Position statement of the EADV Artificial Intelligence (AI) Task Force on AI-assisted smartphone apps and web-based services for skin disease.

BACKGROUND: As the use of smartphones continues to surge globally, mobile applications (apps) have become a powerful tool for healthcare engagement. Prominent among these are dermatology apps powered by Artificial Intelligence (AI), which provide immediate diagnostic guidance and educational resources for skin diseases, including skin cancer. OBJECTIVE: This article, authored by the EADV AI Task Force, seeks to offer insights and recommendations for the present and future deployment of AI-assisted smartphone applications (apps) and web-based services for skin diseases with emphasis on skin cancer detection. METHODS: An initial position statement was drafted on a comprehensive literature review, which was subsequently refined through two rounds of digital discussions and meticulous feedback by the EADV AI Task Force, ensuring its accuracy, clarity and relevance. RESULTS: Eight key considerations were identified, including risks associated with inaccuracy and improper user education, a decline in professional skills, the influence of non-medical commercial interests, data security, direct and indirect costs, regulatory approval and the necessity of multidisciplinary implementation. Following these considerations, three main recommendations were formulated: (1) to ensure user trust, app developers should prioritize transparency in data quality, accuracy, intended use, privacy and costs; (2) Apps and web-based services should ensure a uniform user experience for diverse groups of patients; (3) European authorities should adopt a rigorous and consistent regulatory framework for dermatology apps to ensure their safety and accuracy for users. CONCLUSIONS: The utilisation of AI-assisted smartphone apps and web-based services in diagnosing and treating skin diseases has the potential to greatly benefit patients in their dermatology journeys. By prioritising innovation, fostering collaboration and implementing effective regulations, we can ensure the successful integration of these apps into clinical practice.

A multidisciplinary evaluation, exploration, and advancement of the concept of a traumatic birth experience.

BACKGROUND: Understanding a woman's traumatic birth experience benefits from an approach that considers perspectives from various fields of healthcare and social sciences. AIM: To evaluate and explore the multidisciplinary perspectives surrounding a traumatic birth experience to form a theory and to capture its structure. METHODS: A multidisciplinary advanced principle-based concept analysis was conducted, including the following systematic steps: literature review, assessment of concept maturity, principle-based evaluation, concept exploration and advancement, and formulating a multidisciplinary concept theory. We drew on knowledge from midwifery, psychology, childbirth education, bioethics, obstetric & gender violence, sociology, perinatal psychiatry, and anthropology. RESULTS: Our evaluation included 60 records which were considered as 'mature'. Maturity was determined by the reported concept definition, attributes, antecedents, outcomes, and boundaries. The four broad principles of the philosophy of science epistemology, pragmatics, linguistics, and logic illustrated that women live in a political, and cultural world that includes social, perceptual, and practical features. The conceptual components antecedents, attributes, outcomes, and boundaries demonstrated that a traumatic birth experience is not an isolated event, but its existence is enabled by social structures that perpetuate the diminished and disempowered position of women in medical and institutionalised healthcare regulation and management. CONCLUSION: The traumatic childbirth experience is a distinctive experience that can only occur within a socioecological system of micro-, meso-, and macro-level aspects that accepts and allows its existence and therefore its sustainability - with the traumatic experience of the birthing woman as the central construct.

Attention Analysis in Robotic-Assistive Therapy for Children With Autism.

Children with Autism Spectrum Disorder (ASD) show severe attention deficits, hindering their capacity to acquire new skills. The automatic assessment of their attention response would provide the therapists with an important biomarker to better quantify their behaviour and monitor their progress during therapy. This work aims to develop a quantitative model, to evaluate the attention response of children with ASD, during robotic-assistive therapeutic sessions. Previous attempts to quantify the attention response of autistic subjects during human-robot interaction tasks were limited to restrained child movements. Instead, we developed an accurate quantitative system to assess the attention of ASD children in unconstrained scenarios. Our approach combines gaze extraction (Gaze360 model) with the definition of angular Areas-of-Interest, to characterise periods of attention towards elements of interest in the therapy environment during the session. The methodology was tested with 12 ASD children, achieving a mean test accuracy of 79.5 %. Finally, the proposed attention index was consistent with the therapists' evaluation of patients, allowing a meaningful interpretation of the automatic evaluation. This encourages the future clinical use of the proposed system.

A survey on deep learning for skin lesion segmentation.

Skin cancer is a major public health problem that could benefit from computer-aided diagnosis to reduce the burden of this common disease. Skin lesion segmentation from images is an important step toward achieving this goal. However, the presence of natural and artificial artifacts (e.g., hair and air bubbles), intrinsic factors (e.g., lesion shape and contrast), and variations in image acquisition conditions make skin lesion segmentation a challenging task. Recently, various researchers have explored the applicability of deep learning models to skin lesion segmentation. In this survey, we cross-examine 177 research papers that deal with deep learning-based segmentation of skin lesions. We analyze these works along several dimensions, including input data (datasets, preprocessing, and synthetic data generation), model design (architecture, modules, and losses), and evaluation aspects (data annotation requirements and segmentation performance). We discuss these dimensions both from the viewpoint of select seminal works, and from a systematic viewpoint, examining how those choices have influenced current trends, and how their limitations should be addressed. To facilitate comparisons, we summarize all examined works in a comprehensive table as well as an interactive table available online3.

A reinforcement learning model for AI-based decision support in skin cancer.

We investigated whether human preferences hold the potential to improve diagnostic artificial intelligence (AI)-based decision support using skin cancer diagnosis as a use case. We utilized nonuniform rewards and penalties based on expert-generated tables, balancing the benefits and harms of various diagnostic errors, which were applied using reinforcement learning. Compared with supervised learning, the reinforcement learning model improved the sensitivity for melanoma from 61.4% to 79.5% (95% confidence interval (CI): 73.5-85.6%) and for basal cell carcinoma from 79.4% to 87.1% (95% CI: 80.3-93.9%). AI overconfidence was also reduced while simultaneously maintaining accuracy. Reinforcement learning increased the rate of correct diagnoses made by dermatologists by 12.0% (95% CI: 8.8-15.1%) and improved the rate of optimal management decisions from 57.4% to 65.3% (95% CI: 61.7-68.9%). We further demonstrated that the reward-adjusted reinforcement learning model and a threshold-based model outperformed naïve supervised learning in various clinical scenarios. Our findings suggest the potential for incorporating human preferences into image-based diagnostic algorithms.

Parent organizations' experiences of the pandemic response in maternity care in thirteen European countries.

We surveyed changes to maternity care services in the first 17 months of the COVID-19 pandemic in 13 different European countries, from the perspective of national maternity service (parent) organizations advocating for a human rights approach to maternity services. A qualitative study was conducted in November 2020. An open-question survey was sent to national maternity service (parent) organizations and members of COST Action 18211 in Europe, asking about COVID-19 measures in maternity services (antenatally, intrapartum, postnatally, and overall satisfaction). From the open answers, 16 core issues were extracted. Between February and August 2021, semi-structured interviews with the national representatives of 14 parent member organizations in Europe were conducted, collecting details on overall national situations and changes due to COVID-19 measures. The reported experiences of parent organizations from 13 European countries show wide variations in epidemiological containment measures during the first 17 months of the COVID-19 pandemic. Practices differed between facilities, resulting in emotional disquiet and confusion for parent-patients. Most countries maintained antenatal and postnatal care but restricted psychosocial support (antenatal and birth companions, visitors). Organizations from nine countries reported that women had to wear masks during labor, and all but two countries saw separations of mothers and babies. Most parent organizations described a need for more reliable information for new parents. During the pandemic, non-evidence-based practices were (re-) established in many settings, depriving women and families of many factors which evidence has shown to be essential for a positive birthing experience. Based on the findings, we consider the challenges in maternity services and propose a strategy for future crises.

Regional differences in the quality of maternal and neonatal care during the COVID-19 pandemic in Portugal: Results from the IMAgiNE EURO study.

OBJECTIVE: To compare women's perspectives on the quality of maternal and newborn care (QMNC) around the time of childbirth across Nomenclature of Territorial Units for Statistics 2 (NUTS-II) regions in Portugal during the COVID-19 pandemic. METHODS: Women participating in the cross-sectional IMAgiNE EURO study who gave birth in Portugal from March 1, 2020, to October 28, 2021, completed a structured questionnaire with 40 key WHO standards-based quality measures. Four domains of QMNC were assessed: (1) provision of care; (2) experience of care; (3) availability of human and physical resources; and (4) reorganizational changes due to the COVID-19 pandemic. Frequencies for each quality measure within each QMNC domain were computed overall and by region. RESULTS: Out of 1845 participants, one-third (33.7%) had a cesarean. Examples of high-quality care included: low frequencies of lack of early breastfeeding and rooming-in (8.0% and 7.7%, respectively) and informal payment (0.7%); adequate staff professionalism (94.6%); adequate room comfort and equipment (95.2%). However, substandard practices with large heterogeneity across regions were also reported. Among women who experienced labor, the percentage of instrumental vaginal births ranged from 22.3% in the Algarve to 33.5% in Center; among these, fundal pressure ranged from 34.8% in Lisbon to 66.7% in Center. Episiotomy was performed in 39.3% of noninstrumental vaginal births with variations between 31.8% in the North to 59.8% in Center. One in four women reported inadequate breastfeeding support (26.1%, ranging from 19.4% in Algarve to 31.5% in Lisbon). One in five reported no exclusive breastfeeding at discharge (22.1%; 19.5% in Lisbon to 28.2% in Algarve). CONCLUSION: Urgent actions are needed to harmonize QMNC and reduce inequities across regions in Portugal.

Sharing Worlds: Design of a Real-Time Attention Classifier for Robotic Therapy of ASD Children.

Joint attention is the capacity of sharing attention between two agents and an aspect of the environment, through the use of different cues, namely gaze. This capacity is of paramount importance for social skills. People with Autism Spectrum Disorder (ASD) present certain deficits in joint attention. Therefore, there is an increasing interest in finding therapies to improve this skill. Some of these therapies include robots since they are known to be attractive to people with autism due to their motivation ability and predictability when compared with humans. In this line, we have designed a real-time attention classifier for a triadic robotic therapy, using Gaze360 and geometrical considerations of the scene. We were able to classify the gaze of the therapist and the one of the child during the whole session, even in a highly unconstrained scenario with a single camera, achieving a mean accuracy of 59%. This classifier can be used for the measurement of joint attention, an important metric for the development of adaptive robotic therapies, where increasing levels of difficulty and engagement are provided dependent on the ASD children, who are characterised by high heterogeneity. Future work will pass by the calculation of this metric and integration on a robotic platform for ASD therapy to understand the impact of these robotic therapies in improving ASD symptoms, specifically on how ASD children share their attention with other people present in the rehabilitation scenarios.

Checklist for Evaluation of Image-Based Artificial Intelligence Reports in Dermatology: CLEAR Derm Consensus Guidelines From the International Skin Imaging Collaboration Artificial Intelligence Working Group.

IMPORTANCE: The use of artificial intelligence (AI) is accelerating in all aspects of medicine and has the potential to transform clinical care and dermatology workflows. However, to develop image-based algorithms for dermatology applications, comprehensive criteria establishing development and performance evaluation standards are required to ensure product fairness, reliability, and safety. OBJECTIVE: To consolidate limited existing literature with expert opinion to guide developers and reviewers of dermatology AI. EVIDENCE REVIEW: In this consensus statement, the 19 members of the International Skin Imaging Collaboration AI working group volunteered to provide a consensus statement. A systematic PubMed search was performed of English-language articles published between December 1, 2008, and August 24, 2021, for "artificial intelligence" and "reporting guidelines," as well as other pertinent studies identified by the expert panel. Factors that were viewed as critical to AI development and performance evaluation were included and underwent 2 rounds of electronic discussion to achieve consensus. FINDINGS: A checklist of items was developed that outlines best practices of image-based AI development and assessment in dermatology. CONCLUSIONS AND RELEVANCE: Clinically effective AI needs to be fair, reliable, and safe; this checklist of best practices will help both developers and reviewers achieve this goal.

Theory-Based Approaches to Support Dermoscopic Image Interpretation Education: A Review of the Literature.

Introduction: Efficient interpretation of dermoscopic images relies on pattern recognition, and the development of expert-level proficiency typically requires extensive training and years of practice. While traditional methods of transferring knowledge have proven effective, technological advances may significantly improve upon these strategies and better equip dermoscopy learners with the pattern recognition skills required for real-world practice. Objectives: A narrative review of the literature was performed to explore emerging directions in medical image interpretation education that may enhance dermoscopy education. This article represents the first of a two-part review series on this topic. Methods: To promote innovation in dermoscopy education, the International Skin Imaging Collaborative (ISIC) assembled a 12-member Education Working Group that comprises international dermoscopy experts and educational scientists. Based on a preliminary literature review and their experiences as educators, the group developed and refined a list of innovative approaches through multiple rounds of discussion and feedback. For each approach, literature searches were performed for relevant articles. Results: Through a consensus-based approach, the group identified a number of emerging directions in image interpretation education. The following theory-based approaches will be discussed in this first part: whole-task learning, microlearning, perceptual learning, and adaptive learning. Conclusions: Compared to traditional methods, these theory-based approaches may enhance dermoscopy education by making learning more engaging and interactive and reducing the amount of time required to develop expert-level pattern recognition skills. Further exploration is needed to determine how these approaches can be seamlessly and successfully integrated to optimize dermoscopy education.

Instructional Strategies to Enhance Dermoscopic Image Interpretation Education: a Review of the Literature.

Introduction: In image interpretation education, many educators have shifted away from traditional methods that involve passive instruction and fragmented learning to interactive ones that promote active engagement and integrated knowledge. By training pattern recognition skills in an effective manner, these interactive approaches provide a promising direction for dermoscopy education. Objectives: A narrative review of the literature was performed to probe emerging directions in medical image interpretation education that may support dermoscopy education. This article represents the second of a two-part review series. Methods: To promote innovation in dermoscopy education, the International Skin Imaging Collaborative (ISIC) assembled an Education Working Group that comprises international dermoscopy experts and educational scientists. Based on a preliminary literature review and their experiences as educators, the group developed and refined a list of innovative approaches through multiple rounds of discussion and feedback. For each approach, literature searches were performed for relevant articles. Results: Through a consensus-based approach, the group identified a number of theory-based approaches, as discussed in the first part of this series. The group also acknowledged the role of motivation, metacognition, and early failures in optimizing the learning process. Other promising teaching tools included gamification, social media, and perceptual and adaptive learning modules (PALMs). Conclusions: Over the years, many dermoscopy educators may have intuitively adopted these instructional strategies in response to learner feedback, personal observations, and changes in the learning environment. For dermoscopy training, PALMs may be especially valuable in that they provide immediate feedback and adapt the training schedule to the individual's performance.

Quality of facility-based maternal and newborn care around the time of childbirth during the COVID-19 pandemic: online survey investigating maternal perspectives in 12 countries of the WHO European Region.

BACKGROUND: Multi-country studies assessing the quality of maternal and newborn care (QMNC) during the COVID-19 pandemic, as defined by WHO Standards, are lacking. METHODS: Women who gave birth in 12 countries of the WHO European Region from March 1, 2020 - March 15, 2021 answered an online questionnaire, including 40 WHO Standard-based Quality Measures. FINDINGS: 21,027 mothers were included in the analysis. Among those who experienced labour (N=18,063), 41·8% (26·1%- 63·5%) experienced difficulties in accessing antenatal care, 62% (12·6%-99·0%) were not allowed a companion of choice, 31·1% (16·5%-56·9%) received inadequate breastfeeding support, 34·4% (5·2%-64·8%) reported that health workers were not always using protective personal equipment, and 31·8% (17·8%-53·1%) rated the health workers' number as "insufficient". Episiotomy was performed in 20·1% (6·1%-66·0%) of spontaneous vaginal births and fundal pressure applied in 41·2% (11·5% -100%) of instrumental vaginal births. In addition, 23·9% women felt they were not treated with dignity (12·8%-59·8%), 12·5% (7·0%-23·4%) suffered abuse, and 2·4% (0·1%-26·2%) made informal payments. Most findings were significantly worse among women with prelabour caesarean birth (N=2,964). Multivariate analyses confirmed significant differences among countries, with Croatia, Romania, Serbia showing significant lower QMNC Indexes and Luxemburg showing a significantly higher QMNC Index than the total sample. Younger women and those with operative births also reported significantly lower QMNC Indexes. INTERPRETATION: Mothers reports revealed large inequities in QMNC across countries of the WHO European Region. Quality improvement initiatives to reduce these inequities and promote evidence-based, patient-centred respectful care for all mothers and newborns during the COVID-19 pandemic and beyond are urgently needed. FUNDING: The study was financially supported by the Institute for Maternal and Child Health IRCCS Burlo Garofolo, Trieste, Italy. STUDY REGISTRATION: ClinicalTrials.gov Identifier: NCT04847336.

Individual and country-level variables associated with the medicalization of birth: Multilevel analyses of IMAgiNE EURO data from 15 countries in the WHO European region.

OBJECTIVE: To investigate potential associations between individual and country-level factors and medicalization of birth in 15 European countries during the COVID-19 pandemic. METHODS: Online anonymous survey of women who gave birth in 2020-2021. Multivariable multilevel logistic regression models estimating associations between indicators of medicalization (cesarean, instrumental vaginal birth [IVB], episiotomy, fundal pressure) and proxy variables related to care culture and contextual factors at the individual and country level. RESULTS: Among 27 173 women, 24.4% (n = 6650) had a cesarean and 8.8% (n = 2380) an IVB. Among women with IVB, 41.9% (n = 998) reported receiving fundal pressure. Among women with spontaneous vaginal births, 22.3% (n = 4048) had an episiotomy. Less respectful care, as perceived by the women, was associated with higher levels of medicalization. For example, women who reported having a cesarean, IVB, or episiotomy reported not feeling treated with dignity more frequently than women who did not have those interventions (odds ratio [OR] 1.37; OR 1.61; OR 1.51, respectively; all: P < 0.001). Country-level variables contributed to explaining some of the variance between countries. CONCLUSION: We recommend a greater emphasis in health policies on promotion of respectful and patient-centered care approaches to birth to enhance women's experiences of care, and the development of a European-level indicator to monitor medicalization of reproductive care.

Quality of maternal and newborn care around the time of childbirth for migrant versus nonmigrant women during the COVID-19 pandemic: Results of the IMAgiNE EURO study in 11 countries of the WHO European region.

OBJECTIVE: To describe the perception of quality of maternal and newborn care (QMNC) around the time of childbirth among migrant and nonmigrant women in Europe. METHODS: Women who gave birth at a health facility in 11 countries of the WHO European Region from March 2020 to July 2021 were invited to answer an online questionnaire including demographics and childbirth experience. Data were analyzed and compared for 1781 migrant and 20 653 nonmigrant women. RESULTS: Migrant women who experienced labor perceived slightly more difficulties in attending routine antenatal visits (41.2% vs 39.4%; P = 0.001), more barriers in accessing facilities (32.9% vs 29.9%; P = 0.001), lack of timely care (14.7% vs 13.0%; P = 0.025), inadequate room comfort and equipment (9.2% vs 8.5%; P = 0.004), inadequate number of women per room (9.4% vs 8.6%; P = 0.039), being prevented from staying with their baby as they wished (7.8% vs 6.9%; P = 0.011), or suffering abuse (14.5% vs 12.7%; P = 0.022) compared with nonmigrant women. For women who had a prelabor cesarean, migrant women were more likely not to receive pain relief after birth (16.8% vs.13.5%; P = 0.039) and less likely to provide informal payment (1.8% vs 4.4%; P = 0.005) compared with nonmigrant women. Overall, the QMNC index was not significantly different for migrant compared with nonmigrant women. CONCLUSION: Gaps in overall QMNC were reported by both migrant and nonmigrant women, with improvements to healthcare necessary for all.

Skin cancer classification via convolutional neural networks: systematic review of studies involving human experts.

BACKGROUND: Multiple studies have compared the performance of artificial intelligence (AI)-based models for automated skin cancer classification to human experts, thus setting the cornerstone for a successful translation of AI-based tools into clinicopathological practice. OBJECTIVE: The objective of the study was to systematically analyse the current state of research on reader studies involving melanoma and to assess their potential clinical relevance by evaluating three main aspects: test set characteristics (holdout/out-of-distribution data set, composition), test setting (experimental/clinical, inclusion of metadata) and representativeness of participating clinicians. METHODS: PubMed, Medline and ScienceDirect were screened for peer-reviewed studies published between 2017 and 2021 and dealing with AI-based skin cancer classification involving melanoma. The search terms skin cancer classification, deep learning, convolutional neural network (CNN), melanoma (detection), digital biomarkers, histopathology and whole slide imaging were combined. Based on the search results, only studies that considered direct comparison of AI results with clinicians and had a diagnostic classification as their main objective were included. RESULTS: A total of 19 reader studies fulfilled the inclusion criteria. Of these, 11 CNN-based approaches addressed the classification of dermoscopic images; 6 concentrated on the classification of clinical images, whereas 2 dermatopathological studies utilised digitised histopathological whole slide images. CONCLUSIONS: All 19 included studies demonstrated superior or at least equivalent performance of CNN-based classifiers compared with clinicians. However, almost all studies were conducted in highly artificial settings based exclusively on single images of the suspicious lesions. Moreover, test sets mainly consisted of holdout images and did not represent the full range of patient populations and melanoma subtypes encountered in clinical practice.

A Survey of Feature Extraction in Dermoscopy Image Analysis of Skin Cancer.

Dermoscopy image analysis (DIA) is a growing field, with works being published every week. This makes it difficult not only to keep track of all the contributions, but also for new researchers to identify relevant information and new directions to be explored. Several surveys have been written in the past decade, but these tend to cover all of the steps of a CAD system, which can be overwhelming. Moreover, in these works, each of the steps is briefly discussed due to lack of space. Among the different blocks of the CAD system, the most relevant is the one devoted to feature extraction. This is also the block where existing works exhibit the most variability. Therefore, we believe that it is important to review the state-of-the-art on this matter. This work thoroughly explores the several types of features that have been used in DIA. A discussion on their relevance and limitations, as well as suggestions for future research are provided.

The Reversed Halo Sign: Not Such a Specific Sign of Organising Pneumonia.

The reversed halo sign is defined as a focal rounded area of ground-glass opacity surrounded by a more or less complete ring of consolidation. It is a relatively rare sign and initially considered a specific sign of organising pneumonia. We report the case of a 55-year-old female who was being followed-up in a pulmonology consultation due to a 6 mm nodule which required vigilance. On a re-evaluation chest CT scan, besides a stable 6 mm nodule, a 36 mm mass with the reversed halo sign was diagnosed. The presence of the reversed halo sign misled the multidisciplinary team into the diagnosis of organising pneumonia and initiation of corticotherapy was suggested. However, after further investigation, a final diagnosis of pulmonary tuberculosis was made. Even though this sign is relatively rare, and still considered an important clue to the diagnosis of organising pneumonia in immunocompetent patients, other causes must be excluded before starting treatment. LEARNING POINTS: The reversed halo sign (RHS) is defined as a focal rounded area of ground-glass opacity surrounded by a more or less complete ring of consolidation. It is a relatively rare sign, and still considered an important clue to the diagnosis of organising pneumonia (OP). However, the RHS has been described in other pulmonary diseases.The diagnosis of OP depends upon the demonstration of typical histopathologic features, usually through lung biopsy, and exclusion of other diseases which led, in our case, to a final diagnosis of pulmonary tuberculosis.

Electrochemical Immunosensor for TNFα-Mediated Inflammatory Disease Screening.

Inflammation associated with cancer, neurodegenerative, ocular, and autoimmune diseases has a considerable impact on public health. Tumor necrosis factor alpha (TNFα) is a key mediator of inflammatory responses, responsible for many of the systemic manifestations during the inflammatory process. Thus, inhibition of TNFα is a commonplace practice in the treatment of these disorders. Successful therapy requires the ability to determine the appropriate dose of anti-TNFα drugs to be administered in a timely manner, based on circulating TNFα levels. In this Letter, we report the development of an immunosensor technology able to quantify TNFα at the picogram level in relevant human body fluids, holding the potential to early detect inflammation  and monitor TNFα levels during treatment, enabling TNFα-targeted treatments to be tailored according to the immune status of an individual patient. This immunosensor technology is significantly more rapid and sensitive than conventional enzyme linked immunosorbent assays, maintaining high specificity and requiring small sample volumes. These features might also be advantageous in the context of personalized medicine, as this analytical platform can deliver advanced diagnostics and reduce clinical burden.

Melanoma detection algorithm based on feature fusion.

A Computer Aided-Diagnosis (CAD) System for melanoma diagnosis usually makes use of different types of features to characterize the lesions. The features are often combined into a single vector that can belong to a high dimensional space (early fusion). However, it is not clear if this is the optimal strategy and works on other fields have shown that early fusion has some limitations. In this work, we address this issue and investigate which is the best approach to combine different features comparing early and late fusion. Experiments carried on the datasets PH2 (single source) and EDRA (multi source) show that late fusion performs better, leading to classification scores of Sensitivity = 98% and Specificity = 90% (PH(2)) and Sensitivity = 83% and Specificity = 76% (EDRA).