Discrimination of cancerous from benign pigmented skin lesions based on multispectral autofluorescence lifetime imaging dermoscopy and machine learning.

SIGNIFICANCE: Accurate early diagnosis of malignant skin lesions is critical in providing adequate and timely treatment; unfortunately, initial clinical evaluation of similar-looking benign and malignant skin lesions can result in missed diagnosis of malignant lesions and unnecessary biopsy of benign ones. AIM: To develop and validate a label-free and objective image-guided strategy for the clinical evaluation of suspicious pigmented skin lesions based on multispectral autofluorescence lifetime imaging (maFLIM) dermoscopy. APPROACH: We tested the hypothesis that maFLIM-derived autofluorescence global features can be used in machine-learning (ML) models to discriminate malignant from benign pigmented skin lesions. Clinical widefield maFLIM dermoscopy imaging of 41 benign and 19 malignant pigmented skin lesions from 30 patients were acquired prior to tissue biopsy sampling. Three different pools of global image-level maFLIM features were extracted: multispectral intensity, time-domain biexponential, and frequency-domain phasor features. The classification potential of each feature pool to discriminate benign versus malignant pigmented skin lesions was evaluated by training quadratic discriminant analysis (QDA) classification models and applying a leave-one-patient-out cross-validation strategy. RESULTS: Classification performance estimates obtained after unbiased feature selection were as follows: 68% sensitivity and 80% specificity with the phasor feature pool, 84% sensitivity, and 71% specificity with the biexponential feature pool, and 84% sensitivity and 32% specificity with the intensity feature pool. Ensemble combinations of QDA models trained with phasor and biexponential features yielded sensitivity of 84% and specificity of 90%, outperforming all other models considered. CONCLUSIONS: Simple classification ML models based on time-resolved (biexponential and phasor) autofluorescence global features extracted from maFLIM dermoscopy images have the potential to provide objective discrimination of malignant from benign pigmented lesions. ML-assisted maFLIM dermoscopy could potentially assist with the clinical evaluation of suspicious lesions and the identification of those patients benefiting the most from biopsy examination.

Light-driven photosensitizer uptake increases Candida albicans photodynamic inactivation.

Photodynamic Inactivation (PDI) is based on the use of a photosensitizer (PS) and light that results mainly in the production of reactive oxygen species, aiming to produce microorganism cell death. PS incubation time and light dose are key protocol parameters that influence PDI response; the correct choice of them can increase the efficiency of inactivation. The results of this study show that a minor change in the PDI protocol, namely light-driven incubation leads to a higher photosensitizer and more uniform cell uptake inside the irradiated zone. Furthermore, as the uptake increases, the damage caused by PDI also increases. The proposed light-driven incubation prior to the inactivation illumination dose has advantages when compared to the traditional PDI treatments since it can be more selective and effective. Using a violet light as pre-illumination (light-driven incubation) source and a red-light system as PDI source, it was possible to demonstrate that when compared to the traditional protocol of dark incubation, the pre-illuminated cell culture showed an inactivation increase of 7 log units. These in vitro results performed in Candida albicans cells may result in the introduction of a new protocol for PDI.

Quantification of total carbon in soil using laser-induced breakdown spectroscopy: a method to correct interference lines.

The C cycle in the Brazilian forests is very important, mainly for issues addressed to climate changes and soil management. Assessing and understanding C dynamics in Amazonian soils can help scientists to improve models and anticipate scenarios. New methods that allow soil C measurements in situ are a crucial approach for this kind of region, due to the costs for collecting and sending soil samples from the rainforest to the laboratory. Laser-induced breakdown spectroscopy (LIBS) is a multielemental atomic emission spectroscopy technique that employs a highly energetic laser pulse for plasma production and requires neither sample preparation nor the use of reagents. As LIBS takes less than 10 s per sample measurement, it is considered a promising technique for in situ soil analyses. One of the limitations of portable LIBS systems, however, is the common overlap of the emission lines that cannot be spectrally resolved. In this study a method was developed capable of separating the Al interference from the C emission line in LIBS measurements. Two typical forest Brazilian soils rich in Al were investigated: a spodosol (Amazon Forest) and an oxisol (Atlantic Forest). Fifty-three samples were collected and analyzed using a low-resolution LIBS apparatus to measure the intensities of C lines. In particular, two C lines were evaluated, at 193.03 and 247.86 nm. The line at 247.86 nm showed very strong interference with Fe and Si lines, which made quantitative analysis difficult. The line at 193.03 nm showed interference with atomic and ionic Al emission lines, but this problem could be solved by applying a correction method that was proposed and tested in this work. The line at 247.86 was used to assess the proposed model. The strong correlation (Pearson's coefficient R=0.91) found between the LIBS values and those obtained by a reference technique (dry combustion by an elemental analyzer) supported the validity of the proposed method.