Rakefet Czerninski, DMD, School of Dental Medicine, Department of Oral Medicine, The Hebrew University of Jerusalem and Hadassah University Hospital and Yoav Smith, PhD, Genomic Data Analysis Unit, The Hebrew University of Jerusalem.
Background
Timely analysis of lesions is of utmost importance in order to detect instances where the lesion is malignant and needs to be excised before it metastasizes. In many types of cancer, recovery rates are unequivocally related to early detection, for instance in skin cancer, when the lesion is detected before it has metastasized, complete recovery is seen in 99% of all patients treated.
In current practice of dermal analysis and of oral and dental lesion analysis, the physician typically observes the suspected area with the naked eye and based on parameters of size, color and other aspects of the visual appearance, decides whether the lesion warrants further diagnostic testing or treatment. The examination is therefore highly subjective, and is dependant upon his/her powers of observation and experience. Photographic images are influenced by ambient lighting or different camera and flash affecting the appearance of the digital image.
Even after normalization of the colors of digitized images, the challenge remains to determine whether the colors have been affected over time. This is important for instance, in analysis and diagnosis of clinical oral images, where lesions in which the color becomes more intensely red over time can signify malignancy, infection or inflammation (and less intensity implies healing or reaction to treatment). Changes in the color intensity of a dermal lesion that becomes browner, or more intensely gray, can signify melanoma. Erythema changes can help evaluating any process of inflammation or infection. The need exists for an objective and quantitative method of analyzing the color and the size changes occurring in a lesion over time or compared to normal tissue.
Market
Each year, more than 30,000 new cases of oral and pharyngeal cancer are diagnosed and over 8,000 deaths due to oral cancer occur. The 5-year survival rate for these cancers is only about 50 percent. Methods used to treat oral cancers (surgery, radiation, and chemotherapy) are disfiguring and costly. Early detection is key to increasing the survival rate for these cancers.
The potential also exists to reach these patients via telemedicine.
The Innovation
The invention allows the user to determine the extent and therefore the medical significance of color and size changes that are present in a tissue or organ of the human body by quantification of the intensity of the colors. The technology provides a method for computerized analysis of color and size present in at least one image of the human body, comprising:
- placing a reference label having a plurality of reference colors in the vicinity of a tissue or organ of the human body;
- capturing a color image including the tissue or organ and the reference label, the image is a digital image or is converted into a digital image;
- performing color and size calibration of the image using the LAB color system;
- selecting the borders of an area of interest within the image;
- calculating the intensity and distribution of color within the area of interest using at least one channel of the LAB system, to give at least one color information attribute.
R&D Program
A prototype of the SW and calibration label with user-friendly interface has been developed and tested. The next stage is automation of the calibration process and further improvement of the user interface. The final stage is to prepare the technology for telemedicine via the internet.
Contact
Stuart Bernstein
Business Development, Medical Devices
Tel: +972-2-6777906
Email: stuart@hadasit.co.il
