Medical Imaging Repositories Research Articles

Automated Anatomic Labeling Architecture for Content Discovery in Medical Imaging Repositories.

The combination of textual data with visual features is known to enhance medical image search capabilities. However, the most advanced imaging archives today only index the studies' available meta-data, often containing limited amounts of clinically useful information. This work proposes an anatomic labeling architecture, integrated with an open source archive software, for improved multimodal content discovery in real-world medical imaging repositories. The proposed solution includes a technical specification for classifiers in an extensible medical imaging archive, a classification database for querying over the extracted information, and a set of proof-of-concept convolutional neural network classifiers for identifying the presence of organs in computed tomography scans. The system automatically extracts the anatomic region features, which are saved in the proposed database for later consumption by multimodal querying mechanisms. The classifiers were evaluated with cross-validation, yielding a best F1-score of 96% and an average accuracy of 97%. We expect these capabilities to become common-place in production environments in the future, as automated detection solutions improve in terms of accuracy, computational performance, and interoperability.

Medical Image Processor and Repository – MIPAR

The recent progress in medical image analysis has given birth to image-guided therapy, virtual reality and augmented reality. These among other innovations have greatly improved healthcare delivery, improved quality of life and also saved lives.. The advances in Internet and network technology have produced web technologies which make it possible to offer platform or software as a service via the web, making it possible for end users to access computer resources or specialized computer tools remotely. However, the high cost of image acquisition, the limited availability of medical image analysts, and the limited collaborative efforts between medical experts and scientists are major challenges for medical image analysis in the developing world. The aim of this project was to develop a medical image e-infrastructure called Medical Image Processor and Analysis (MIPAR) to contain a repository of medical images acquired from Africa and a platform for processing medical images. The backend of MIPAR which is resident on a High-Performance Computing infrastructure was built using FutureGateway, a framework for building science gateway. The image upload and download module was built upon the framework of Open Access Repository with the front end developed using HTML, CSS and BootStrap. JavaScript and JQuery were used for scripting. User's access to the server is controlled with HTTP response and a Client-Server Architecture, while the image processing tools on the server side communicate with PHP using Representational State Architecture (REST) API. MIPAR was tested using brain MRI images. Images were submitted remotely via MIPAR's web interface and kept in the repository. Image processing facility of MIPAR was tested using the brain extraction module, a 3D image of approximately 43MB was successfully brain extracted within 60 s. MIPAR allows users to donate, download and process medical images at no cost. It is our hope that such useful and unique tool will encourage collaboration, improve diagnosis, improve patient management, and promote open science in Africa.

Convolutional neural networks and hash learning for feature extraction and of fast retrieval of pulmonary nodules

With a widespread use of digital imaging data in hospitals, the size of medical image repositories is increasing rapidly. This causes difficulty in managing and querying these large databases leading to the need of content based medical image retrieval (CBMIR) systems. A major challenge in CBMIR systems is the ?semantic gap? that exists between the low level visual information captured by imaging devices and high level semantic information perceived by the human. Using deep convolution neural network (CNN) to construct the CBMIR system can fully characterize the high level semantic features information for medical image retrieval. The existing network mostly used for the natural images can?t produce a good result directly applied to medical image. This paper used UNet method to preprocessing under the guidance of medical knowledge. Then, multi-scale receiving field convolution module is used to extract features of the segmented images with different sizes. Finally, encoded the features and used a coarse to fine search strategy with an average search accuracy of 0.73.

Multimodal medical imaging repository and information system in epilepsy surgery

Surgery is an option for patients with drug-resistant epilepsy, but it requires a comprehensive assessment. Electroencephalography (EEG), magnetic resonance imaging (MRI), and functional MRI (fMRI) are used to localize the epileptogenic zone, which directly affects the surgery outcome. Accessing EEG, MRI, and fMRI results and patient information simultaneously using traditional methods might result in misinformation and increase the workload of clinicians. In this study, we developed a modern web-based repository system for the preoperative evaluation of epilepsy disorder, including multimodal medical images and patient information. Our dedicated system is enriched with clinical metadata that are not currently available and is managed with an online application. It overcomes the identified problem and minimizes possible medical errors. In conclusion, this system has the potential to accelerate surgical procedures, get reliable results, and improve the seizure outcome. It is an extensive solution for epilepsy hospitals and clinical research centers. It may serve as a standard template for archiving multiple imaging techniques.

An Intelligent Cloud Storage Gateway for Medical Imaging.

Historically, medical imaging repositories have been supported by indoor infrastructures. However, the amount of diagnostic imaging procedures has continuously increased over the last decades, imposing several challenges associated with the storage volume, data redundancy and availability. Cloud platforms are focused on delivering hardware and software services over the Internet, becoming an appealing solution for repository outsourcing. Although this option may bring financial and technological benefits, it also presents new challenges. In medical imaging scenarios, communication latency is a critical issue that still hinders the adoption of this paradigm. This paper proposes an intelligent Cloud storage gateway that optimizes data access times. This is achieved through a new cache architecture that combines static rules and pattern recognition for eviction and prefetching. The evaluation results, obtained from experiments over a real-world dataset, show that cache hit ratios can reach around 80%, leading to reductions of image retrieval times by over 60%. The combined use of eviction and prefetching policies proposed can significantly reduce communication latency, even when using a small cache in comparison to the total size of the repository. Apart from the performance gains, the proposed system is capable of adjusting to specific workflows of different institutions.

SiNC: Saliency-injected neural codes for representation and efficient retrieval of medical radiographs.

Medical image collections contain a wealth of information which can assist radiologists and medical experts in diagnosis and disease detection for making well-informed decisions. However, this objective can only be realized if efficient access is provided to semantically relevant cases from the ever-growing medical image repositories. In this paper, we present an efficient method for representing medical images by incorporating visual saliency and deep features obtained from a fine-tuned convolutional neural network (CNN) pre-trained on natural images. Saliency detector is employed to automatically identify regions of interest like tumors, fractures, and calcified spots in images prior to feature extraction. Neuronal activation features termed as neural codes from different CNN layers are comprehensively studied to identify most appropriate features for representing radiographs. This study revealed that neural codes from the last fully connected layer of the fine-tuned CNN are found to be the most suitable for representing medical images. The neural codes extracted from the entire image and salient part of the image are fused to obtain the saliency-injected neural codes (SiNC) descriptor which is used for indexing and retrieval. Finally, locality sensitive hashing techniques are applied on the SiNC descriptor to acquire short binary codes for allowing efficient retrieval in large scale image collections. Comprehensive experimental evaluations on the radiology images dataset reveal that the proposed framework achieves high retrieval accuracy and efficiency for scalable image retrieval applications and compares favorably with existing approaches.

An efficient architecture to support digital pathology in standard medical imaging repositories

In the past decade, digital pathology and whole-slide imaging (WSI) have been gaining momentum with the proliferation of digital scanners from different manufacturers. The literature reports significant advantages associated with the adoption of digital images in pathology, namely, improvements in diagnostic accuracy and better support for telepathology. Moreover, it also offers new clinical and research applications. However, numerous barriers have been slowing the adoption of WSI, among which the most important are performance issues associated with storage and distribution of huge volumes of data, and lack of interoperability with other hospital information systems, most notably Picture Archive and Communications Systems (PACS) based on the DICOM standard.This article proposes an architecture of a Web Pathology PACS fully compliant with DICOM standard communications and data formats. The solution includes a PACS Archive responsible for storing whole-slide imaging data in DICOM WSI format and offers a communication interface based on the most recent DICOM Web services. The second component is a zero-footprint viewer that runs in any web-browser. It consumes data using the PACS archive standard web services. Moreover, it features a tiling engine especially suited to deal with the WSI image pyramids. These components were designed with special focus on efficiency and usability. The performance of our system was assessed through a comparative analysis of the state-of-the-art solutions. The results demonstrate that it is possible to have a very competitive solution based on standard workflows.

Intelligent generator of big data medical imaging repositories

The production of medical imaging data has grown tremendously in the last decades. Nowadays, even small institutions produce a considerable amount of studies. Furthermore, the general trend in new imaging modalities is to produce more data per examination. As a result, the design and implementation of tomorrow's storage and communication systems must deal with big data issues. The research on technologies to cope with big data issues in large scale medical imaging environments is still in its early stages. This is mostly due to the difficulty of implementing and validating new technological approaches in real environments, without interfering with clinical practice. Therefore, it is crucial to create test bed environments for research purposes. This study proposes a methodology for creating simulated medical imaging repositories, based on the indexing of model datasets, extraction of patterns and modelling of study production. The system creates a model from a real-world repository's representative time window and expands it according to on-going research needs. In addition, the solution provides distinct approaches to reducing the size of the generated datasets. The proposed system has already been used by other research projects in validation processes that aim to assess the performance and scalability of developed systems.

Medical image retrieval using deep convolutional neural network

With a widespread use of digital imaging data in hospitals, the size of medical image repositories is increasing rapidly. This causes difficulty in managing and querying these large databases leading to the need of content based medical image retrieval (CBMIR) systems. A major challenge in CBMIR systems is the semantic gap that exists between the low level visual information captured by imaging devices and high level semantic information perceived by human. The efficacy of such systems is more crucial in terms of feature representations that can characterize the high-level information completely. In this paper, we propose a framework of deep learning for CBMIR system by using deep convolutional neural network (CNN) that is trained for classification of medical images. An intermodal dataset that contains twenty-four classes and five modalities is used to train the network. The learned features and the classification results are used to retrieve medical images. For retrieval, best results are achieved when class based predictions are used. An average classification accuracy of 99.77% and a mean average precision of 0.69 is achieved for retrieval task. The proposed method is best suited to retrieve multimodal medical images for different body organs.

Repository of 3D images for education and everyday clinical practice purposes

Abstract Novel, easy-automation technologies such as three-dimensional (3D) printing and reverse engineering can improve the training of medical and allied health professionals and everyday clinical practice. This paper aims at the presentation of its own concept of the repository of medical images for education and everyday clinical practice purposes. Presented concept of the repository constitutes a relatively novel solution, but its further development may lead to the novel family of commercial initiatives aiming at joining common efforts toward optimization of 3D-based technologies in everyday clinical practice and online e-learning system.

Classification of Brain MRI Tumor Images: A Hybrid Approach

Nowadays, brain tumor has been proved as a life threatening disease which cause even to death. Various classification techniques have been identified for Brain MRI Tumor Images. In this paper brain tumor from MR Images with the help of hybrid approach has been carried out. This hybrid approach includes discrete wavelet transform (DWT) to be used for extraction of features, Genetic algorithm for diminishing the number of features and support vector machine (SVM) for brain tumor classification. Images are downloaded from SICAS Medical Image Repository which classified images as benign or malign type. The proposed hybrid approach is implemented in MATLAB 2015a platform. Parameters used for analyzing the images are given as: entropy, smoothness, root mean square error (RMS), kurtosis and correlation. The simulation analysis approach results shows that hybrid approach offers better performance by improving accuracy and minimizing the RMS error in comparison with the state-of-the-art techniques in the similar context.

Repository of images for reverse engineering and medical simulation purposes

Novel technologies such as 3D printing (additive manufacturing), 3D scanning and reverse engineering may significantly improve application of the principles of medicine in current clinical practice. This paper aims at presentation of the own concept of the repository of medical images based on 3D printing and reverse engineering technology. The proposed concept of the repository can constitute a beginning of the novel family of commercial techniques needed for development and optimization of reverse engineering toward printing the fully clinically functional solutions.

Pattern recognition for cache management in distributed medical imaging environments.

Traditionally, medical imaging repositories have been supported by indoor infrastructures with huge operational costs. This paradigm is changing thanks to cloud outsourcing which not only brings technological advantages but also facilitates inter-institutional workflows. However, communication latency is one main problem in this kind of approaches, since we are dealing with tremendous volumes of data. To minimize the impact of this issue, cache and prefetching are commonly used. The effectiveness of these mechanisms is highly dependent on their capability of accurately selecting the objects that will be needed soon. This paper describes a pattern recognition system based on artificial neural networks with incremental learning to evaluate, from a set of usage pattern, which one fits the user behavior at a given time. The accuracy of the pattern recognition model in distinct training conditions was also evaluated. The solution was tested with a real-world dataset and a synthesized dataset, showing that incremental learning is advantageous. Even with very immature initial models, trained with just 1 week of data samples, the overall accuracy was very similar to the value obtained when using 75% of the long-term data for training the models. Preliminary results demonstrate an effective reduction in communication latency when using the proposed solution to feed a prefetching mechanism. The proposed approach is very interesting for cache replacement and prefetching policies due to the good results obtained since the first deployment moments.

A Routing Mechanism for Cloud Outsourcing of Medical Imaging Repositories.

Web-based technologies have been increasingly used in picture archive and communication systems (PACS), in services related to storage, distribution, and visualization of medical images. Nowadays, many healthcare institutions are outsourcing their repositories to the cloud. However, managing communications between multiple geo-distributed locations is still challenging due to the complexity of dealing with huge volumes of data and bandwidth requirements. Moreover, standard methodologies still do not take full advantage of outsourced archives, namely because their integration with other in-house solutions is troublesome. In order to improve the performance of distributed medical imaging networks, a smart routing mechanism was developed. This includes an innovative cache system based on splitting and dynamic management of digital imaging and communications in medicine objects. The proposed solution was successfully deployed in a regional PACS archive. The results obtained proved that it is better than conventional approaches, as it reduces remote access latency and also the required cache storage space.

A Centralized Platform for Geo-Distributed PACS Management

Picture Archive and Communication System (PACS) is a globally adopted concept and plays a fundamental role in patient care flow within healthcare institutions. However, the deployment of medical imaging repositories over multiple sites still brings several practical challenges namely related to operation and management (O&M). This paper describes a Web-based centralized console that provides remote monitoring, testing, and management over multiple geo-distributed PACS. The system allows the PACS administrator to define any kind of service or operation, reducing the need for local technicians and providing a 24/7 monitoring solution.

Dicoogle, a Pacs Featuring Profiled Content Based Image Retrieval

Content-based image retrieval (CBIR) has been heralded as a mechanism to cope with the increasingly larger volumes of information present in medical imaging repositories. However, generic, extensible CBIR frameworks that work natively with Picture Archive and Communication Systems (PACS) are scarce. In this article we propose a methodology for parametric CBIR based on similarity profiles. The architecture and implementation of a profiled CBIR system, based on query by example, atop Dicoogle, an open-source, full-fletched PACS is also presented and discussed. In this solution, CBIR profiles allow the specification of both a distance function to be applied and the feature set that must be present for that function to operate. The presented framework provides the basis for a CBIR expansion mechanism and the solution developed integrates with DICOM based PACS networks where it provides CBIR functionality in a seamless manner.

Two Decades of Endovascular Abdominal Aortic Aneurysm Repair: Enormous Progress With Serious Lessons Learned

The modern open surgical management of abdominal aortic aneurysm (AAA) has changed little since its inception in the 1950s. Endoaneurysmorrhaphy, first described by Rudolph Matas in 1888, involved ligating the branches of an aneurysm from within the aneurysm sac. Approximately 25 years later at the

A RESTful Image Gateway for Multiple Medical Image Repositories

Mobile technologies are increasingly important components in telemedicine systems and are becoming powerful decision support tools. Universal access to data may already be achieved by resorting to the latest generation of tablet devices and smartphones. However, the protocols employed for communicating with image repositories are not suited to exchange data with mobile devices. In this paper, we present an extensible approach to solving the problem of querying and delivering data in a format that is suitable for the bandwidth and graphic capacities of mobile devices. We describe a three-tiered component-based gateway that acts as an intermediary between medical applications and a number of Picture Archiving and Communication Systems (PACS). The interface with the gateway is accomplished using Hypertext Transfer Protocol (HTTP) requests following a Representational State Transfer (REST) methodology, which relieves developers from dealing with complex medical imaging protocols and allows the processing of data on the server side.

Predictors of Abdominal Aortic Aneurysm Sac Enlargement After Endovascular Repair

Conclusion: In the Unites States compliance with endovascular aneurysm repair (EVAR) device guidelines is low. Rates of post-EVAR aneurysm sac enlargement are high. Summary: Companies marketing EVAR devices measure technical factors such as delivery accuracy, sealing ability, and fixation strength in the laboratory. These measurements are used to generate instructions for use (IFU) that are packaged with each EVAR device sold in the United States. The randomized trials comparing EVAR with open aneurysm repair have used EVAR devices in accordance with IFUs. Many physicians may perform EVAR without adherence to IFUs. It is unknown how often EVAR is performed outside the IFU for the device and longer-term results of EVAR performed outside IFUs are unknown. However, a significant portion of late deaths after EVAR are due to aneurysm rupture (De Bruin JL et al, N Engl J Med 2010;362;1881-9, and Wyss TR, Ann Surg 2010;252;805-12) and aneurysm rupture after EVAR is linked with aneurysm sac enlargement. The authors sought to determine compliance with IFUs in EVAR placement over the last decade and to determine relationships between anatomic characteristics at baseline of aorta and iliac arteries and the subsequent incidence of aortic aneurysm sac enlargement after EVAR. The authors examined patients undergoing EVAR between January 1, 1999, and December 31, 2008, using a medical imaging repository at M2S, Inc. (West Lebanon, NH). The deidentified data on all patients in the prospectively acquired M2S database was used for those who underwent a CT scan before EVAR and had at least one CT scan after EVAR between 1999 and 2008 in the United States. Patients were excluded if they underwent an EVAR in the context of premarketing or postmarketing studies in which M2S served as a core imaging laboratory. Preoperative aorta iliac anatomic characteristics were reviewed for each patient. The specific endovascular device implanted was not available from this database. Morphologic measurements were therefore compared with the most liberal and most conservative published anatomic guidelines stated in manufacturer's IFUs. Primary outcome was post-EVAR aneurysm sac enlargement >5 mm. There were 10,228 patients included in the database and used for the study. Of these, 59% had a maximum abdominal aortic aneurysm diameter below the 55-mm threshold at which intervention is recommended over surveillance. Forty-two percent of patients had anatomy that met the most conservative definition of IFUs and 69% met the most liberal definition of device IFUs. The 5-year post-EVAR rate of AAA sac enlargement was 41%. Independent predictors of AAA sac enlargement included age >80 years, endoleak, aortic neck diameter >28 mm, aortic neck angle >60°, and common iliac artery diameter >20 mm. Comment: Some physicians placing EVAR devices may be acting in what some could consider an irresponsible fashion. In this study, most EVAR devices were placed in abdominal aortic aneurysms that were below the threshold for intervention justified by randomized trials. Even with the most liberal definition of IFUs almost one-third of devices were placed outside manufacturer's recommendations, and in this database, >40% of aortas treated with EVAR had aneurysm sac enlargement at 5 years. The data are disturbing and raise significant questions about aneurysm treatment in the United States and about long-term risk of aneurysm rupture in patients treated with EVAR in the United States. Physicians placing EVAR devices need to do better job of policing themselves. If not, one can be sure someone else will! One can only speculate that it won't take long for the malpractice attorneys to stumble on to this.

The state of the art of medical imaging technology: from creation to archive and back.

Medical imaging has learnt itself well into modern medicine and revolutionized medical industry in the last 30 years. Stemming from the discovery of X-ray by Nobel laureate Wilhelm Roentgen, radiology was born, leading to the creation of large quantities of digital images as opposed to film-based medium. While this rich supply of images provides immeasurable information that would otherwise not be possible to obtain, medical images pose great challenges in archiving them safe from corrupted, lost and misuse, retrievable from databases of huge sizes with varying forms of metadata, and reusable when new tools for data mining and new media for data storing become available. This paper provides a summative account on the creation of medical imaging tomography, the development of image archiving systems and the innovation from the existing acquired image data pools. The focus of this paper is on content-based image retrieval (CBIR), in particular, for 3D images, which is exemplified by our developed online e-learning system, MIRAGE, home to a repository of medical images with variety of domains and different dimensions. In terms of novelties, the facilities of CBIR for 3D images coupled with image annotation in a fully automatic fashion have been developed and implemented in the system, resonating with future versatile, flexible and sustainable medical image databases that can reap new innovations.