Introduction: Imaging application is one of the most important technologies for detecting pathological condition in humans and animals. But these technologies are expensive, time consuming and can be harmful to patients. Thermographic imaging has been used in the research and development of the clinical application. Thermographic imaging is non-invasive, and inexpensive technology. This research and development uses thermographic imaging to detect bone cancer and anterior cruciate ligament in canines and hyperthyroid in felines. The algorithms with the best sensitivity metrics are designed from the result of previous research [Subedi; 2014, Liu; 2012, Fu; 2014] and used in the development of application software.
Objectives: Primary objective of this research and development is to develop a clinical application which can be used as a pre-screening tool for detecting diseases in cats and dogs. The clinical application uses thermographic imaging.
Clinical Application Development: Image database was created at first which is used as the training set while running the test. Algorithms are stores as an XML (extensible markup language) and these algorithms are based on the previous research [Subedi; 2014, Liu; 2012, Fu; 2014].Clinical Application has the graphical user interface where user load the test image, enter the demographic information, minimum and maximum temperature and select the animal, possible disease, body part and camera view, hit the Run Test button to run the experiment. First step of the experiment is to perform color normalization to ensure that all the images are mapped to a common temperature scale. Next, based on the parameters selected by the user in the GUI, feature extraction and pattern classification is performed. Results are displayed in an html file which includes demographic information of animal, diagnostic class, sensitivity of the algorithm used and the test image.
Testing and Results: GUI testing, system testing and correctness testing has been performed to ensure the software is functioning as expected. Results from correctness testing show that for canine bone cancer and elbow or knee body part anterior and posterior view give the best results as sensitivity for these two camera views are 94%. For canine bone cancer of full or limb body part, anterior camera view gives the best result because the algorithm for anterior camera view has 100% sensitivity. For canine bone cancer of shoulder or hip body part only lateral view is available in the application which is identifying all 10 images correctly as shown in table as sensitivity for this algorithm is 95%. For wrist body part, it is clear that posterior camera view is giving best result as algorithm used for this has 100% sensitivity. For feline hyperthyroid, unshaved images gives the best result as sensitivity of the algorithm used for unshaved is 88%.For canine anterior lateral view is giving the best results among other camera views as algorithm used for lateral camera view has 79% sensitivity.
Conclusion: Results show that the clinical application can be used as a pre-screening tool for the detection of diseases in cats and dogs. The clinical application is easy to scale; so if more images are received or new algorithm is developed with a better success rate; the new images and algorithms can be integrated in the application easily without any changes in the coding.
|Advisor:||Umbaugh, Scott E.|
|Commitee:||LeAnder, Robert, Smith, Scott|
|School:||Southern Illinois University at Edwardsville|
|Department:||Electrical and Computer Engineering|
|School Location:||United States -- Illinois|
|Source:||MAI 55/01M(E), Masters Abstracts International|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be