Dermoscopy

Advancements in dermatological artificial intelligence research require high-quality and comprehensive datasets that mirror real-world clinical scenarios. We introduce a collection of 18,946 dermoscopic images spanning from 2010 to 2016, collated at the Hospital Clínic in Barcelona, Spain. The BCN20000 dataset aims to address the problem of unconstrained classification of dermoscopic images of skin cancer, including lesions in hard-to-diagnose locations such as those found in nails and mucosa, large lesions which do not fit in the aperture of the dermoscopy device, and hypo-pigmented lesions. Our dataset covers eight key diagnostic categories in dermoscopy, providing a diverse range of lesions for artificial intelligence model training. Furthermore, a ninth out-of-distribution (OOD) class is also present on the test set, comprised of lesions which could not be distinctively classified as any of the others. By providing a comprehensive collection of varied images, BCN20000 helps bridge the gap between the training data for machine learning models and the day-to-day practice of medical practitioners. Additionally, we present a set of baseline classifiers based on state-of-the-art neural networks, which can be extended by other researchers for further experimentation.

Actinic keratoses are dysplastic proliferations of keratinocytes with potential for malignant transformation. Clinically, actinic keratoses present as macules, papules, or hyperkeratotic plaques with an erythematous background that occur on photoexposed areas. At initial stages, they may be better identified by palpation rather than by visual inspection. They may also be pigmented and show variable degrees of infiltration; when multiple they then constitute the so-called field cancerization. Their prevalence ranges from 11% to 60% in Caucasian individuals above 40 years. Ultraviolet radiation is the main factor involved in pathogenesis, but individual factors also play a role in the predisposing to lesions appearance. Diagnosis of lesions is based on clinical and dermoscopic examination, but in some situations histopathological analysis may be necessary. The risk of transformation into squamous cell carcinoma is the major concern regarding actinic keratoses. Therapeutic modalities for actinic keratoses include topical medications, and ablative and surgical methods; the best treatment option should always be individualized according to the patient.

ObjectiveMost skin lesions first present in primary care, where distinguishing rare melanomas from benign lesions can be challenging. Dermoscopy improves diagnostic accuracy among specialists and is promoted for use by primary care physicians (PCPs). However, when used by untrained clinicians, accuracy may be no better than visual inspection. This study aimed to undertake a systematic review of literature reporting use of dermoscopy to triage suspicious skin lesions in primary care settings, and challenges for implementation.DesignA systematic literature review and narrative synthesis.Data sourcesWe searched MEDLINE, Cochrane Central, EMBASE, Cumulative Index to Nursing and Allied Health Literature, and SCOPUS bibliographic databases from 1 January 1990 to 31 December 2017, without language restrictions.Inclusion criteriaStudies including assessment of dermoscopy accuracy, acceptability to patients and PCPs, training requirements, and cost-effectiveness of dermoscopy modes in primary care, including trials, diagnostic accuracy and acceptability studies.Results23 studies met the review criteria, representing 49 769 lesions and 3708 PCPs, all from high-income countries. There was a paucity of studies set truly in primary care and the outcomes measured were diverse. The heterogeneity therefore made meta-analysis unfeasible; the data were synthesised through narrative review. Dermoscopy, with appropriate training, was associated with improved diagnostic accuracy for melanoma and benign lesions, and reduced unnecessary excisions and referrals. Teledermoscopy-based referral systems improved triage accuracy. Only three studies examined cost-effectiveness; hence, there was insufficient evidence to draw conclusions. Costs, training and time requirements were considered important implementation barriers. Patient satisfaction was seldom assessed. Computer-aided dermoscopy and other technological advances have not yet been tested in primary care.ConclusionsDermoscopy could help PCPs triage suspicious lesions for biopsy, urgent referral or reassurance. However, it will be important to establish further evidence on minimum training requirements to reach competence, as well as the cost-effectiveness and patient acceptability of implementing dermoscopy in primary care.Trial registration numberCRD42018091395.

Skin cancer is one of the most prevalent and potentially lethal diseases worldwide, with early detection being critical for patient survival. This study presents a novel framework that leverages transfer learning, pruning, SMOTE, data augmentation, and the advanced Avg-TopK pooling method to improve the accuracy and efficiency of skin cancer classification using dermoscopic images. The HAM10000 dataset was used to evaluate the performance of various transfer learning models, with Xception as the top performer. A layer-based pruning strategy was proposed to optimize the model and reduce its complexity. SMOTE and data augmentation were applied to address the class imbalance within the dataset, significantly improving the model's generalization across all skin lesion classes. The utilization of the Avg-TopK pooling technique further enhanced model accuracy by preserving crucial image features during the downsampling process. The proposed approach achieved an overall accuracy of 91.52%, surpassing several state-of-the-art models. Following pruning, the model's parameter count was reduced by approximately 35%, from 20.9 million to 13.5 million, improving efficiency and performance. This framework demonstrates the effectiveness of combining model pruning, oversampling, and advanced pooling methods to build robust and efficient skin cancer classification systems suitable for clinical applications.