Abstract:Reflectance imaging of biological tissues with visible and near-infrared light has the significant potential to provide a noninvasive and safe imaging modality for diagnosis of dysplastic and malignant lesions in the superficial tissue layers. The difficulty in the extraction of optical and structural parameters lies in the lack of efficient methods for accurate modeling of light scattering in biological tissues of turbid nature. We present a parallel Monte Carlo method for accurate and efficient modeling of reflectance images from turbid tissue phantoms. A parallel Monte Carlo code has been developed with the message passing interface and evaluated on a computing cluster with 16 processing elements. The code was validated against the solutions of the radiative transfer equation on the bidirectional reflection and transmission functions. With this code we investigated numerically the dependence of reflectance image on the imaging system and phantom parameters. The contrasts of reflectance images were found to be nearly independent of the numerical aperture (NA) of the imaging camera despite the fact that reflectance depends on the NA. This enables efficient simulations of the reflectance images using an NA at 1.00. Using heterogeneous tissue phantoms with an embedded region simulating a lesion, we investigated the correlation between the reflectance image profile or contrast and the phantom parameters. It has been shown that the image contrast approaches 0 when the single-scattering albedos of the two regions in the heterogeneous phantoms become matched. Furthermore, a zone of detection has been demonstrated for determination of the thickness of the embedded region and optical parameters from the reflectance image profile and contrast. Therefore, the utility of the reflectance imaging method with visible and near-infrared light has been firmly established. We conclude from these results that the optical parameters of the embedded region can be determined inversely from reflectance images acquired with full-field illumination at multiple incident angles or multiple wavelengths. (C) 2007 American Association of Physicists in Medicine.

Multivariate analysis of Monte Carlo generated images for diagnosis of dysplastic lesions

In Vivo Measurement and Modeling of Multispectral Reflectance Images for Melanoma Diagnosis

Physics Based Contrast Marking and Inpainting Based Local Texture Comparison for Clustered Microcalcification Detection

Combined Classifier for Discriminating Cancerous Tissue from Normal Tissue Using Light-Induced Autofluorescence

Distribution Quantification on Dermoscopy Images for Computer-Assisted Diagnosis of Cutaneous Melanomas

Classification of in Vivo Autofluorescence Spectra Using Support Vector Machines

Recovering Skin Reflectance And Geometry For Diagnosis Of Melanoma

Early Melanoma Diagnosis With Sequential Dermoscopic Images

Numerical study of reflectance imaging using a parallel Monte Carlo method.

Novel strategy for applying hierarchical density-based spatial clustering of applications with noise towards spectroscopic analysis and detection of melanocytic lesions

Unsupervised sub-segmentation for pigmented skin lesions.

Biological indexes based reflectional asymmetry for classifying cutaneous lesions.

In Vivo Multispectral Imaging for Diagnosis of Melanoma

Melanoma Diagnosis with Spatio-Temporal Feature Learning on Sequential Dermoscopic Images

Classification of melanocytic lesions using direct illumination multispectral imaging

Combining hyperspectral imaging techniques with deep learning to aid in early pathological diagnosis of melanoma

Deep Learning With Optical Coherence Tomography for Melanoma Identification and Risk Prediction

Malignancy Prediction and Lesion Identification from Clinical Dermatological Images

A Computer-Aided Spectroscopic System for Early Diagnosis of Melanoma

Optical transfer diagnosis of pigmented lesions

Multimodality Computer-Aided Breast Cancer Diagnosis with FFDM and DCE-MRI