Dermatologists are starting to make use of Computer-Aided Diagnosis based on deep learning algorithms, which can provide them with an objective judgement during evaluation of equivocal lesions. DL algorithms can be trained to classify skin lesions with datasets of diverse nature like traditional RGB, clinical and dermoscopic images, or more experimentally, with images from other modalities, such as multispectral imaging. In this work, we have evaluated and customized the different DL approaches that exist in the state of the art to classify a dataset of +500 images acquired on skin lesions. The images were acquired with a staring multispectral imaging prototype in the visible and near-infrared ranges. The best results were obtained for a customized model VGG-16 that combined 3D convolutional layers, 3D maxpooling layers and dropout regularization, leading to an overall accuracy of 71%.
Spectral reflectance of the eye fundus was evaluated in adult healthy patients through a fast visible and near-infrared multispectral fundus camera. Spectral signatures were analyzed for different ocular structures of the retina and the choroid.
We present a multispectral fundus camera that performs fast imaging of the ocular posterior pole in the visible and near-infrared (400 to 1300 nm) wavelengths through 15 spectral bands, using a flashlight source made of light-emitting diodes, and CMOS and InGaAs cameras. We investigate the potential of this system for visualizing occult and overlapping structures of the retina in the unexplored wavelength range beyond 900 nm, in which radiation can penetrate deeper into the tissue. Reflectance values at each pixel are also retrieved from the acquired images in the analyzed spectral range. The available spectroscopic information and the visualization of retinal structures, specifically the choroidal vasculature and drusen-induced retinal pigment epithelium degeneration, which are hardly visible in conventional color fundus images, underline the clinical potential of this system as a new tool for ophthalmic diagnosis.
Eye fundus photography routinely used in clinical practice is restricted to color imaging of the retina. In the last years, hyperspectral imaging has shown to be a powerful tool for the spectral analysis of biological tissue. In this study, we present a fully custom-made fast hyperspectral fundus camera based on light emitting diodes (LED) with 15 different wavelengths of emission and with extended spectral sensitivity towards the near infrared (NIR) (from 400 nm to 1300 nm), which allows imaging deeper retinal layers, including the choroid, than current clinical devices. These new features will be very useful for a better understanding of ocular diseases as well as aiding in their diagnosis.
The effective and non-invasive diagnosis of skin cancer is a hot topic in biophotonics since the current gold standard, biopsy followed by histological examination, is a slow and costly procedure for the healthcare system. Therefore, authors have put their efforts in characterizing skin cancer quantitatively through optical and photonic techniques such as 3D topography and multispectral imaging. Skin relief is an important biophysical feature that can be difficult to appreciate by touch, but can be precisely characterized with 3D imaging techniques, such as fringe projection. Color and spectral features given by skin chromophores, which are routinely analyzed by the naked eye and through dermoscopy, can also be quantified by means of multispectral imaging systems. In this study, the outcomes of these two imaging modalities were combined in a machine learning process to enhance classification of melanomas and nevi obtained from the two systems when operating isolately. The results suggest that the combination of 3D and multispectral data is relevant for the medical diagnosis of skin cancer.
We study experimentally the operating conditions of a semiconductor laser diode subjected to different amounts of optical feedback in order to find a stable and cost-efficient solution for speckle reduction in double-pass retinal imaging.
The double pass imaging method is used to obtain the point spread function of a patient’s eye; however it suffers from speckle formation. Here we present a comparison of speckle formation in double pass imaging using three different semiconductor-based light sources.
This paper shows the simulations of the usage of a LED cluster as the illumination source for a multispectral imaging
system covering the range of wavelengths from 350 to 1650 nm. The system can be described as being composed of two
modules determined by the spectral range of the imaging sensors responses, one of them covering the range from 350-
950nm (CCD camera) and the other one covering the wavelengths from 900-1650nm (InGaAs camera). A well known
method of reflectance estimation, the pseudo-inverse method, jointly with the experimentally measured data of the
spectral responses of the cameras and the spectral emission of the LED elements are used for the simulations. The
performance of the system for spectral estimation under ideal conditions and realistic noise influence is evaluated
through different spectral and colorimetric metrics like the GFC, RMS error and CIEDE2000 color difference formula.
The results show that is expectable a rather good performance of the real setup. However, they also reveal a difference in
the performances of the modules. The second module has poorer performance due to the less narrow spectral emission
and less number of LED elements that covers the near-infrared spectral range.
This work is focused on the study and comparison of the performance for color measurements of different systems based on optoelectronic imaging sensors. We used two different configurations of the imaging system, one with three acquisition channels and the other with more spectral bands, in order to measure the color associated to each pixel of the captured scene. We applied different methodologies to obtain the XYZ tristumulus values from the measured digital signals. The different techniques included an absolute spectral and colorimetric characterization of the system and also direct transformations between both sets, which used several mathematical fittings such as the pseudo-inverse technique, a non-linear estimation method and the principal component analysis. The proposed configurations were experimentally tested imaging the patches of the Gretagmacbeth ColorChecker DC and Color Rendition charts placed in
a light booth, and measuring the corresponding colors. The results obtained showed that optoelectronic imaging systems can be used in order to perform rather accurate color measurements with high spatial resolution. Specifically, the best results in terms of CIELab color differences were achieved by using a multispectral configuration of the imaging system with seven spectral bands and directly transforming the digital signals into XYZ tristimulus values by means of the pseudo-inverse technique.
The near infrared spectral region (NIR) is useful in many applications. These include agriculture, the food and chemical industry, and textile and medical applications. In this region, spectral reflectance measurements are currently made with conventional spectrophotometers. These instruments are expensive since they use a diffraction grating to obtain monochromatic light. In this work, we present a multispectral imaging based technique for obtaining the reflectance spectra of samples in the NIR region (800 - 1000 nm), using a small number of measurements taken through different channels of a conventional CCD camera. We used methods based on the Wiener estimation, non-linear methods and principal component analysis (PCA) to reconstruct the spectral reflectance. We also analyzed, by numerical simulation, the number and shape of the filters that need to be used in order to obtain good spectral reconstructions. We obtained the reflectance spectra of a set of 30 spectral curves using a minimum of 2 and a maximum of 6 filters under the influence of two different halogen lamps with color temperatures Tc1 = 2852K and Tc2 = 3371K. The results obtained show that using between three and five filters with a large spectral bandwidth (FWHM = 60 nm), the reconstructed spectral reflectance of the samples was very similar to that of the original spectrum. The small amount of errors in the spectral reconstruction shows the potential of this method for reconstructing spectral reflectances in the NIR range.
An algorithm is proposed for the spectral and colorimetric characterization of digital still cameras (DSC) which allows to use them as tele-colorimeters with CIE-XYZ color output, in cd/m2. The spectral characterization consists of the calculation of the color-matching functions from the previously measured spectral sensitivities. The colorimetric characterization consists of transforming the RGB digital data into absolute tristimulus values CIE-XYZ (in cd/m2) under variable and unknown spectroradiometric conditions. Thus, at the first stage, a gray balance has been applied over the RGB digital data to convert them into RGB relative colorimetric values. At a second stage, an algorithm of luminance adaptation vs. lens aperture has been inserted in the basic colorimetric profile. Capturing the ColorChecker chart under different light sources, the DSC color analysis accuracy indexes, both in a raw state and with the corrections from a linear model of color correction, have been evaluated using the Pointer'86 color reproduction index with the unrelated Hunt'91 color appearance model. The results indicate that our digital image capture device, in raw performance, lightens and desaturates the colors.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.