa densidade óssea a partir de imagens tomográficas de vértebras / The Euler-Poincaré characteristic applied to identify low bone density from vertebral tomographic images

A densidade mineral óssea (DMO) é o exame padrão-ouro para o diagnóstico da osteoporose. No entanto, sabe-se que apenas essa medida não é suficiente para identificar completamente a fragilidade óssea e o conseqüente risco de fratura, tornando-se necessário a investigação da estrutura óssea. OBJETIVOS: Avaliar se a característica de Euler-Poincaré (CEP) para analisar a conectividade do osso trabecular poderia fornecer um suporte adicional na identificação da deterioração da estrutura óssea. MATERIAIS E MÉTODOS: Analisou-se um conjunto de imagens formando disectors, obtidas da tomografia computadorizada de vértebras lombares, a partir dos quais foi estimada a característica Euler-Poincarè.(CEP). Para lidar com o processamento de imagens dos disectors, foi desenvolvido um programa de computador usando o GTK+ para MS-Windows. Os resultados foram comparados com a DMO. RESULTADOS: Verificou-se que a medida da CEP está correlacionada com os resultados obtidos por meio da DMO para as vértebras lombares. Ficou demonstrado também que a área de conectividade das trabéculas que é propagada ao longo dos disectors corrobora para assegurar que os resultados da CEP sejam consistentes com a medida da DMO. CONCLUSÕES: A aplicação da CEP na análise das tomografias vertebrais poderá vir a se constituir num método para avaliar a estrutura óssea trabecular, e sua correlação com a resistência mecânica do osso, sendo necessários porém mais estudos para confirmar esses dados.

The bone mineral density (BMD) is the gold standard test for the diagnosis of osteoporosis. Nevertheless, it is known that only this measurement is not sufficient to completely identify bone fragility and the consequent risk of fracture, becoming necessary the investigation of the bone structure. OBJECTIVES: Evaluate if the Euler-Poincaré characteristic (EPC) for the analysis of the connectivity of the trabecular bone could supply an additional support in identifying the deterioration of the bone structure. MATERIALS AND METHODS: A group of images forming dissectors were analyzed, obtained from the computerized tomography of the lumbar vertebrae, from which the Euler-Poincarè characteristic (EPC) was estimated. To deal with the processing of images of the dissectors, a computer program was developed using the GTK+ for MS-Windows. The results were compared to the BMD. RESULTS: It was verified that the measurement of the EPC is correlated with the results obtained by means of the BMD for the lumbar vertebrae. It was also demonstrated that the area of connectivity of the trabecular that is propagated over the dissectors corroborates to ensure that the results of the EPC are consistent with the BMD measurement. CONCLUSIONS: The application of ECP in the analysis of the vertebral tomographies could constitute a method to evaluate the trabecular bone structure, and its correlation with the mechanical resistance of the bone, however being necessary more studies to confirm this data.

Study on Medical Image Courses Teaching / 中华医学教育探索杂志

The contents and features of medical imaging technology and medical image processing course are introduced in this paper,relationship of the two courses and their different parts are mainly discussed.Contents and the multimedia courseware are then designed.

Investigation of a Wavelet Transform Based Noise Filtering Approach for Medical Images / 医疗卫生装备

Objective To investigate a wavelet-transform-based approach that reduces the noise of medical images, and to compare the difference of the effects by different wavelet types. Methods A soft threshold approach based on the modification of local coefficient of wavelets was proposed. Firstly, a local modulus extrema distribution of the image, M j,m,n is obtained using wavelet transform. Then the modulus maximum was calculated and a threshold Tm was defined according to the statistical properties of the local modulus extrema distribution. If the extremum of the wavelet transform was greater than or equal to the threshold Tm, the corresponding wavelet coefficient was kept unchanged; while if the extremum of the wavelet transform was less than the threshold Tm, its corresponding wavelet coefficient was calculated using the soft threshold approach. Lastly, an inverse wavelet transform was performed according to the wavelet coefficients of these two parts so that the image could be reconstructed. Results The proposed approach could filter out the noise in medical images effectively, and the effects of noise reduction by different wavelets were different. Conclusion A useful wavelet threshold noise reduction algorithm can be obtained by wavelet multi-dimensional decomposition of image with proper selection of wavelet base function, and comparatively ideal effect of noise reduction can be achieved using this algorithm.

Edge detection of medical image and methods for tracking & showing of gray level in special regions / 医疗卫生装备

By using the image processing function and programming tomography,the following functions are realized: the contour line,edge line and gradation histogram of medical images are displayed;the selected regions of medical images are freely zoomed;the gray level of discretional regions is displayed using medical image processing tool and the distance of the given points is freely measured.

Design & Development of 3D Medical Image Processing System using VTK / 대한의료정보학회지

Recently, medical image processing systems which take advantages of computer information technology have been used usefully in the hospitals. In particular, the 3D volume rendering technique with a sets of 2D medical images can visualize the organs of human body into 3D image virtually and such systems have started to be used widely to find diseases and to build up the plan of treatment. In this paper, we introduce a medical image processing system which uses a public graphic library VTK so that the system can be developed in the low price with the same power compared to other commercial systems. The system supports 2D and 3D image processing such as image editing, filtering, 3D image reconstruction and etc. by accepting multiple 2D original medical images.

Infrastructure for image guided surgery / 中国医学物理学杂志

Image guided surgery (IGS) requires an integrated environment for seamless acquisition, visualization,manipulation, display, registration and storage of complex data sets. The infrastructure to support IGS integrates image acquisition, networks, presurgical planning, surgical navigation, and archival storage elements. This paper describes the principal components of an integrated IGS environment and an implementation in a large academic medical center. The IGS infrastructure is illustrated for practical applications in neurosurgical case examples.