No abstract

Early diagnosis of precancerous and malignant lesions is critical for the improving of the current poor survival rate of patients with a variety of tumors. The development of new high-specificity and high-sensitivity imaging technologies can play an important role in the early diagnosis, accurate staging, and treatment of cancer. Bio-optical signals are the result of the optical functions of the biological systems, occurring naturally or induced by the measurement. The identification of the state of human skin tissues is discussed here. The Bio-optical signals recorded in vitro have been analyzed by extracting various statistical features. Using MATLAB programs, various statistical features are extracted from both normal and pathology spectra. Different features like mean, summation, skewness, etc were extracted. The values of the feature vector reveal information regarding tissue state. These parameters have been analyzed for the discrimination between normal and pathology conditions. For analysis, a specific data set has been considered. Using K-means clustering, signal classification was done in MATLAB. Clustering is the process of partitioning or grouping a given set of patterns into disjoint clusters. This is done such that patterns in the same cluster are alike and patterns belonging to two different clusters are different. Clustering has been a widely studied problem in a variety of application domains including neural networks, statistics etc.

No abstract

Early diagnosis of precancerous and malignant lesions is critical for the improving of the current poor survival rate of patients with a variety of tumors. The development of new high-specificity and high-sensitivity imaging technologies can play an important role in the early diagnosis, accurate staging, and treatment of cancer. Bio-optical signals are the result of the optical functions of the biological systems, occurring naturally or induced by the measurement. The identification of the state of human skin tissues is discussed here. The Bio-optical signals recorded in vitro have been analyzed by extracting various statistical features. Using MATLAB programs, various statistical features are extracted from both normal and pathology spectra. Different features like mean, energy etc were extracted. The values of the feature vector reveal information regarding tissue state. Then a classical multilayer feed forward neural network with back propagation algorithm is employed to serve as a classifier of the feature vector, giving 100% successful results for the specific data set considered. Furthermore, a small time needed to acquire and analyze the fluorescence spectra together with high rates of success, proves our method very attractive for real time applications.