! Alert Updates on COVID-19 assistance. Learn more
RESEARCH & TECHNOLOGY
The patented FAIS-Moffitt Centrosome Image Analysis software has proven effective for use in medical image analysis and has seamlessly integrated with health records to improve efficiency. Working in partnership with the Moffitt Cancer Center, our research pioneers new visions for cancer treatments. One valuable application of our technology within the biomedical imaging field is our Centrosome Image System that discriminates cancer from normal cells in cell-based and histology specimens. The revolutionary Centrosome Image System can be used for the early diagnosis of cancer, prognosis of cancers’ progression and in evaluation of cancer treatment response.
Background - The Centrosome
Background - The Centrosome
The Structure of the Centrosome.
The Centrosome is a small region near the nucleus of cells that plays a critical role in organizing microtubules, while maintaining cellular polarity and chromosome segregation during mitosis. Chromosome instability commonly seen in human malignant tumors could result from defects in the centrosome. Centrosome abnormalities are found in many types of cancers including breast, lung, pancreas, colorectal, prostate, ovarian, bone, head and neck, and are rarely observed in normal tissue. Centrosomal defects have been found to occur at a very early premalignant stage of tumor formation, prior to the appearance of detectable lesions. Centrosomal defects also increase in severity during tumor progression.
The Centrosome Image System analyzes centrosome abnormalities in cell-based and histology specimens. We offer immunofluorescence centrosome labeling, acquisition of centrosome images following quantitative calculation, statistical analysis of centrosomal features, and classification for diagnosis or prognosis.
In doing this, Centrosome Image System may allow for individual prediction of cancer treatment response, and potentially serve as anticancer drug sensitivity test for lung, breast (especially, for triple-negative breast cancers), prostate, ovarian, pancreas and other cancers. The Centrosome Image System provides valuable imaging information for clinical practice and could become the adjuvant therapy choice for patients’ survival rate improvement.
The Centrosome Image System analyzes centrosome abnormalities in cell-based and histology specimens. We offer immunofluorescence centrosome labeling, acquisition of centrosome images following quantitative calculation, statistical analysis of centrosomal features, and classification for diagnosis or prognosis.
In doing this, Centrosome Image System may allow for individual prediction of cancer treatment response, and potentially serve as anticancer drug sensitivity test for lung, breast (especially, for triple-negative breast cancers), prostate, ovarian, pancreas and other cancers. The Centrosome Image System provides valuable imaging information for clinical practice and could become the adjuvant therapy choice for patients’ survival rate improvement.
Software
Our Software has proven effective for use in medical image analysis and has seamlessly integrated with health records to improve efficiency. One valuable application of our technology within the biomedical imaging field is our Centrosome Image System. It discriminates cancer from normal cells in cell-based & histology specimens.
The revolutionary Centrosome Image System can be used for the early diagnosis of cancer, prognosis of cancers’ progression & in evaluation of cancer treatment. FAIS-Moffit Centrosome Image Analysis patent has been issued.
Our Software has proven effective for use in medical image analysis and has seamlessly integrated with health records to improve efficiency. One valuable application of our technology within the biomedical imaging field is our Centrosome Image System. It discriminates cancer from normal cells in cell-based & histology specimens.
The revolutionary Centrosome Image System can be used for the early diagnosis of cancer, prognosis of cancers’ progression & in evaluation of cancer treatment. FAIS-Moffit Centrosome Image Analysis patent has been issued.
Project
Automatic ROI Selection in Centrosome Analysis for Very Earlier Cancer Diagnosis
Region of interest (ROI) selection is commonly used in medical image processing. ROI selection has been carried out manually at many hospitals for many years and this method is still used. Manual ROI selection is time inefficient, therefore economically inefficient. Manual ROI selection is also subjective and lack of consistency. Because of these reasons, people started to work on semi-automatic ROI selection and fully-automatic ROI selection. As database become larger and larger, fully-automatic ROI selection will be the only viable choice.
Our team started from using manual ROI selection, and developed a semi-automatic ROI selection later on. It worked well but not as efficient as fully-automatic ROI selection, especially for a large database. A fully-automatic ROI selection software have been developed recently. Its accurate rate reached about 97% on our database. With selected ROI, our image processing work will be easier and the performance will be better. That is obvious as it is much easier to work on a smaller and simpler ROI image that only includes one cell and its centrosomes than working on a huge, complicated frame image with many cells and other tissues.
For more details, please read the "Automatic ROI Selection in Centrosome Analysis" presentation.
Automatic ROI Selection in Centrosome Analysis for Very Earlier Cancer Diagnosis
Region of interest (ROI) selection is commonly used in medical image processing. ROI selection has been carried out manually at many hospitals for many years and this method is still used. Manual ROI selection is time inefficient, therefore economically inefficient. Manual ROI selection is also subjective and lack of consistency. Because of these reasons, people started to work on semi-automatic ROI selection and fully-automatic ROI selection. As database become larger and larger, fully-automatic ROI selection will be the only viable choice.
Our team started from using manual ROI selection, and developed a semi-automatic ROI selection later on. It worked well but not as efficient as fully-automatic ROI selection, especially for a large database. A fully-automatic ROI selection software have been developed recently. Its accurate rate reached about 97% on our database. With selected ROI, our image processing work will be easier and the performance will be better. That is obvious as it is much easier to work on a smaller and simpler ROI image that only includes one cell and its centrosomes than working on a huge, complicated frame image with many cells and other tissues.
For more details, please read the "Automatic ROI Selection in Centrosome Analysis" presentation.