Develop various algorithms for the measurements of medical image stereological topology, geometric sizes, tumor distortion, and tumor distributions. The project focuses on the stereological metrics of of tumor volume, interfacial or intrinsic surface and volume ratio, tumor major characteristics length scales (dimensions), and tumor distortion and distribution topology.
voluMeasure Project
Prof. Ge Wang, Ph.D., Prof. Simon Kao, M.D., and Prof. Jun Ni, Ph.D.
Department of Radiology, Carver College of Medicine, University of Iowa, USA
Development of clinic and education software to measure tumor volume size. The software is called voluMeasure. The software is based on grid-point counting algorithm that enable radiologist or researchers to quickly calculate the tumor volume size at multiple levels. The software is programmed by Java. The project development team members are Dr. Ge Wang, Dr. Simon Kao, Dr. Jun Ni, Lili Huang, Wenli He, Tao He, The software has been used in research, clinics and education. The software was presented at RSNA|05. The current users are from researchers and radiologists from Biochemistry Department, Yale University, USA, PharmaBlood Company in Florida, USA, India Institute of Medical Sciences, India, School of Biology, University of Newcastle upon Tyne Newcastle, U. K., Duke University, USA, Northwestern University, USA, New Jersey Medical School, USA, and Medical School, University of Kansas., USA.
Volume estimation of tumor sizes is important for diagnostics and therapeutic evaluation using radiological imaging modalities. There are many stereological methodologies, which have been developed for that purpose. At the present, medical researchers use general-purpose image analysis tools to measure the tumor volume. Although these tools are well established, generally speaking these software tools are useful to researchers but not suitable in the clinical practice. There is an immediate need for volumetric measurement software at low-cost and high efficiency with cross platform applicability.
Theoretically, medical professionals may manually trace the boundary of the tumor area in each cross-sectional image reconstructed by X-ray computed topography (CT), magnetic resonance imaging (MRI), or other imaging techniques. The accumulation of area values based on pixel counting leads to the volume of a particular structure. Although the manual tracing approach is accurate, simple, and easy to be implemented, it is usually tedious and inefficient. In addition to the traditional volumetry, several widely-used highly simplified methods exist. A determination of the largest linear dimension has been proposed in the Response Evaluation Criteria In Solid Tumors (RECIST) recently (http://www.eortc.be/recist). This method is based on the one dimensional measurement along the main characteristic principle axis. Another technique involves calculating the product of the two largest, orthogonal diameters on a single axial image that contains the largest region of interest, as suggested in the Word Health Organization (WHO) criteria for tumor response to treatment. This method is also called the bi-dimensional or cross-dimensional method. Both of these methods are relatively easy to implement, but they are often inaccurate, particularly in the cases of irregularly shaped structures.
Recently, people have developed an alternative method, which is also referred to as point-counting stereology (abbreviated as PCS). For example, Kao and Wang introduced a point-counting stereological method, and developed such a program in the Interactive Data Language (IDL) (Research Systems Inc.) on a PC for estimating volume. Their method is a fixed-grid stereological volumetry approach. This method is very promising, since it assumes little knowledge about object shape or orientation. The point-counting method was favorably compared with the traditional tracing method. Although the point-counting method is rapid and satisfactory, it requires an expensive IDL programming environment. Also, there are various other point-counting techniques available, such as the random-grid stereological volumetry technique. It is our goal to develop an efficient technique for evaluating volumes of tumors with desirable functionality and user friendly interfaces at low software cost, acceptable accuracy, and cross-platform compatibility. VoluMeasure is a java-enhanced stereological volume measurement system using the point-counting approach . The point-counting algorithm is implemented in the J2SE 1.4 programming environment. A user-friendly interface and several novel features are integrated.
User Guide
Download
Window 98/2000/ME/XP version
LINUX/UNIX/MacOS version
Order Information
License Agreement
Medical Imaging and Radiology Informatics (MIRI)
Medical Imaging Information System (MIIS) 
Parallel Computing for Medical Imaging (PCMI) 
Modeling of Biotransport in
Biological Systems (MBBS)
Optical Imaging Tomography and Applications (OITA) 
Image Stereology and Clinic Applications (ISCA)
Stereotactic Atlas for the Anatomic Topology (SAAT)
Coupled Diffusions for Image Enhencement (DDIE)