Effectiveness of an image analyzing AI-based Digital Health Technology to identify Non-Melanoma Skin Cancer and other skin lesions: results of the DERM-003 study.

Introduction: Identification of skin cancer by an Artificial Intelligence (AI)-based Digital Health Technology could help improve the triage and management of suspicious skin lesions. Methods: The DERM-003 study (NCT04116983) was a prospective, multi-center, single-arm, masked study that aimed to demonstrate the effectiveness of an AI as a Medical Device (AIaMD) to identify Squamous Cell Carcinoma (SCC), Basal Cell Carcinoma (BCC), pre-malignant and benign lesions from dermoscopic images of suspicious skin lesions. Suspicious skin lesions that were suitable for photography were photographed with 3 smartphone cameras (iPhone 6S, iPhone 11, Samsung 10) with a DL1 dermoscopic lens attachment. Dermatologists provided clinical diagnoses and histopathology results were obtained for biopsied lesions. Each image was assessed by the AIaMD and the output compared to the ground truth diagnosis. Results: 572 patients (49.5% female, mean age 68.5 years, 96.9% Fitzpatrick skin types I-III) were recruited from 4 UK NHS Trusts, providing images of 611 suspicious lesions. 395 (64.6%) lesions were biopsied; 47 (11%) were diagnosed as SCC and 184 (44%) as BCC. The AIaMD AUROC on images taken by iPhone 6S was 0.88 (95% CI: 0.83-0.93) for SCC and 0.87 (95% CI: 0.84-0.91) for BCC. For Samsung 10 the AUROCs were 0.85 (95% CI: 0.79-0.90) and 0.87 (95% CI, 0.83-0.90), and for the iPhone 11 they were 0.88 (95% CI, 0.84-0.93) and 0.89 (95% CI, 0.86-0.92) for SCC and BCC, respectively. Using pre-determined diagnostic thresholds on images taken on the iPhone 6S the AIaMD achieved a sensitivity and specificity of 98% (95% CI, 88-100%) and 38% (95% CI, 33-44%) for SCC; and 94% (95% CI, 90-97%) and 28% (95 CI, 21-35%) for BCC. All 16 lesions diagnosed as melanoma in the study were correctly classified by the AIaMD. Discussion: The AIaMD has the potential to support the timely diagnosis of malignant and premalignant skin lesions.

Clinical outcomes and operational impact of a medical photography based teledermatology service with over 8,000 patients in the UK.

A new, store-and-forward, fully digitised Teledermatology (TD) pathway was designed and implemented in an urban setting for non-two-week wait routine patients. In total 8,352 new patients had a TD consultation over 37 months. Of these, 4,748 (56.8%) were referred back to their GP, 1,634 (19.6%) were referred directly for a surgical procedure and 1,970 (23.6%) for a face-to-face review with a Dermatologist (F2F). The average waiting time for a TD appointment was 3 vs. 30 weeks for a routine F2F appointment. Between 2019 and 2018, TD referrals rose by 38%, routine dermatology referrals reduced by 16% and cancer referrals increased by 6%. Using medical photographers proved to be effective with only two cases (0.02%) of images being of insufficient quality to form a clinical opinion. Hitherto, savings for the local Commissioning Groups were estimated at £671,218. Last financial year savings (2019-2020) were £284,671. The average cost savings per TD patient appointment was £80.36. Savings in the Trust's overhead costs were £53,587. TD consultants reviewed almost twice the number of patients vs. F2F for the same amount of consultant programmed activities. 95% of surveyed patients would be likely or extremely likely to recommend this service to friends and family.

Detection of Malignant Melanoma Using Artificial Intelligence: An Observational Study of Diagnostic Accuracy.

BACKGROUND: Malignant melanoma can most successfully be cured when diagnosed at an early stage in the natural history. However, there is controversy over screening programs and many advocate screening only for high-risk individuals. OBJECTIVES: This study aimed to evaluate the accuracy of an artificial intelligence neural network (Deep Ensemble for Recognition of Melanoma [DERM]) to identify malignant melanoma from dermoscopic images of pigmented skin lesions and to show how this compared to doctors' performance assessed by meta-analysis. METHODS: DERM was trained and tested using 7,102 dermoscopic images of both histologically confirmed melanoma (24%) and benign pigmented lesions (76%). A meta-analysis was conducted of studies examining the accuracy of naked-eye examination, with or without dermoscopy, by specialist and general physicians whose clinical diagnosis was compared to histopathology. The meta-analysis was based on evaluation of 32,226 pigmented lesions including 3,277 histopathology-confirmed malignant melanoma cases. The receiver operating characteristic (ROC) curve was used to examine and compare the diagnostic accuracy. RESULTS: DERM achieved a ROC area under the curve (AUC) of 0.93 (95% confidence interval: 0.92-0.94), and sensitivity and specificity of 85.0% and 85.3%, respectively. Avoidance of false-negative results is essential, so different decision thresholds were examined. At 95% sensitivity DERM achieved a specificity of 64.1% and at 95% specificity the sensitivity was 67%. The meta-analysis showed primary care physicians (10 studies) achieve an AUC of 0.83 (95% confidence interval: 0.79-0.86), with sensitivity and specificity of 79.9% and 70.9%; and dermatologists (92 studies) 0.91 (0.88-0.93), 87.5%, and 81.4%, respectively. CONCLUSIONS: DERM has the potential to be used as a decision support tool in primary care, by providing dermatologist-grade recommendation on the likelihood of malignant melanoma.

Assessment of Accuracy of an Artificial Intelligence Algorithm to Detect Melanoma in Images of Skin Lesions.

Importance: A high proportion of suspicious pigmented skin lesions referred for investigation are benign. Techniques to improve the accuracy of melanoma diagnoses throughout the patient pathway are needed to reduce the pressure on secondary care and pathology services. Objective: To determine the accuracy of an artificial intelligence algorithm in identifying melanoma in dermoscopic images of lesions taken with smartphone and digital single-lens reflex (DSLR) cameras. Design, Setting, and Participants: This prospective, multicenter, single-arm, masked diagnostic trial took place in dermatology and plastic surgery clinics in 7 UK hospitals. Dermoscopic images of suspicious and control skin lesions from 514 patients with at least 1 suspicious pigmented skin lesion scheduled for biopsy were captured on 3 different cameras. Data were collected from January 2017 to July 2018. Clinicians and the Deep Ensemble for Recognition of Malignancy, a deterministic artificial intelligence algorithm trained to identify melanoma in dermoscopic images of pigmented skin lesions using deep learning techniques, assessed the likelihood of melanoma. Initial data analysis was conducted in September 2018; further analysis was conducted from February 2019 to August 2019. Interventions: Clinician and algorithmic assessment of melanoma. Main Outcomes and Measures: Area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity of the algorithmic and specialist assessment, determined using histopathology diagnosis as the criterion standard. Results: The study population of 514 patients included 279 women (55.7%) and 484 white patients (96.8%), with a mean (SD) age of 52.1 (18.6) years. A total of 1550 images of skin lesions were included in the analysis (551 [35.6%] biopsied lesions; 999 [64.4%] control lesions); 286 images (18.6%) were used to train the algorithm, and a further 849 (54.8%) images were missing or unsuitable for analysis. Of the biopsied lesions that were assessed by the algorithm and specialists, 125 (22.7%) were diagnosed as melanoma. Of these, 77 (16.7%) were used for the primary analysis. The algorithm achieved an AUROC of 90.1% (95% CI, 86.3%-94.0%) for biopsied lesions and 95.8% (95% CI, 94.1%-97.6%) for all lesions using iPhone 6s images; an AUROC of 85.8% (95% CI, 81.0%-90.7%) for biopsied lesions and 93.8% (95% CI, 91.4%-96.2%) for all lesions using Galaxy S6 images; and an AUROC of 86.9% (95% CI, 80.8%-93.0%) for biopsied lesions and 91.8% (95% CI, 87.5%-96.1%) for all lesions using DSLR camera images. At 100% sensitivity, the algorithm achieved a specificity of 64.8% with iPhone 6s images. Specialists achieved an AUROC of 77.8% (95% CI, 72.5%-81.9%) and a specificity of 69.9%. Conclusions and Relevance: In this study, the algorithm demonstrated an ability to identify melanoma from dermoscopic images of selected lesions with an accuracy similar to that of specialists.

Tiger-like mycosis fungoides: an unusual clinical presentation of a rare variant of mycosis fungoides.

Mycosis fungoides (MF) is the most common type of cutaneous T-cell lymphoma. Mycosis fungoides classically presents in the skin as patches, plaques, tumors, or erythroderma, progressing to involve the lymph nodes and peripheral blood. The many clinical variants, with different histologic patterns, and the subtle early clinical and histologic changes may delay early diagnosis and present a diagnostic challenge for clinicians. The greatest challenge in diagnosis is the pre-mycotic stage, which may closely resemble eczematous or psoriasiform dermatitis clinically and histologically. The persistence of lesions and inadequate response to treatment are the first warning signs. Later stages of MF have a poor prognosis with poor therapeutic response and fatal outcome. We describe a 72-year-old man, who presented with a two-year history of an unusual eruption, which started on the abdomen, around the waistline, and gradually spread to involve his back, trunk, and buttocks. Clinically, the skin eruption presented as tiger-like stripes. The diagnosis was confirmed after histopathologic examination. The patient was treated with NB-UVB phototherapy with marked improvement.

Genome-wide association study in frontal fibrosing alopecia identifies four susceptibility loci including HLA-B*07:02.

Frontal fibrosing alopecia (FFA) is a recently described inflammatory and scarring type of hair loss affecting almost exclusively women. Despite a dramatic recent increase in incidence the aetiopathogenesis of FFA remains unknown. We undertake genome-wide association studies in females from a UK cohort, comprising 844 cases and 3,760 controls, a Spanish cohort of 172 cases and 385 controls, and perform statistical meta-analysis. We observe genome-wide significant association with FFA at four genomic loci: 2p22.2, 6p21.1, 8q24.22 and 15q2.1. Within the 6p21.1 locus, fine-mapping indicates that the association is driven by the HLA-B*07:02 allele. At 2p22.1, we implicate a putative causal missense variant in CYP1B1, encoding the homonymous xenobiotic- and hormone-processing enzyme. Transcriptomic analysis of affected scalp tissue highlights overrepresentation of transcripts encoding components of innate and adaptive immune response pathways. These findings provide insight into disease pathogenesis and characterise FFA as a genetically predisposed immuno-inflammatory disorder driven by HLA-B*07:02.

Severe inflammatory response and vasculitis leading to quadruple limb amputations.

A systemic inflammatory response causing multi-organ failure and requiring multiple amputations was refractory to all treatments except Anakinra, and the cause remains unclear.

Mohs micrographic surgery concordance between Mohs surgeons and dermatopathologists.

BACKGROUND: Mohs micrographic surgery (MMS) is the preferred treatment modality for high-risk nonmelanoma skin cancer because of the high cure rates and tissue-sparing effect. Its outcome is highly dependent on the expertise and accuracy of the Mohs surgeon in the interpretation of frozen sections. OBJECTIVE: This retrospective study evaluated the level of concordance between Mohs surgeons and dermatopathologists in reading histology slides from MMS procedures. METHODS AND MATERIALS: A Mohs surgeon read 170 randomly selected slides for a quality assurance audit during excision, and then a dermatopathologist blindly read them at a separate time. Absence or presence of tumour and the final diagnosis were recorded on a standardized form. RESULTS: An overall concordance of 99.4% was demonstrated. True discordance was recorded in only one of 170 cases. Intraepidermal atypia was the most challenging scenario for Mohs surgeons. CONCLUSIONS: The high rate of agreement in this study confirms that adequately trained MMS surgeons have sufficient expertise and training for accurate and precise frozen sections interpretation.

Patients are happy to be informed of their final non-melanoma skin cancer results by post.

During the past year, because of increasing pressure to see more patients, we have started to write to our patients informing them in a letter of their final skin cancer histology results following surgery for non-melanoma skin cancers: basal cell carcinoma and squamous cell carcinoma only. A questionnaire-based study was carried out to assess whether patients were happy to receive information concerning their non-melanoma skin cancer diagnosis in a carefully worded letter. One-hundred fifty patients were involved with a diagnosis of "completely excised non-melanoma skin cancer (NMSC)" that had previously received their final diagnosis by post. Seventy-seven (51.3%) patients responded to the questionnaire. Eighty-seven percent felt that they had been given the cancer diagnosis in an appropriate manner; 90 percent reported that they had understood the explanation about their skin cancer. In addition, 81 percent stated that they had been sufficiently involved in the discussion about their skin cancer and its treatment. Patients gave an average rating of 7.76 (1 = poor and 10 = excellent) for the overall experience of care that they had received. By writing to the patient with their final histology results, we have reduced the number of follow-up appointments without reducing the quality of patient care.

Microcystic adnexal carcinoma: a case series treated with mohs micrographic surgery and identification of patients in whom paraffin sections may be preferable.

BACKGROUND: Microcystic adnexal carcinoma (MAC) is a rare cutaneous tumor characterized by aggressive local infiltration, including a high propensity for perineural invasion (PNI). OBJECTIVES: To report our experience in treating MAC using Mohs micrographic surgery (MMS) with frozen sections and to identify patients in whom that technique may have limitations. MATERIALS & METHODS: A review of records between 1992 and 2008. RESULTS: Nine patients with MAC were identified. All tumors were located on the face. PNI was noted in the diagnostic biopsies of two patients with periocular MAC, in both of whom tumor persisted after MMS. The mean duration of follow-up was 5.4 years. CONCLUSIONS: MMS with frozen sections is reliable for treating primary MAC in which PNI is not present on a diagnostic biopsy. Previous surgery and PNI were associated with greater risk of persistence in periocular MAC. In these patients, it may be appropriate to consider MMS with paraffin-embedded sections, possibly as a layer after apparent clearance on frozen sections. Further excision of orbital contents should be considered in periocular MAC that infiltrate the deep orbital fat or are noted to have PNI.

An unusual presentation of a common disease.

We present a case of perforating granuloma annulare (PGA), in which we show the natural history of lesions and outline the different clinical types. Our patient responded well to intralesional triamcinolone acetonide 10mg/ml injections. Although she was otherwise well, PGA can be associated with diabetes mellitus in up to 17 percent of cases. Differential clinico-histopathological diagnosis, specifically in relation to necrobiosis lipoidica diabeticorum is being discussed. Treatment for PGA is difficult; apart from topical and/or intralesional steroids (clearance up to 54% in one case series), other options include Psoralen plus UVA (PUVA), systemic isotretinoin, chloroquine or hydroxylchloroquine, sulphapyridine, dapsone, topical application of imidazole creams, and liquid nitrogen. Spontaneous remission has also been reported in up to 77 percent in one case series.