Return to search

High resolution microendoscopy for quantitative diagnosis of esophageal neoplasia

Esophageal cancer is the eighth most common cancer in the world. Cancers of the esophagus account for 3.8% of all cases of cancers, with approximately 482,300 new cases reported in 2008 worldwide. In the United States alone, it is estimated that approximately 18,000 new cases will be diagnosed in 2013, and 15,210 deaths are expected. Despite advances in surgery and chemoradiation therapy, these advances have not led to a significant increase in survival rates, primarily because diagnosis often at an advanced and incurable stage when treatment is more difficult and less successful. Accurate, objective methods for early detection of esophageal neoplasia are needed.
Here, quantitative classification algorithms for high resolution miscroendoscopic images were developed to distinguish between esophageal neoplastic and non-neoplastic tissue. A clinical study in 177 patients with esophageal squamous cell carcinoma (ESCC) was performed to evaluate the diagnostic performance of the classification algorithm in collaboration with the Mount Sinai Medical Center in the United States, the First Hospital of Jilin University in China, and the Cancer Institute and Hospital, the Chinese Academy of Medical Science in China. The study reported a sensitivity and specificity of 93% and 92%, respectively, in the training set, 87% and 97%, respectively, in the test set, and 84% and 95%, respectively, in an independent validation set. Another clinical study in 31 patients with Barrett’s esophagus resulted in a sensitivity of 84% and a specificity of 85%. Finally, a compact, portable version of the high resolution microendoscopy (HRME) device using a consumer-grade camera was developed and a series of biomedical experimental studies were carried out to assess the capability of the device.

Identiferoai:union.ndltd.org:RICE/oai:scholarship.rice.edu:1911/72038
Date16 September 2013
CreatorsShin, Dong Suk
ContributorsRichards-Kortum, Rebecca R.
Source SetsRice University
LanguageEnglish
Detected LanguageEnglish
Typethesis, text
Formatapplication/pdf

Page generated in 0.0016 seconds