Image-Based Artificial Intelligence in Psoriasis Assessment: The Beginning of a New Diagnostic Era?

Psoriasis, a chronic inflammatory skin disease, affects millions of people worldwide. It imposes a significant burden on patients' quality of life and healthcare systems, creating an urgent need for optimized diagnosis, treatment, and management. In recent years, image-based artificial intelligence (AI) applications have emerged as promising tools to assist physicians by offering improved accuracy and efficiency. In this review, we provide an overview of the current landscape of image-based AI applications in psoriasis. Emphasis is placed on machine learning (ML) algorithms, a key subset of AI, which enable automated pattern recognition for various tasks. Key AI applications in psoriasis include lesion detection and segmentation, differentiation from other skin conditions, subtype identification, automated area involvement, and severity scoring, as well as personalized treatment selection and response prediction. Furthermore, we discuss two commercially available systems that utilize standardized photo documentation, automated segmentation, and semi-automated Psoriasis Area and Severity Index (PASI) calculation for patient assessment and follow-up. Despite the promise of AI in this field, many challenges remain. These include the validation of current models, integration into clinical workflows, the current lack of diversity in training-set data, and the need for standardized imaging protocols. Addressing these issues is crucial for the successful implementation of AI technologies in clinical practice. Overall, we underscore the potential of AI to revolutionize psoriasis management, highlighting both the advancements and the hurdles that need to be overcome. As technology continues to evolve, AI is expected to significantly improve the accuracy, efficiency, and personalization of psoriasis treatment.

The SLICE-3D dataset: 400,000 skin lesion image crops extracted from 3D TBP for skin cancer detection.

AI image classification algorithms have shown promising results when applied to skin cancer detection. Most public skin cancer image datasets are comprised of dermoscopic photos and are limited by selection bias, lack of standardization, and lend themselves to development of algorithms that can only be used by skilled clinicians. The SLICE-3D ("Skin Lesion Image Crops Extracted from 3D TBP") dataset described here addresses those concerns and contains images of over 400,000 distinct skin lesions from seven dermatologic centers from around the world. De-identified images were systematically extracted from sensitive 3D Total Body Photographs and are comparable in optical resolution to smartphone images. Algorithms trained on lower quality images could improve clinical workflows and detect skin cancers earlier if deployed in primary care or non-clinical settings, where photos are captured by non-expert physicians or patients. Such a tool could prompt individuals to visit a specialized dermatologist. This dataset circumvents many inherent limitations of prior datasets and may be used to build upon previous applications of skin imaging for cancer detection.

Health Economic Consequences Associated With COVID-19-Related Delay in Melanoma Diagnosis in Europe.

Importance: The COVID-19 pandemic resulted in delayed access to medical care. Restrictions to health care specialists, staff shortages, and fear of SARS-CoV-2 infection led to interruptions in routine care, such as early melanoma detection; however, premature mortality and economic burden associated with this postponement have not been studied yet. Objective: To determine the premature mortality and economic costs associated with suspended melanoma screenings during COVID-19 pandemic lockdowns by estimating the total burden of delayed melanoma diagnoses for Europe. Design, Setting, and Participants: This multicenter economic evaluation used population-based data from patients aged at least 18 years with invasive primary cutaneous melanomas stages I to IV according to the American Joint Committee on Cancer (AJCC) seventh and eighth editions, including melanomas of unknown primary (T0). Data were collected from January 2017 to December 2021 in Switzerland and from January 2019 to December 2021 in Hungary. Data were used to develop an estimation of melanoma upstaging rates in AJCC stages, which was verified with peripandemic data. Years of life lost (YLL) were calculated and were, together with cost data, used for financial estimations. The total financial burden was assessed through direct and indirect treatment costs. Models were building using data from 50 072 patients aged 18 years and older with invasive primary cutaneous melanomas stages I to IV according to the AJCC seventh and eighth edition, including melanomas of unknown primary (T0) from 2 European tertiary centers. Data from European cancer registries included patient-based direct and indirect cost data, country-level economic indicators, melanoma incidence, and population rates per country. Data were analyzed from July 2021 to September 2022. Exposure: COVID-19 lockdown-related delay of melanoma detection and consecutive public health and economic burden. As lockdown restrictions varied by country, lockdown scenario was defined as elimination of routine medical examinations and severely restricted access to follow-up examinations for at least 4 weeks. Main Outcomes and Measures: Primary outcomes were the total burden of a delay in melanoma diagnosis during COVID-19 lockdown periods, measured using the direct (in US$) and indirect (calculated as YLL plus years lost due to disability [YLD] and disability-adjusted life-years [DALYs]) costs for Europe. Secondary outcomes included estimation of upstaging rate, estimated YLD, YLL, and DALY for each European country, absolute direct and indirect treatment costs per European country, proportion of the relative direct and indirect treatment costs for the countries, and European health expenditure. Results: There were an estimated 111 464 (range, 52 454-295 051) YLL due to pandemic-associated delay in melanoma diagnosis in Europe, and estimated total additional costs were $7.65 (range, $3.60 to $20.25) billion. Indirect treatment costs were the main cost driver, accounting for 94.5% of total costs. Estimates for YLD in Europe resulted in 15 360 years for the 17% upstaging model, ranging from 7228 years (8% upstaging model) to 40 660 years (45% upstaging model). Together, YLL and YLD constitute the overall disease burden, ranging from 59 682 DALYs (8% upstaging model) to 335 711 DALYs (45% upstaging model), with 126 824 DALYs for the real-world 17% scenario. Conclusions and Relevance: This economic analysis emphasizes the importance of continuing secondary skin cancer prevention measures during pandemics. Beyond the personal outcomes of a delayed melanoma diagnosis, the additional economic and public health consequences are underscored, emphasizing the need to include indirect economic costs in future decision-making processes. These estimates on DALYs and the associated financial losses complement previous studies highlighting the cost-effectiveness of screening for melanoma.