Enhanced visualization of blood and pigment in multispectral skin dermoscopy

Multispectral Imaging for Skin Diseases Assessment-State of the Art and Perspectives

Skin optical inspection is an imperative procedure for a suspicious dermal lesion since very early skin cancer detection can guarantee total recovery. Dermoscopy, confocal laser scanning microscopy, optical coherence tomography, multispectral imaging, multiphoton laser imaging, and 3D topography are the most outstanding optical techniques implemented for skin examination. The accuracy of dermatological diagnoses attained by each of those methods is still debatable, and only dermoscopy is frequently used by all dermatologists. Therefore, a comprehensive method for skin analysis has not yet been established. Multispectral imaging (MSI) is based on light-tissue interaction properties due to radiation wavelength variation. An MSI device collects the reflected radiation after illumination of the lesion with light of different wavelengths and provides a set of spectral images. The concentration maps of the main light-absorbing molecules in the skin, the chromophores, can be retrieved using the intensity values from those images, sometimes even for deeper-located tissues, due to interaction with near-infrared light. Recent studies have shown that portable and cost-efficient MSI systems can be used for extracting skin lesion characteristics useful for early melanoma diagnoses. This review aims to describe the efforts that have been made to develop MSI systems for skin lesions evaluation in the last decade. We examined the hardware characteristics of the produced devices and identified the typical structure of an MSI device for dermatology. The analyzed prototypes showed the possibility of improving the specificity of classification between the melanoma and benign nevi. Currently, however, they are rather adjuvants tools for skin lesion assessment, and efforts are needed towards a fully fledged diagnostic MSI device.

Infrared Macrothermoscopy Patterns-A New Category of Dermoscopy

(1) Background: The authors developed a new non-invasive dermatological infrared macroimaging analysis technique (MacroIR) that evaluates microvascular, inflammatory, and metabolic changes that may be dermoscopy complimentary, by analyzing different skin and mucosal lesions in a combined way-naked eye, polarized light dermatoscopy (PLD), and MacroIR-and comparing results; (2) Methods: ten cases were evaluated using a smartphone coupled with a dermatoscope and a macro lens integrated far-infrared transducer into specific software to capture and organize high-resolution images in different electromagnetic spectra, and then analyzed by a dermatologist; (3) Results: It was possible to identify and compare structures found in two dermoscopic forms. Visual anatomical changes were correlated with MacroIR and aided skin surface dermatological analysis, presenting studied area microvascular, inflammatory, and metabolic data. All MacroIR images correlated with PLD, naked eye examination, and histopathological findings; (4) Conclusion: MacroIR and clinic dermatologist concordance rates were comparable for all dermatological conditions in this study. MacroIR imaging is a promising method that can improve dermatological diseases diagnosis. The observations are preliminary and require further evaluation in larger studies.

Preoperative assessment of cutaneous melanoma thickness by multispectral dermoscopy

Preoperative assessment of Breslow thickness by means of sonography and clinical and dermoscopic criteria in white light dermoscopy has been reported, but up until now, the use of multispectral dermoscopy has not been investigated. Aim of this research is to determine whether multispectral dermoscopy and more specifically pigment maps can be used as a predictive marker for Breslow thickness in melanoma. Pigment maps are generated in real time from multispectral dermoscopic images and help to visualize the presence of pigment in a lesion. Multispectral images of 110 melanomas were collected, using a digital handheld multispectral dermatoscope, and assessed independently by five observers for the presence or absence of deep pigment compared with the surrounding skin. According to histopathological examination, the mean Breslow thickness of all 110 melanomas was 1.04 mm (ranging from 0.1 to 14 mm). The group of melanomas where deep pigment was visualized on the multispectral image (n = 78) had a significantly higher Breslow thickness (1.19 mm) than the group where no deep pigment was observed (n = 32, mean Breslow 0.68 mm) (P = 0.025). This study is unique in preoperative assessment of tumour thickness by means of multispectral dermoscopy. Our data indicate that the presence of deep pigment as visualized in digital dermoscopic skin parameter maps identifies a group of thicker melanomas. Further prospective research is needed to validate these pigment maps, generated by multispectral dermoscopy as a measure to predict invasiveness in melanoma.

Study protocol of the BASINEL Study: a pragmatic randomised controlled trial investigating treatment versus no treatment of low-risk basal cell carcinomas in older persons

INTRODUCTION

Basal cell carcinomas (BCCs) represent 70% of all skin cancers. These tumours do not metastasise but are locally invasive if left untreated. There is a high incidence of BCC in the elderly, and clinicians frequently face important treatment dilemmas. The approach to BCC in the elderly should be investigated thoroughly.

METHODS AND ANALYSIS

Data on health-related quality of life (HrQoL), survival and complication rate will be examined in a treatment and a non-treatment arm (1:1 allocation). In the non-treatment arm, in vivo biological behaviour of low-risk BCCs in elderly patients will be examined. The main objective is to combine tumour characteristics with demographic data, in order to determine whether treatment will positively affect the patients' HrQoL within a predetermined time frame. A monocentric randomised controlled trial (RCT) was designed at the Ghent University Hospital. The study population consists of patients with the minimum age of 75 years and a new diagnosis of (a) low-risk BCC(s). Patients in the treatment arm will receive standard care. Patients in the non-treatment arm will be closely monitored: the tumour will be intensively evaluated using multispectral dermoscopy, reflectance confocal microscopy and high-definition optical coherence tomography. All patients will be asked to fill in a questionnaire concerning their HrQoL at consecutive time points. Patient-reported side effects will be evaluated via an additional questionnaire.Primary outcomes will include the difference in HrQoL and the difference in complication risks (treatment vs non-treatment) at different time points of the study. Secondary endpoints are the evolution of the BCCs in the non-treatment arm and the long-term survival in both study arms. Tertiary endpoint is the treatment effectiveness in the treatment arm. The sample size calculation was performed and resulted in a target sample size of 272 patients in this study with a 1:1 allocation.

ETHICS AND DISSEMINATION

Subjects can withdraw from participating in this study at any time for any reason without any consequences. Approval for this study was received from the Ethics Committee of the Ghent University Hospital on 26 August 2021.The results of this RCT will be submitted for publication in one or more international, peer-reviewed medical journals, regardless of the nature of the study results.

TRIAL REGISTRATION NUMBER

ClinicalTrials.gov (NCT05110924).

Dermoscopy and skin imaging light sources: a comparison and review of spectral power distribution and color consistency

SIGNIFICANCE

Dermoscopes incorporate light, polarizers, and optical magnification into a handheld tool that is commonly used by dermatologists to evaluate skin findings. Diagnostic accuracy is improved when dermoscopes are used, and some major artificial intelligence (AI) projects have been accomplished using dermocopic images. Color rendering consistency and fidelity are crucial for clinical diagnostics, AI, and image processing applications.

AIM

With many devices available on the market, our objective was to measure the emission spectra of various dermoscopes, compare them with other light sources, and illustrate variations in reflected colors from images of a reference sample.

APPROACH

A spectrometer measured the spectral power distribution (SPD) produced by four dermoscope models and three alternate light sources, illustrating differences in the emission spectra. Most dermoscopes use light-emitting diodes (LEDs), which are inconsistent when compared with one another. An LED was compared with halogen, xenon-arc, and daylight sources. Images of a micro ColorChecker were acquired from several sources, and three specific colors were selected to compare in CIELAB color space. Color consistency and color fidelity measured by color rendering index (CRI) and TM-30-18 graphical vectors show variation in saturation and chroma fidelity.

RESULTS

A marked degree of variation was observed in both the emission and reflected light coming from different dermoscopes and compared with other sources. The same chromophores appeared differently depending on the light source used.

CONCLUSIONS

A lack of uniform illumination resulted in inconsistent image color and likely impacted metamerism and visibility of skin chromophores in real-world settings. Artificial light in skin examinations, especially LEDs, may present challenges for the visual separation of specific colors. Attention to LEDs SPD may be important, especially as the field increases dependency on machine/computer vision.

Data-Efficient Sensor Upgrade Path Using Knowledge Distillation

Deep neural networks have achieved state-of-the-art performance in image classification. Due to this success, deep learning is now also being applied to other data modalities such as multispectral images, lidar and radar data. However, successfully training a deep neural network requires a large reddataset. Therefore, transitioning to a new sensor modality (e.g., from regular camera images to multispectral camera images) might result in a drop in performance, due to the limited availability of data in the new modality. This might hinder the adoption rate and time to market for new sensor technologies. In this paper, we present an approach to leverage the knowledge of a teacher network, that was trained using the original data modality, to improve the performance of a student network on a new data modality: a technique known in literature as knowledge distillation. By applying knowledge distillation to the problem of sensor transition, we can greatly speed up this process. We validate this approach using a multimodal version of the MNIST dataset. Especially when little data is available in the new modality (i.e., 10 images), training with additional teacher supervision results in increased performance, with the student network scoring a test set accuracy of 0.77, compared to an accuracy of 0.37 for the baseline. We also explore two extensions to the default method of knowledge distillation, which we evaluate on a multimodal version of the CIFAR-10 dataset: an annealing scheme for the hyperparameter α and selective knowledge distillation. Of these two, the first yields the best results. Choosing the optimal annealing scheme results in an increase in test set accuracy of 6%. Finally, we apply our method to the real-world use case of skin lesion classification.