Past Research |
Analysis of Confocal Microendoscopic Images for Automatic
Detection of Ovarian Cancer
Student: Mehul Patel
|
Diagnosis of disease is often accomplished via evaluation of tissue biopsy samples. However, misdiagnosis or
inability to make a diagnosis is often encountered due to improper selection of tissue. This is costly to the patient and the health-care system as a
whole. The novel, confocal microendoscope allows cellular-level imaging (optical biopsy) of tissues in
regions that are accessible by endoscopes such as, lungs, bladder, cervix, uterus, colon, peritoneum and ovaries. This aids physicians in the
diagnosis of pathologies in situ and could, in the ideal case, replace tissue extraction biopsy. At the very least, this capability
could improve the selection of tissue for biopsy and increase the accuracy of
diagnosis.
In this project, images of human
ovaries (ex vivo) obtained with the confocal microendoscope imaging
system were analyzed to detect pathologies.
Here are some examples of confocal microendoscopic images of normal human ovaries:
Examples of Abnormal Human Ovaries:
This work was a collaborative effort with Prof. Art Gmitro in the
Dept. of Radiology.
Publications:
-
Mehul Patel, Jeffrey J. Rodriguez, and Arthur F. Gmitro, "Image
classification based on focus," 2008 IEEE Intl. Conf. on Image Processing,
San Diego, CA, Oct. 12-15, 2008, pp. 397-400. [ PDF ]
-
Mehul Patel, Jeffrey J. Rodriguez, and Arthur F. Gmitro, "Effect
of gray-level re-quantization on co-occurrence based texture analysis," 2008
IEEE Intl. Conf. on Image Processing, San Diego, CA, Oct. 12-15, 2008, pp. 585-588. [ PDF ]
|
|