Dermatologie (Heidelberg, Germany)最新文献

Background: The diagnosis of actinic keratosis (AK), basal cell carcinoma (BCC), and psoriasis may present a challenge in everyday dermatological practice. Clinical and dermoscopic assessments often reach their limits, especially in ambiguous or anatomically difficult-to-access lesions. Biopsies are often impractical, and objective tools for treatment monitoring are lacking.

Objective: To investigate the potential of line-field confocal optical coherence tomography (LC-OCT) combined with artificial intelligence (AI) for noninvasive diagnosis, differentiation, and longitudinal monitoring.

Materials and methods: Analysis and evaluation of LC-OCT imaging data from various studies. Application of AI-based algorithms for the detection of vascular patterns, epidermal changes, and BCC identification using heatmap-supported decision tools.

Results and discussion: The LC-OCT enables high-resolution, real-time visualization of dermoepidermal structures as well as vascular architecture. In combination with AI, objective parameters such as PRO score, atypia, and vascular morphology can be quantified and monitored over time. AI-assisted diagnostics significantly improve diagnostic accuracy-especially in BCC and among less experienced users. However, implementation requires clear guidelines, standardization, and well-defined legal and ethical frameworks.

Conclusion: The LC-OCT combined with AI is a promising tool for more precise, standardized, and personalized dermatological diagnostics. Particularly in AK, BCC, and psoriasis, it has the potential to enhance care, reduce the need for invasive procedures, and provide novel insights into tumor and inflammation biology.

Background: Keratinocyte skin cancers (KC), including basal cell carcinoma (BCC) and the spectrum of squamous cell carcinoma (SCC), represent the most common malignancies worldwide. Accurate diagnosis and margin assessment are critical for optimal treatment but remain challenging when relying solely on clinical and dermoscopic evaluation. Reflectance confocal microscopy (RCM) offers high-resolution, noninvasive imaging of skin architecture at near-histological levels, potentially improving diagnostic precision and reducing the need for (re)biopsies.

Objective: To evaluate the current diagnostic criteria, diagnostic utility, advantages, and limitations of confocal microscopy in the assessment of nonmelanoma skin cancer (NMSC).

Methods: A narrative review of recent literature and current guidelines was conducted, focusing on RCM for the detection, characterization, and monitoring of BCC and KC. Key diagnostic criteria, imaging protocols, and clinical applications were analyzed, including preoperative mapping, differentiation from benign lesions, and follow-up after treatment.

Results: RCM provides real-time visualization of tumor architecture, cellular morphology, and vascular patterns, facilitating the identification of hallmark features such as basaloid tumor islands in BCC and keratinocyte atypia in SCC. Studies report improved diagnostic accuracy and reduced rates of unnecessary excisions. Limitations include restricted imaging depth and the requirement for operator expertise.

Conclusion: RCM represents a powerful adjunct to conventional diagnostic methods for NMSC. Its ability to deliver high-resolution, noninvasive imaging enhances diagnostic confidence, supports surgical planning, and aids in treatment monitoring, ultimately improving patient care. Combining multiple noninvasive techniques with proper training is key to enhancing this approach and promoting its routine clinical use.

Primary cutaneous B‑cell lymphomas (PCBCL) belong to the group of extranodal non-Hodgkin lymphomas and by definition only manifest on the skin at the time of diagnosis. These lymphomas can be broadly classified into indolent and aggressive subtypes. Indolent PCBCLs include primary cutaneous marginal zone lymphoma, primary cutaneous follicular B‑cell lymphoma and Epstein-Barr virus-positive mucocutaneous ulcer, which present solitary or multifocal lesions and are associated with a favorable prognosis. Treatment usually involves excision or radiotherapy, although a watch-and-wait concept can also be discussed. Conversely, aggressive PCBCLs, such as primary cutaneous diffuse large B‑cell lymphoma, leg type, and primary cutaneous intravascular large B‑cell lymphoma, carry a high risk of systemic dissemination. Consequently, they require systemic treatment. The standard first-line therapy is typically the R‑CHOP regimen, which may be supplemented with radiotherapy.

Reflectance confocal microscopy (RCM) is a noninvasive imaging technique offering high lateral resolution and is increasingly used in dermatology for differentiating benign from malignant melanocytic skin lesions such as nevi, melanoma, and lentigo maligna. By directing focused laser light into the skin and capturing reflected signals through a confocal aperture, horizontal optical sections of the epidermis and upper dermis can be generated with a resolution of 1-3 µm. Hallmarks of malignancy-such as pagetoid cells, nonedged papillae, and irregular nests-can be detected in vivo. The penetration depth of the technique is limited to approximately 250-300 µm. One of the key strengths of RCM lies in its ability to delineate tumor margins in lentigo maligna prior to surgical or laser-based therapy. In addition to reflectance imaging, fluorescence-based visualization is possible following topical or intradermal administration of contrast agents. Recent advances integrate artificial intelligence (AI) to assist in the automated analysis of RCM images. Early AI-based systems, such as MED-Net, have demonstrated promising sensitivity and specificity for identifying melanocytic structures. Overall, RCM-especially when combined with validated AI algorithms-represents a valuable extension of diagnostic capabilities in modern dermatologic oncology.

Line-field confocal optical coherence tomography (LC-OCT) is an innovative and noninvasive imaging technique that combines the advantages of reflectance confocal microscopy (RCM) and optical coherence tomography (OCT). Previously used for the diagnosis of basal cell carcinoma and squamous cell carcinomas, recent studies also show promising results in its application in melanocytic tumors. While in LC-OCT benign nevi exhibit a regular epidermal architecture with a honeycomb and cobblestone pattern as well as clearly defined dermoepidermal junctions (DEJ), melanomas, in contrast, exhibit an irregular cellular architecture with pagetoid ascending cells, destruction of the DEJ, and clefting. Therefore, LC-OCT expands the diagnostic spectrum for the differentiated assessment of conspicuous melanocytic lesions.

Ex vivo confocal laser scanning microscopy is a modern diagnostic tool that enables rapid microscopic evaluation of freshly excised skin samples without tissue loss. By combining reflectance and fluorescence imaging with digital staining algorithms, the technique produces section images that closely resemble conventional hematoxylin-eosin (HE) staining. In dermatological practice, it is primarily used for the diagnosis and margin assessment of basal cell carcinomas. The method also shows promise for squamous cell carcinomas, its precursors, and inflammatory dermatoses. Current limitations include the assessment of pigmented lesions and the differentiation of inflammatory cell infiltrates. Future developments such as artificial intelligence (AI)-assisted image analysis or fluorescence-labeled antibodies could further expand its range of applications.

Background: The incidence of malignant melanoma has increased exponentially in recent decades. The combination of total body photography (TBP) and digital dermoscopy (DD) improved the early detection of melanoma in previous studies. Furthermore, the use of artificial intelligence (AI) can improve the diagnostic accuracy of dermatologists.

Objectives: In the present work, the applications of AI in TBP, DD, and high-resolution dermoscopy in clinical practice are evaluated.

Materials and methods: A literature review was conducted of previous studies on the use of TBP, DD, and high-resolution dermoscopy and the integration of AI in these devices for the early detection of melanoma.

Results and conclusion: The use of AI can improve the diagnostic accuracy of TBP, DD, and high-resolution dermoscopy. As most of the studies to date have been conducted in an artificial setting, more studies are needed to evaluate the potential use of AI in the clinical setting. In addition, staff must be trained and informed about the legal framework regarding the use of AI in everyday clinical practice.