The label noise in AI-aided cancer diagnostics has various origins but often poses a challenge to the data analysis. Misclassified samples in the training set can lead to low accuracy of predictions. In this work, we present strategies of reducing the label noise in the context of dermatofluoroscopy (two-photon fluorescence excitation spectroscopy for early diagnosis of malignant melanoma) and support vector machines (SVMs). The experiments performed on real data set composed of 265 pigmented skin lesions confirm the hypothesis of reduced model accuracy in the presence of label noise. Relabeling and especially removing the supporting vector examples from the training set (100 skin lesions) allow for building models of very high predictive accuracy in diagnosing malignant melanoma as shown on independent data set (165 skin lesions). Furthermore, in the limit of very low data quantity, relabeling of supporting vectors and ensembling are shown to yield models that are more robust to label noise.
The penetration of spherical and rod-like gold nanoparticles into human skin is reported. Several skin preparation techniques are applied, including cryo techniques, such as plunge freezing and freeze drying, and the use of wet cells. Their advantages and drawbacks for observing nanoparticle uptake are discussed. Independent of the particle shape no uptake into intact skin is observed by a combination of imaging approaches, including scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX), and scanning X-ray microscopy (STXM). These results are discussed along with suitable skin preparation approaches. Experiments on barrier-disrupted skin, i.e. mechanical lesions made by pricking, indicate, however, that gold particles can be identified deep in the dermis, as follows from STXM studies on wet skin samples.
In order to investigate the penetration depth of silver nanoparticles (Ag NPs) inside the skin, porcine ears treated with Ag NPs are measured by two-photon tomography with a fluorescence lifetime imaging microscopy (TPT-FLIM) technique, confocal Raman microscopy (CRM), and surface-enhanced Raman scattering (SERS) microscopy. Ag NPs are coated with poly-N-vinylpyrrolidone and dispersed in pure water solutions. After the application of Ag NPs, porcine ears are stored in the incubator for 24 h at a temperature of 37°C. The TPT-FLIM measurement results show a dramatic decrease of the Ag NPs’ signal intensity from the skin surface to a depth of 4 μm. Below 4 μm, the Ag NPs’ signal continues to decline, having completely disappeared at 12 to 14 μm depth. CRM shows that the penetration depth of Ag NPs is 11.1±2.1 μm. The penetration depth measured with a highly sensitive SERS microscopy reaches 15.6±8.3 μm. Several results obtained with SERS show that the penetration depth of Ag NPs can exceed the stratum corneum (SC) thickness, which can be explained by both penetration of trace amounts of Ag NPs through the SC barrier and by the measurements inside the hair follicle, which cannot be excluded in the experiment.
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.