Differences in the annotation between facial images and videos for training an artificial intelligence for skin type determination.

BACKGROUND: The Grand-AID research project, consisting of GRANDEL-The Beautyness Company, the dermatology department of Augsburg University Hospital and the Chair of IT Infrastructure for Translational Medical Research at Augsburg University, is currently researching the development of a digital skin consultation tool that uses artificial intelligence (AI) to analyze the user's skin and ultimately perform a personalized skin analysis and a customized skin care routine. Training the AI requires annotation of various skin features on facial images. The central question is whether videos are better suited than static images for assessing dynamic parameters such as wrinkles and elasticity. For this purpose, a pilot study was carried out in which the annotations on images and videos were compared. MATERIALS AND METHODS: Standardized image sequences as well as a video with facial expressions were taken from 25 healthy volunteers. Four raters with dermatological expertise annotated eight features (wrinkles, redness, shine, pores, pigmentation spots, dark circles, skin sagging, and blemished skin) with a semi-quantitative and a linear scale in a cross-over design to evaluate differences between the image modalities and between the raters. RESULTS: In the videos, most parameters tended to be assessed with higher scores than in the images, and in some cases significantly. Furthermore, there were significant differences between the raters. CONCLUSION: The present study shows significant differences between the two evaluation methods using image or video analysis. In addition, the evaluation of the skin analysis depends on subjective criteria. Therefore, when training the AI, we recommend regular training of the annotating individuals and cross-validation of the annotation.

Dynamic optical coherence tomography of chronic venous ulcers.

BACKGROUND: Chronic ulcers, especially venous leg ulcers, are a major burden on the healthcare system. To date there are only few non-invasive established procedures for evaluation of blood perfusion in wounds. Dynamic optical coherence tomography (D-OCT) provides images of the skin's superficial vascularisation. OBJECTIVES: This study aims to investigate if and how the D-OCT measurement of chronic wounds can provide new information about the vascularisation during the healing process. METHODS: We examined 16 venous ulcers over 16 weeks and evaluated the vessel morphology and density using D-OCT at the wound bed, borders, two centimetres adjacent to the wound und at non-ulcerated skin on the contralateral leg. RESULTS: In D-OCT scans clumps were unique and the most common vessel type in the wound area of venous ulcers, whereas lines and serpiginous vessels were the most common in non-ulcerated skin. At the wound border mottle and cluster patterns occurred more frequently. Healthy skin showed a significant increase of mesh pattern. Vessel density significantly increased at the wound area compared to non-ulcerated skin. During the healing process the wound border showed the most vascular changes while only an increase in curves was observed in the wound centre. Non-healing wounds had fewer dots and blobs at the borders, fewer dots, coils, clumps, lines and serpiginous vessels at the centre and fewer dots in adjacent skin. Temperature analysis showed higher temperatures in non-ulcerated skin, followed by the wound margin and centre. Non-healing wounds showed the lowest temperatures in the wound centre. CONCLUSIONS: These results highlight the non-invasive use of D-OCT for the examination and monitoring of wound healing in chronic venous ulcers. D-OCT imaging of blood vessels may offer the potential to detect disorders of wound healing at an early stage, differentiate ulcers of different genesis and to tailor more individualized, patient-oriented therapy.

Diagnose und Therapie aktinischer Keratosen: Diagnosis and therapy of actinic keratosis.

Actinic keratosis (AK) is considered a chronic and recurring in situ skin neoplasia, with a possible transformation into invasive squamous cell carcinoma (SCC). Among others, predominant risk factors for development of AK are UV-light exposure and immunosuppression. Basal epidermal keratinocyte atypia (AK I) and proliferation (PRO Score) seem to drive malignant turnover, rather than clinical appearance of AK (Olsen I-III). Due to the invasiveness of punch biopsy, those histological criteria are not regularly assessed. Non-invasive imaging techniques, such as optical coherence tomography (OCT), reflectance confocal microscopy (RCM) and line-field confocal OCT (LC-OCT) are helpful to distinguish complex cases of AK, Bowen's disease and SCC. Moreover, LC-OCT can visualize the epidermis and the papillary dermis at cellular resolution, allowing real-time PRO Score assessment. The decision-making for implementation of therapy is still based on clinical risk factors, ranging from lesion- to field-targeted and ablative to non-ablative regimes, but in approximately 85% of the cases a recurrence of AK can be observed after a 1-year follow-up. The possible beneficial use of imaging techniques for a non-invasive follow-up of AK to detect recurrence or invasive progression early on should be subject to critical evaluation in further studies.

Diagnosis and therapy of actinic keratosis.

Actinic keratosis (AK) is considered a chronic and recurring in situ skin neoplasia, with a possible transformation into invasive squamous cell carcinoma (SCC). Among others, predominant risk factors for development of AK are UV-light exposure and immunosuppression. Basal epidermal keratinocyte atypia (AK I) and proliferation (PRO score) seem to drive malignant transformation, rather than clinical appearance of AK (Olsen I-III). Due to the invasiveness of punch biopsy, those histological criteria are not regularly assessed. Non-invasive imaging techniques, such as optical coherence tomography (OCT), reflectance confocal microscopy (RCM) and line-field confocal OCT (LC-OCT) are helpful to distinguish complex cases of AK, Bowen's disease, and SCC. Moreover, LC-OCT can visualize the epidermis and the papillary dermis at cellular resolution, allowing real-time PRO score assessment. The decision-making for implementation of therapy is still based on clinical risk factors, ranging from lesion- to field-targeted and ablative to non-ablative regimens, but in approximately 85% of the cases a recurrence of AK can be observed after a 1-year follow-up. The possible beneficial use of imaging techniques for a non-invasive follow-up of AK to detect recurrence or invasive progression early on should be subject to critical evaluation in further studies.

Line-field confocal optical coherence tomography, a novel non-invasive tool for the diagnosis of onychomycosis.

BACKGROUND AND OBJECTIVES: Onychomycosis is common and important to distinguish from other nail diseases. Rapid and accurate diagnosis is necessary for optimal patient treatment and outcome. Non-invasive diagnostic tools have increasing potential for nail diseases including onychomycosis. This study evaluated line-field confocal optical coherence tomography (LC-OCT) as a rapid non-invasive tool for diagnosing onychomycosis as compared to confocal laser scanning microscopy (CLSM), optical coherence tomography (OCT), and conventional methods. PATIENTS AND METHODS: In this prospective study 86 patients with clinically suspected onychomycosis and 14 controls were examined using LC-OCT, OCT, and CLSM. KOH-preparation, fungal culture, PCR, and histopathology were used as comparative conventional methods. RESULTS: LC-OCT had the highest sensitivity and negative predictive value of all methods used, closely followed by PCR and OCT. Specificity and positive predictive value of LC-OCT were as high as with CLSM, while OCT scored much lower. The gold standard technique, fungal culture, showed the lowest sensitivity and negative predictive value. Only PCR and culture allowed species differentiation. CONCLUSIONS: LC-OCT enables quick and non-invasive detection of onychomycosis, with advantages over CLSM and OCT, and similar diagnostic accuracy to PCR but lacking species differentiation. For accurate nail examination, LC-OCT requires well-trained and experienced operators.

How efficient is laser therapy for telangiectasias, spider veins, and cherry angiomas?-A study using dynamic optical coherence tomography.

OBJECTIVES: A clinical study to investigate the effectiveness of pulsed dye laser (PDL) versus Nd:YAG laser in the treatment of telangiectasias, spider veins and cherry angiomas. Dynamic optical coherence tomography (D-OCT) was introduced as an innovative follow-up tool for evaluation of blood flow within superficial vessels and to allow visualization of morphological changes of the vasculature in vivo. The final aim of this study was to demonstrate a possible treatment benefit comparing both laser types. MATERIALS AND METHODS: Vessel structures of 102 skin lesions were documented photographically and dermoscopically. Subsequently, lesions were imaged using optical coherence tomography before laser therapy (a), directly after the treatment (p) and after a follow-up 4-6 weeks after laser treatment. All lesions were treated using either a 595 nm PDL or a 1064 nm Nd:YAG laser. Two main vessel parameters, namely density and diameter, and their possible changes during follow-up were observed in 150/300/500 µm penetration depth using D-OCT and were subsequently compared between both treatment groups. Other analyzed vessel parameters were depth of the plexus, mean diameter, mean density, top edge of the vessel, columns, and spikes. RESULTS: Both laser types are suitable options for the treatment of vascular skin lesions, with the most significant effect on cherry angiomas. PDL shows better results treating smaller vessels in upper skin regions, in comparison to Nd:YAG laser, achieving better results on deeper vessels, like spider veins. Using the applied laser settings, there was no statistically significant effect on telangiectasias. CONCLUSION: D-OCT represents a new, noninvasive imaging method to evaluate blood flow and vessel morphology in the follow-up of telangiectasias, spider veins, and cherry angiomas, which underwent laser therapy.

Optical coherence tomography-guided Nd:YAG laser treatment and follow-up of basal cell carcinoma.

OBJECTIVES: Basal cell carcinoma (BCC) is the most common skin tumor with an annually increasing incidence. Standard care requires several visits for diagnosis and treatment. Optical coherence tomography (OCT) as a diagnostic tool increases the sensitivity (95%) and specificity (77%) of the diagnosis of BCC. Although laser therapy is not the standard of care, the long-pulsed 1064 nm Nd:YAG laser seems to be a promising option. However, data are scarce. The published papers had a short follow-up (FU) time and used to some extent inferior methods to detect complete tumor clearance. To address this research gap, this study evaluates the efficiency of laser treatment by FU OCT. We pursue a patient-focused approach and combine OCT with Nd:YAG laser treatment in one procedure. MATERIALS AND METHODS: The study was conducted as a prospective, single-center trial that recruited biopsy-confirmed or OCT-proven BCC with a tumor thickness of less than 1.2 mm. Patients underwent two or three repeated sessions with the Nd:YAG laser (5-6 mm spot, fluence of 120-140 J/cm2 , pulse duration of 8-10 milliseconds). Each BCC was assessed at baseline, and 3 and 12 months after laser treatment by clinical image, dermoscopy, and OCT. Incomplete tumor clearance (ITC) was defined as a clearly detectable BCC on the OCT image or a biopsy-confirmed BCC in the treated area. RESULTS: Forty-five patients completed the 12-month FU (46.7% women; median age of 74.0 [52-88] years) with a total number of 78 BCC lesions. At baseline, all patients had their BCC diagnosed by OCT (tumor thickness of 0.6 [0.4; 0.8] mm), 15.4% lesions were additionally diagnosed by histopathology. The most common subtype of BCC was superficial (48.7%), followed by nodular (47.4%) and infiltrative (3.8%). ITC rate after the treatment using Nd:YAG laser was 30.8% (95% CI: 20.8%-42.2%) (24/78) after 3 months and 7.4% (95% CI: 2.1%-17.9%) (4/54) after 12 months. ITC was not associated with histological subtype, tumor thickness, or location. If ITC was detected, the lesion was treated again. Out of 19 lesions with at least one additional laser treatment, 7 lesions (36.8%) suffered from incomplete tumor removal. In 46.7% of the treated lesions, the cosmetic outcome was rated as moderate or severe scarring after 12 months. CONCLUSION: Our results demonstrate that the ITC rate of BCC treated with the Nd:YAG laser is much higher (up to one-third) than reported, although the laser settings were identical to prior studies. This is especially evident at the 3-month FU. In addition, we witnessed a larger number of side effects and a worse cosmetic outcome compared to previous studies.

Implementation of a dermatoscopy curriculum during residency at Augsburg University Hospital in Germany.

BACKGROUND AND OBJECTIVES: To date, there is no structured program for dermatoscopy training during residency in Germany. Whether and how much dermatoscopy training is acquired is left to the initiative of each resident, although dermatoscopy is one of the core competencies of dermatological training and daily practice. The aim of the study was to establish a structured dermatoscopy curriculum during residency at the University Hospital Augsburg. PATIENTS AND METHODS: An online platform with dermatoscopy modules was created, accessible regardless of time and place. Practical skills were acquired under the personal guidance of a dermatoscopy expert. Participants were tested on their level of knowledge before and after completing the modules. Test scores on management decisions and correct dermatoscopic diagnosis were analyzed. RESULTS: Results of 28 participants showed improvements in management decisions from pre- to posttest (74.0% vs. 89.4%) and in dermatoscopic accuracy (65.0% vs. 85.6%). Pre- vs. posttest differences in test score (7.05/10 vs. 8.94/10 points) and correct diagnosis were significant (p < 0.001). CONCLUSIONS: The dermatoscopy curriculum increases the number of correct management decisions and dermatoscopy diagnoses. This will result in more skin cancers being detected, and fewer benign lesions being excised. The curriculum can be offered to other dermatology training centers and medical professionals.

Artificial intelligence-based PRO score assessment in actinic keratoses from LC-OCT imaging using Convolutional Neural Networks.

BACKGROUND AND OBJECTIVES: The histological PRO score (I-III) helps to assess the malignant potential of actinic keratoses (AK) by grading the dermal-epidermal junction (DEJ) undulation. Line-field confocal optical coherence tomography (LC-OCT) provides non-invasive real-time PRO score quantification. From LC-OCT imaging data, training of an artificial intelligence (AI), using Convolutional Neural Networks (CNNs) for automated PRO score quantification of AK in vivo may be achieved. PATIENTS AND METHODS: CNNs were trained to segment LC-OCT images of healthy skin and AK. PRO score models were developed in accordance with the histopathological gold standard and trained on a subset of 237 LC-OCT AK images and tested on 76 images, comparing AI-computed PRO score to the imaging experts' visual consensus. RESULTS: Significant agreement was found in 57/76 (75%) cases. AI-automated grading correlated best with the visual score for PRO II (84.8%) vs. PRO III (69.2%) vs. PRO I (66.6%). Misinterpretation occurred in 25% of the cases mostly due to shadowing of the DEJ and disruptive features such as hair follicles. CONCLUSIONS: The findings suggest that CNNs are helpful for automated PRO score quantification in LC-OCT images. This may provide the clinician with a feasible tool for PRO score assessment in the follow-up of AK.

Management of suspected and confirmed COVID-19 (SARS-CoV-2) vaccine hypersensitivity.

BACKGROUND: Systemic allergic reactions to vaccines are very rare. In this study we assessed the management and outcome of suspected SARS-CoV-2 vaccine hypersensitivity. METHODS: Totally, 334 individuals underwent an allergy work up regarding SARS-CoV-2 vaccination (group A: 115 individuals suspected to be at increased risk for vaccine-related reactions before vaccination and group B: 219 patients with reactions after COVID vaccination). The large majority of the SPT/IDT with the vaccines were negative; however, we identified in 14.1% (n = 47) a possible sensitization to the SARS-CoV-2 vaccine and/or its ingredients defined as one positive skin test. Of the 219 individuals (group B) who experienced symptoms suspicious for a hypersensitivity reaction after vaccination, 214 were reported after the first vaccination with a mRNA vaccine (157 mRNA (Comirnaty®, 38 Spikevax®) and 18 with a vector vaccine (Vaxzevria®), 5 cases were after the second vaccination. RESULTS: The symptom profile in group B was as follows: skin symptoms occurred in 115 cases (n = 59 angioedema, n = 50 generalized urticaria and n = 23 erythema/flush. Seventy individuals had cardiovascular, 53 respiratory and 17 gastrointestinal symptoms. Of the overall 334 individuals, 78 patients tolerated (re)-vaccination (out of skin test positive/negative 7/19 from group A and 17/35 from group B). CONCLUSION: Proven IgE-mediated hypersensitivity to SARS-CoV-2 vaccines is extremely rare and not increased in comparison with reported hypersensitivity to other vaccines. The value of skin tests is unclear and nonspecific reactions, in particular when intradermal testing is applied, should be considered.

[Optical coherence tomography for the differential diagnosis of unclear nail pigmentation]. / Optische Kohärenztomographie zur differenzialdiagnostischen Abklärung unklarer Nagelpigmentierungen.

In daily practice, nail pigmentation can be a diagnostic challenge, especially if the dermoscopic findings are nonspecific. We present examples of cases, in which optical coherence tomography-a rapid, noninvasive imaging method-showed typical changes that were indicative for the diagnosis.

Line-field confocal optical coherence tomography-Practical applications in dermatology and comparison with established imaging methods.

BACKGROUND: Non-invasive diagnostic techniques in dermatology gained increasing popularity in the last decade. Reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) are meanwhile established in research and clinical routine. While OCT is mainly indicated for detecting non-melanoma skin cancer, RCM has proven its usefulness additionally in distinguishing melanocytic lesions. Line-field confocal optical coherence tomography (LC-OCT) is an emerging tool combining the principles of both above-mentioned methods. METHODS: Healthy skin at different body sites and exemplary skin lesions (basal cell carcinoma, malignant melanoma, actinic keratosis) were examined using dermoscopy, RCM, OCT and LC-OCT. Standard features for RCM and OCT and comparable features for LC-OCT were analysed. RESULTS: LC-OCT has a lower penetration depth but superior resolution compared to OCT. In comparison with RCM, which provides only horizontal sections, LC-OCT creates both vertical and horizontal images in real time and has nearly the same cellular resolution. DISCUSSION: Our preliminary experiences suggest that LC-OCT combines the advantages of RCM and OCT, with optimal resolution and penetration depth to diagnose all types of skin cancer. Larger systematic studies are needed to further characterize the field of use of this device and its sensitivity and specificity compared to histology.

[Confocal line-field OCT]. / Konfokale Line-Field-OCT.

Optical coherence tomography (OCT) and confocal laser microscopy (CLSM) are established non-invasive methods in clinical dermatological routine diagnosis. Whereas CLSM is especially useful to distinguish between nevi and melanoma, OCT is suitable for the diagnosis and differentiation of non-melanoma skin cancer. Line-field confocal optical coherence tomography (LC-OCT) is a new innovative device, which has better cellular resolution than OCT and a higher penetration depth than CLSM. Similar to CLSM, LC-OCT also allows 3D images in real time to be taken. Therefore LC-OCT is very useful for the examination of skin lesions of all kinds, since it unites the features of CLSM and OCT.

[Advances in optical coherence tomography]. / Neues von der optischen Kohärenztomographie.

Optical coherence tomography (OCT) has been able to establish itself in recent years not only in academic-scientific, but also in everyday dermatological practice. Its focus lies on epithelial tumors of the skin, which can be diagnosed intuitively and within a few seconds. Thus, basal cell carcinomas, actinic keratoses, and different stages of field cancerization can be diagnosed and monitored for response to therapy or possible recurrence. This often helps to avoid invasive sample extraction. Recently, the field of OCT and its latest advancement, dynamic OCT (D-OCT), has been expanded to include non-oncologic dermatological diseases. This encompasses inflammatory dermatoses and the analysis of physiological skin parameters such as hydration. Thanks to automated vascular imaging and the measurement of objective parameters such as epidermal thickness, blood flow at depth, optical attenuation coefficient, and skin roughness, more and more characteristics of the skin can be studied in a noninvasive and standardized way. New potential areas of application are eczema, contact allergic dermatitis, psoriasis, rosacea, telangiectasia, acute and chronic wounds, melasma and nevus flammeus but also melanocytic lesions.

Cumulative Sum Analysis for the Learning Curve of Optical Coherence Tomography Assisted Diagnosis of Basal Cell Carcinoma.

The amount of training needed to correctly interpret optical coherence tomography scans of the skin is undefined. The aim of this study was to illustrate how cumulative sum charts can be used to determine how many optical coherence tomography scans novice assessors should evaluate in order to obtain competence in diagnosing basal cell carcinoma. Four hundred lesions suspected for non-melanoma skin cancer were evaluated by optical coherence tomography in combination with clinical photographs, using a 5-point confidence scale. The diagnostic error rate (sum of false-negative and false-positive optical coherence tomography results/total number of cases) was used to evaluate performance, with histopathological diagnosis as the reference standard. Acceptable and unacceptable error rates were set at 16% and 25%, respectively. Adequate performance was reached after assessing 183-311 scans, dependent on the cut-off for a positive test result. In conclusion, cumulative sum analysis is useful to monitor the progress of optical coherence tomography trainees. The caseload necessary for training is substantial.

The vascular morphology of melanoma is related to Breslow index: An in vivo study with dynamic optical coherence tomography.

BACKGROUND: Malignant melanoma is an aggressive skin cancer, which can lead to metastasis development. Vascularization enhancement is fundamental for tumor growth, worsening the prognosis. Dynamic optical coherence tomography (D-OCT) enables the in vivo evaluation of vascular patterns in skin lesions. OBJECTIVE: In vivo evaluation of the melanoma vessel morphology by means of D-OCT and correlation with Breslow index. METHODS: Retrospective analysis of histologically proven melanomas, evaluated by D-OCT at three different depths (150, 300 and 500 µm), was performed. Vessels were classified according to morphology (dots, blobs, coiled, line, curved, serpiginous), distribution (regular, irregular) and the presence/type of branches. The data were correlated with Breslow thickness. RESULTS: A total of 127 melanomas were evaluated. Dotted vessels were recorded at all depths, and their irregular distribution was associated with lesions thicker than 1.0 mm (from 75% to 91%), compared with thin ones (42%) at 150 µm (P = 0.031), and from 33% to 57% vs 18% at 300 µm (P = 0.021). Serpiginous and branching vessels with bulges were predominantly seen in melanomas thicker than 2 mm at 150 µm (from 14% to 27%, P < 0.001) and 300 µm of depth (from 36% to 54%, P < 0.001). LIMITATIONS: Background noise hampered vessel detection at 500 µm. No correlation with dermoscopy/histology. CONCLUSION: Vascular pattern evaluation at 150 and 300 µm provided data on tumor microvascular asset and its pattern of progression in accordance with Breslow thickness. Since vascular progression is theoretically linked with tumor aggressiveness, the study of vascular pattern related with melanoma metastatization capability is warranted.

In vivo differentiation of common basal cell carcinoma subtypes by microvascular and structural imaging using dynamic optical coherence tomography.

The subtype of basal cell carcinoma (BCC) influences the choice of treatment. Optical coherence tomography (OCT) is a non-invasive imaging tool, and a recent development of an angiographic version of OCT has extended the application of OCT to image the cutaneous microvasculature (so-called dynamic OCT, D-OCT). This study explores D-OCT's ability to differentiate the common BCC subtypes by microvascular and structural imaging. Eighty-one patients with 98 BCC lesions, consisting of three subtypes: 27 superficial BCC (sBCC), 55 nodular BCC (nBCC) and 16 infiltrative BCC (iBCC) were D-OCT scanned at three European dermatology centres. Blinded evaluations of microvascular and structural features were performed, followed by extensive statistical analysis of risk ratio (RR) and multiple correspondence analysis. nBCC lesions displayed most characteristic structural and vascular features. Serpiginous vessels, branching vessels, vessels creating a circumscribed figure and sharply demarcated hyporeflective ovoid structures in the dermis were all associated with a higher risk of the subtype being nBCC. The presence of highly present lines and dark peripheral borders at the margin of ovoid structures was negatively associated with iBCC. Lastly, the finding of hyporeflective ovoid structures protruding from epidermis correlated with sBCC. We identified various microvascular and structural D-OCT features that may aid non-invasive identification of BCC subtypes. This would allow clinicians to individualize and optimize BCC treatment as well as aid follow-up of non-surgical treatment.