Lisa Kinnard

Learn More
A multiple circular path convolution neural network (MCPCNN) architecture specifically designed for the analysis of tumor and tumor-like structures has been constructed. We first divided each suspected tumor area into sectors and computed the defined mass features for each sector independently. These sector features were used on the input layer and were(More)
This study attempted to accurately segment the masses and distinguish malignant from benign tumors. The masses were segmented using a technique that combines pixel aggregation with likelihood analysis. We found that the segmentation method can delineate the tumor body as well as tumor peripheral regions covering typical mass boundaries and some spiculation(More)
A number of interrelated factors can affect the precision and accuracy of lung nodule size estimation. To quantify the effect of these factors, we have been conducting phantom CT studies using an anthropomorphic thoracic phantom containing a vasculature insert to which synthetic nodules were inserted or attached. Ten repeat scans were acquired on different(More)
This work is a part of our more general effort to probe the interrelated factors impacting the accuracy and precision of lung nodule measurement tasks. For such a task a low-bias size estimator is needed so that the true effect of factors such as acquisition and reconstruction parameters, nodule characteristics and others can be assessed. Towards this goal,(More)
Our purpose in this work was to develop an automatic boundary detection method for mammographic masses and to rigorously test this method via statistical analysis. The segmentation method utilized a steepest change analysis technique for determining the mass boundaries based on a composed probability density cost function. Previous investigators have shown(More)
This study attempted to accurately segment tumors in mammograms. Although this task is considered to be a preprocessing step in a computer analysis program, it plays an important role for further analysis of breast lesions. The region of interest (ROI) was segmented using the pixel aggregation and region growing techniques combined with maximum likelihood(More)
The purpose of this work was to develop an automatic boundary detection method for mammographic masses and to observe the method's performance on different four of the five margin groups as defined by the ACR, namely, spiculated, ill-defined, circumscribed, and obscured. The segmentation method utilized a maximum likelihood steep change analysis technique(More)
We have digitized mammography films of African-American patients treated in the Howard University Hospital Radiology Department and have developed a database using these images. Two hundred and sixty cases totaling more than 5,000 images have been scanned with a high resolution Kodak LS85 laser scanner. The database system and web-based search engine were(More)
  • 1