Service Description

Develop computational algorithms to simulate human visual performance in reading X-rays and medical images. The evaluation results can then be used to optimize your systems design.

Medical image quality varies with the image capture systems, dosages, image processing algorithms, and will affect the diagnostic performance of image reading experts. Usually, human observers have to look at the concerned images and to evaluate the image quality. In mammography, for example, ACR phantom images and CDMAM phantom images are read by image reading experts to obtain the quality scores of the mammography images. Such a method, however, is costly and time consuming, and the variability of the results can be quite large due to human involvement and testing methods employed.

Here we can help you to develop computational algorithms to simulate the detection performance of an average human observer, so that the phantom images (and other medical images) can be evaluated computationally without using real human observers.

Key Benefits

• Reduced human involvements in the tedious evaluation process.

• Automated computational tool to reduce costs and time.

• Streamlined processes to optimize the design of medical imaging systems.

Capabilities

•  Develop computational algorithms to simulate human visual  performance on specific evaluation tasks.

•  Prototype evaluation tools.

•  Conduct confirmation studies and tool refinements.