A JAVA environment for medical image data analysis: Initial application for brain PET quantitation

Abstract

Analysis software for medical image data tends to be expensive and usable only in a restricted environment. Therefore the aim of the current project was to implement a flexible framework for medical image processing and visualization which is portable among platforms and open to different data formats including DICOM 3.0. The software was designed as a set of tools which encapsulate specialized functionality. The tools are full stand alone applications, but they are also able to present a co-operating environment within which images and other information are communicated in real time. Currently, the emphasis is on quantitative analysis of PET data by kinetic modelling. However, general viewing capabilities are included, and the design is flexible enough that other types of processing can easily be integrated by simply plugging in Java classes. The software is successfully applied to PET data quantitation in clinical research studies and even patient studies. Portability was aimed at by encoding the programs in Java. Experience shows that the implementation of such a complex and computationally demanding Java application is feasible. Although there are major portability issues to consider, configurations can be found on which the software runs stably and at a speed comparable to C code if just-in-time compilation is available.

Effects of Early Life Stress on [11C]DASB Positron Emission Tomography Imaging of Serotonin Transporters in Adolescent Peer- and Mother-Reared Rhesus Monkeys

“…MRTM2 parametric imaging was performed by voxelwise weighted linear least-squares fitting with weights equal to the inverse of the data variance. The data variance was obtained by the method based on the noise equivalent counts (Pajevic et al, 1998) implemented in pixelwise modeling software, PMOD [version 2.5, PMOD group, Zurich, Switzerland (Mikolajczyk et al, 1998)]. …”

18 F-choline in experimental soft tissue infection assessed with autoradiography and high-resolution PET

“…The raw data were reconstructed using the iterative one pass listmode-expectation maximisation (OPL-EM) algorithm (0.5 mm bin size, 200×200×440 matrix size) incorporating resolution recovery. Volumes of interest (VOIs) were subsequently placed on various tissue areas using the software PMOD [11]. The uptake values were expressed in kBq/ml using the syringe calibration.…”

“…Promising results for staging patients with prostate carcinoma have been reported with labelled choline compounds. Some studies have used carbon-11 labelled choline [4,5,6], whose clinical usefulness is limited by the short half-life of 11 C. This problem is overcome by 18 F-labelled choline (FCH), which has recently been introduced and has also shown promising results in the evaluation of patients with prostate cancer [7,8,9,10]. For every tumour imaging compound it is important to know the accumulation characteristics in non-tumorous lesions, of which inflammatory/infectious foci are among the most important.…”

Early metabolic responses in temozolomide treated low-grade glioma patients

“…Quantification of PET and MR data was done using PMOD [12] on digitized images as described in detail in our recent publication [13]. For calculation of the MR tumor volume we used the FLAIR sequence which provides the best delineation between tumor and adjacent brain [3,14].…”