Researchers at the Center for Computational Imaging and Personalized Diagnostics, including Anant Madabhushi, the F. Alex Nason Professor of Biomedical Engineering, were issued patents for their work. Cheng Lu, senior research associate with the Center for Computational Imaging and Personalized Diagnostics, was part of the team that worked on one of the patents.
“Computerized analysis of computed tomography (CT) imagery to quantify tumor infiltrating lymphocytes (TILs) in non-small cell lung cancer (NSCLC)”
Madabhushi and colleagues affiliated with the Center for Computational Imaging and Personalized Diagnostics were recently issued a patent United States Serial Number: 10,346,975 titled “Computerized analysis of computed tomography (CT) imagery to quantify tumor infiltrating lymphocytes (TILs) in non-small cell lung cancer (NSCLC).”
Tumor infiltrating lymphocytes (TILs) are a part of the dynamic immune microenvironment. Increased levels of TILs are associated with better clinical outcomes in diverse human neoplasms, including melanoma, colorectal cancer, triple-negative carcinomas and non-small cell lung cancer (NSCLC). Clinical trials with immune checkpoint inhibitors report significant increase in TILs in responders to treatment in follow up biopsies. However, since biopsies are invasive, time-consuming, expensive, and may expose a patient to significant side effects, it would be beneficial to more accurately and non-invasively determine which patients are more likely to exhibit increased levels of TILs. This patent relates to methods, apparatus and other embodiments that predict tumor infiltrating lymphocyte (TIL) density from pre-surgical computed tomography images of a region of tissue demonstrating non-small cell lung cancer.
Other co-inventors at Case Western Reserve University include Mehdi Alilou, a senior research associate, and Niha Beig, a graduate student.
“Multi-pass adaptive voting for nuclei detection in histopathological images”
Madabhushi and Lu were recently issued a patent United States Serial Number: 10,360,434 titled “Multi-pass adaptive voting for nuclei detection in histopathological images.”
Nuclei detection is an initial step in the development of computer-aided diagnosis and automated tissue grading schemes in the context of digital pathology images. Accurate nuclei detection in images that have poor staining or noise is a challenging task. Nuclear clusters that result from tissue sectioning artifacts also increase the challenge in accurately detecting nuclei. Manually identifying cellular nuclei, including identifying the location and extent of melanocyte invasion or breast cancer nuclei, is subjective and time consuming. Since nuclei detection is often a critical step in computer-aided diagnosis and prognosis schemes in the context of digital pathology images, it would be beneficial to more accurately detect and count nuclei in histopathological images. One example method includes accessing a histopathology image that includes a plurality of pixels, generating a gradient field map based on the histopathology image, generating a refined gradient field map based on the gradient field map, calculating a voting map for a member of the plurality of pixels based, at least in part, on the refined gradient field map and a voting kernel, generating an aggregated voting map based on the voting map, computing a global threshold, and identifying a nuclear centroid based on the global threshold and the aggregated voting map.