Dinu Dragan

Medical Image on the Go

The idea for softcopy viewing of medical image outside the radiology reading room spread among the scientists in various fields for several years. An image could be read on workstation of all types, from desktop across movable to handheld. Benefits are numerous and continue to grow as physicians use them discovering new usage cases. Proposed solutions vary with PACS architecture invasion level, communication and storage image formats, and utilization. We employ JPEG2000 standard because of its high (lossy/lossless) compression ratio with minimal spatial distortion, retrieval-oriented storage, and streaming. It is embedded in PACS as the DICOM Private Data Element containing JPIP parameter string, so-called DICOM2000. The DICOM2000 message is transparent for standard DICOM devices at the slightest level of invasion. Thanks to sophisticated JPEG2000 streaming, medical image becomes suitable for any resolution and quality display and (wireless) networks. The solution is validated on the ACR/NEMA standard test set of PACS images.

Request redirection paradigm in medical image archive implementation

It is widely recognized that the JPEG2000 facilitates issues in medical imaging: storage, communication, sharing, remote access, interoperability, and presentation scalability. Therefore, JPEG2000 support was added to the DICOM standard Supplement 61. Two approaches to support JPEG2000 medical image are explicitly defined by the DICOM standard: replacing the DICOM image format with corresponding JPEG2000 codestream, or by the Pixel Data Provider service, DICOM supplement 106. The latest one supposes two-step retrieval of medical image: DICOM request and response from a DICOM server, and then JPIP request and response from a JPEG2000 server. We propose a novel strategy for transmission of scalable JPEG2000 images extracted from a single codestream over DICOM network using the DICOM Private Data Element without sacrificing system interoperability. It employs the request redirection paradigm: DICOM request and response from JPEG2000 server through DICOM server. The paper presents programming solution for implementation of request redirection paradigm in a DICOM transparent manner.

Quality evaluation of medical image compression: What to measure?

In this paper we present a continuation of our work on three dimensional quality evaluation space for medical image compression. Quality dimensions are: presentation-objective, presentation-subjective, and technical-objective. Each quality dimension is defined by a quality vector. We present which attributes of medical image compression should be included in quality vectors.

DICOM/JPEG2000 CLIENT/SERVER ARCHITECTURE IMPLEMENTATION

The paper presents client/server architecture based on DICOM standard (Digital Imaging and Communications in Medicine). Standard DICOM transfer syntax is extended to support JPEG2000 streaming. This grants DICOM clients ability to query and retrieve medical images in resolution and quality suitable to their display capabilities which is very useful when DICOM client runs on a device which has smaller display resolution, like palmtop computers or mobile phones. The benefits of this client/server architecture are analysed. The paper presents client/server prototype based on the extended DICOM transfer syntax from the end-user’s perspective. The applications based on these prototypes are meant for use in Hospital for Lung Diseases “Dr Vasa Savic Zrenjanin”.

Introducing an acceptability metric for image compression in PACS - A model

In this paper we introduce a model which describes the process of choosing individual metrics which are used to establish if a medical image compression is acceptable for a specific PACS or not. Image compression is an integral part of a PACS. There is no universal way for measuring acceptance of still image compression for a specific PACS. Many individual metrics have been proposed for measuring still image quality. They can be used individually or together. The proposed model should help the process of choosing individual metrics which are adequate for specific PACS. Results of individual metrics should improve the decision making regarding medical image compression during PACS development and implementation.

Binary Classification of Images for Applications in Intelligent 3D Scanning

Three-dimensional (3D) scanning techniques based on photogrammetry, also known as Structure-from-Motion (SfM), require many two-dimensional (2D) images of an object, obtained from different viewpoints, in order to create its 3D reconstruction. When these images are acquired using closed-space 3D scanning rigs, which are composed of large number of cameras fitted on multiple pods, flash photography is required and image acquisition must be well synchronized to avoid the problem of ‘misfired’ cameras. This paper presents an approach to binary classification (as ‘good’ or ‘misfired’) of images obtained during the 3D scanning process, using four machine learning methods—support vector machines, artificial neural networks, k-nearest neighbors algorithm, and random forests. Input to the algorithms are histograms of regions determined to be of interest in the detection of image misfires. The considered algorithms are evaluated based on the prediction accuracy that they achieved on our dataset. The average prediction accuracy of 94.19% is obtained using the random forests approach under cross-validation. Therefore, the application of the proposed approach allows the development of an ‘intelligent’ 3D scanning system which can automatically detect camera misfiring and repeat the scanning process without the need for human intervention.

Tools for Ubiquitous PACS System

Ubiquitous computing in healthcare (ubiquitous healthcare) is a very appealing and beneficial goal. It is a mean to achieve patient-centric healthcare and a mean to improve healthcare in general. PACS system is important part of a modern day hospital. To achieve ubiquitous healthcare, it is necessary to build ubiquitous PACS system. Thus a traditionally closed system based on large and location-dependent workstations will be accessible to all users regardless of their location and devices they use. This is achievable only by improving PACS system functionality without sacrificing its interoperability.