Skip to main content

CIA Lab Has Four Manuscripts for Publication

The CIA Lab had four new manuscripts accepted for publication in Proceedings of SPIE Medical Imaging to be held in San Diego in February 2014.

 “Segmentation and Automated Measurement of Chronic Wound Images: Probability Map Approach” by Fauzi, Khansa, Catignani, Gordillo, Sen, and Gurcan, details a new method to characterize chronic wounds containing granulation, slough, and eschar tissues. The work is innovative due to defining wound characteristics specific to the probability map used for segmentation, the computationally efficient regions that grow with channel selection, and the ability to auto-calibrate the measurements based on the content of the image. The computer accuracy was 82.4% while the inter-reader agreement was 85.5%. This project was done in collaboration with the Department of Plastic Surgery-The Ohio State University Wexner Medical Center, Department of Internal Medicine-The Ohio State University Wexner Medical Center, Department of Surgery-The Ohio State University Wexner Medical Center, OSU Comprehensive Wound Center-The Ohio State University Wexner Medical Center, and the Faculty of Engineering, Multimedia University in Cyberjaya, Selangor, Malaysia.

 “Classification of Glioblastoma and Metastasis for Neuropathology Intraoperative Diagnosis: A Multi-Resolution Textural Approach to Model the Background” by Fauzi, Gokozan, Elder, Puduvalli, Otero, and Gurcan, details a texture based classification of glioblastoma and metastatic cancer. The regions of interest in glioblastoma appear between the nuclei as anisotropic thin linear structures while the features are more homogeneous for metastasis. Discrete Wavelet Frames were used to characterize the texture after the nuclei regions were segmented out using the decomposition segmentation algorithm. The accuracy for the glioblastomas was 80% while the metastasis was 87.5% resulting in an overall accuracy of 83.5%. This project was done in collaboration with the Department of Pathology-The Ohio State University, Department of Neurological Surgery-The Ohio State University, Division of Neuro-oncology-The Ohio State University Wexner Medical Center, and the Faculty of Engineering, Multimedia University in Cyberjaya, Selangor, Malaysia.

 “Hot Spot Detection in Breast Cancer Slides with Image Filtering” by Niazi, Downs-Kelly, and Gurcan, proposes a solution to the computationally demanding problem of hot spot detection from larger images. This new computational framework uses image filtering to detect the positively stained pixels. The Ki-67 image is segmented using VMS and then an image dependent filter is generated which in turn generates a density map. The smoothness of this density map simplifies the detection of the local maxima which is what correlates to the number of hot spots in the image. This method detected hot spots with 81% precision compared to 57% intra-reader variability. This project was done in collaboration with the Department of Anatomic Pathology-Cleveland Clinic.


 “Grading Vascularity from Histopathological Images Based on Traveling Salesman Distance and Vessel Size” by Niazi, Hemminger, Kurt, Lozanski, and Gurcan, details a new method to determine characteristics of vessels. Currently, the quantification is done manually and is not easily reproduced and is only semi-quantitative. The new method automatically segments the blood vessels by maxlink thresholding and minimum graph cuts. Then, there is morphological post-processing which reduces blast and small spurious objects from the images. 20 features were extracted to classify the images into one of four grades. These features included the first four moments of the distribution of the area of blood vessels, the edge weights in the minimum spanning tree of the blood vessels, the shortest distance between blood vessels, the homogeneity of the shortest distance, and blood vessel orientation. This method resulted in 68.18% grading accuracy as opposed to intra- and inter-reader variability of 66.87% and 40%. This project was done in collaboration with the Department of Pathology-The Ohio State University. 

Comments

Post a Comment

Popular posts from this blog

CIALAB encouraging talented young minds with summer internships

CIALAB is pleased to introduce the three interns namely Tong Gan, Rosana Rodriguez Milanes and Michael Priddy working through summer’09. Rosana Rodriguez Milanes - I am a third year undergraduate student in Electronic Engineering from Universidad del Norte, Colombia. My experience as a volunteer foreign student in the Clinical Image Analysis Laboratory has been an edifying, gratifying and enriching. Being able to participate, to learn and to collaborate in the Clinical Image Analysis Laboratory during the past two weeks has allowed me to improve my analytical and interpretative skills in processing histopathological and MRI images. I have been able to learn about segmentation, region growing, splitting and merging algorithms development. I have also had the privilege of knowing and interacting with excellent engineers who have helped me improve my skills as a foreign student. I am grateful for the opportunity that the Ohio State University has given me to collaborate and to learn with
1 comment
Read more

Recent publications

The CIA lab has recently had 4 articles published in PLOS One and the Journal of Urology. Automated Staging Of T1 Bladder Cancer Using Digital Pathologic H&E Images: A Deep Learning approach (Journal of Urology). The paper discusses the need for accurately gauging tumor cell intrusion into Lamina Propria in an effort to substage bladder cancer. It explains how transfer learning in conjunction with Convolutional Neural Networks can be used to accurately identify different bladder layers and then compute the distance between tumor nuclei and Lamina Propria. The article is available here:  https://www.jurology.com/article/S0022-5347(18)41148-2/pdf Identifying tumor in pancreatic neuroendocrine neoplasms from Ki67 images using transfer learning (PLOS One). This paper examines a proposed methodology to automatically differentiate between NET and non-tumor regions based on images of Ki67 stained biopsies. It also uses transfer learning to exploit a rich set of features developed
6 comments
Read more

Drs. Goceri, Kus, and Senaras Present at OSUMC Research Day

On April 25th, Drs. Goceri, Kus, and Senaras presented their research at the OSUMC Research Day.  Dr. Evgin Goceri presented “Automatic and Robust Segmentation of Liver and Its Vessels from MR Datasets for Pre-Evaluation of Liver Translation” which proposed a robust and fully automated method for segmenting the liver and its vessels from MR images. Dr. Goceri’s study presented a novel approach to reducing processing time by employing binary regularization of the level set function. The fully-automatic segmentation of liver and its vessels with the proposed method was more efficient than manual approach and the other methods in the literature in terms of processing time and accuracy. Dr. Pelin Kus presented “Segmentation and Quantification of Tissue Necrosis in Tuberculosis” which focuses on how the immune system of patients infected with M. tuberculosis responds by using many types of cells including macrophages that form granulomas within the pulmonary tissue. Segmentation a
3 comments
Read more