Mesothelioma is a rare though often fatal disease, often associated with exposure to asbestos. The differentiation of malignant mesothelioma from benign mesothelial proliferation is often achieved by assessing tissue invasiveness from histological examination of biopsy specimens. Pleural effusion (fluid between lung and the chest wall) cytology is often a less invasive procedure that can be made available in most hospitals. Diagnosis of this malignancy from cytology specimens is often not possible, given that tissue invasiveness cannot be assessed. Our proposal details a pilot project to be conducted in collaboration with the System Chair of Pathology and Laboratory Medicine of Western Pennsylvania Health System (Jan F. Silverman, M.D.), as well as the Anatomic Pathology Medical Director of the Cleveland Clinic (Tarik M. Elsheikh, M.D.) to test an image analysis approach most recently developed by Dr. Rohde's laboratory at Carnegie Mellon University (CMU) for cancer detection directly from pleural effusion cytology. Preliminary results resulting from the application of Dr. Rohde's methods to data provided by Dr. Silverman (detailed in the application) indicates the method can potentially be useful for diagnosis of malignant mesothelioma. The objective of the pilot study would be to test the accuracy of the method developed by Dr. Rohde's laboratory in differentiating malignant mesothelioma from benign mesothelial proliferation directly from effusion cytology with a larger cohort of patients from two institutions (WPAHS and CC). If successful, the completion of the project could have a significant impact in reducing overall costs and avoiding health risks associated with invasive procedures. The successful completion of this project could also pave the way for the application of similar techniques on several similar cytopathology diagnostic and prognostic challenges.
The goal of this project proposal is to test advanced computational image analysis methods for the automated detection of malignant mesothelioma directly from pleural effusion cytology smears. Mesothelioma is a rare and often fatal disease whose diagnosis usually depends on determining invasion from an image of a tissue section obtained through an invasive biopsy. If successful, the technique has the potential to greatly facilitate diagnosis, reduce costs, and health risks associated with invasive biopsies in this and other cytopathology challenges.
|Pantanowitz, Liron; Liu, Chi; Huang, Yue et al. (2017) Impact of Altering Various Image Parameters on Human Epidermal Growth Factor Receptor 2 Image Analysis Data Quality. J Pathol Inform 8:39|
|Thorpe, Matthew; Park, Serim; Kolouri, Soheil et al. (2017) A Transportation Lp Distance for Signal Analysis. J Math Imaging Vis 59:187-210|
|Kolouri, Soheil; Park, Serim; Thorpe, Matthew et al. (2017) Optimal Mass Transport: Signal processing and machine-learning applications. IEEE Signal Process Mag 34:43-59|
|Hanna, Matthew G; Liu, Chi; Rohde, Gustavo K et al. (2017) Predictive Nuclear Chromatin Characteristics of Melanoma and Dysplastic Nevi. J Pathol Inform 8:15|
|Kolouri, Soheil; Tosun, Akif B; Ozolek, John A et al. (2016) A continuous linear optimal transport approach for pattern analysis in image datasets. Pattern Recognit 51:453-462|
|Kolouri, Soheil; Park, Se Rim; Rohde, Gustavo K (2016) The Radon Cumulative Distribution Transform and Its Application to Image Classification. IEEE Trans Image Process 25:920-34|
|Liu, Chi; Shang, Fei; Ozolek, John A et al. (2016) Detecting and segmenting cell nuclei in two-dimensional microscopy images. J Pathol Inform 7:42|
|Tosun, Akif Burak; Yergiyev, Oleksandr; Kolouri, Soheil et al. (2015) Detection of malignant mesothelioma using nuclear structure of mesothelial cells in effusion cytology specimens. Cytometry A 87:326-33|
|Rohde, Gustavo K (2015) Computational analysis of cell images. Cytometry A 87:294-5|