Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 150
Filtrar
1.
J Cutan Pathol ; 2021 Feb 11.
Artigo em Inglês | MEDLINE | ID: mdl-33576022

RESUMO

BACKGROUND: Novel solutions are needed for expediting margin assessment to guide BCC surgeries. Ex-vivo fluorescence confocal microscopy (FCM) is starting to be used in freshly-excised surgical specimens to examine BCC margins in real-time. Training and educational process are needed for this novel technology to be implemented into clinic. OBJECTIVE: To test a training and reading process, and measure diagnostic accuracy of clinicians with varying expertise level in reading ex-vivo FCM images. METHODS: An international 3-center study was designed for training and reading to assess BCC surgical margins and residual subtypes. Each center included a lead dermatologic/Mohs surgeon (clinical developer of FCM), and 3 additional readers (dermatologist, dermatopathologist, dermatologic/Mohs surgeon), who use confocal in clinical practice. Testing was conducted on 30 samples. RESULTS: Overall, the readers achieved 90% average sensitivity, 78% average specificity in detecting residual BCC margins, showing high and consistent diagnostic reading accuracy. Those with expertise in dermatologic surgery and dermatopathology showed the strongest potential for learning to assess FCM images. LIMITATIONS: Small dataset, variability in mosaic quality between centers. CONCLUSION: Suggested process is feasible and effective. This process is proposed for wider implementation, to facilitate wider adoption of FCM to potentially expedite BCC margin assessment to guide surgery in real-time. This article is protected by copyright. All rights reserved.

3.
Sci Data ; 8(1): 34, 2021 01 28.
Artigo em Inglês | MEDLINE | ID: mdl-33510154

RESUMO

Prior skin image datasets have not addressed patient-level information obtained from multiple skin lesions from the same patient. Though artificial intelligence classification algorithms have achieved expert-level performance in controlled studies examining single images, in practice dermatologists base their judgment holistically from multiple lesions on the same patient. The 2020 SIIM-ISIC Melanoma Classification challenge dataset described herein was constructed to address this discrepancy between prior challenges and clinical practice, providing for each image in the dataset an identifier allowing lesions from the same patient to be mapped to one another. This patient-level contextual information is frequently used by clinicians to diagnose melanoma and is especially useful in ruling out false positives in patients with many atypical nevi. The dataset represents 2,056 patients (20.8% with at least one melanoma, 79.2% with zero melanomas) from three continents with an average of 16 lesions per patient, consisting of 33,126 dermoscopic images and 584 (1.8%) histopathologically confirmed melanomas compared with benign melanoma mimickers.


Assuntos
Melanoma , Neoplasias Cutâneas , Inteligência Artificial , Humanos , Melanoma/diagnóstico por imagem , Melanoma/patologia , Melanoma/fisiopatologia , Metadados , Pele/patologia , Neoplasias Cutâneas/diagnóstico por imagem , Neoplasias Cutâneas/patologia , Neoplasias Cutâneas/fisiopatologia
4.
Arch Dermatol Res ; 2021 Jan 30.
Artigo em Inglês | MEDLINE | ID: mdl-33515277

RESUMO

Patient expectations of the scar after Mohs micrographic surgery (MMS) are often not realistic, leading to subsequent psychosocial sequelae such as anxiety, depression, and avoidance of social situations. When patient expectations are not met, this may also contribute to a decrease in patient satisfaction after surgery. Therefore, altering expectation levels may change patient satisfaction and psychosocial distress levels after surgery. To assess whether patient satisfaction improves in patients after MMS when patients view the surgical defect prior to reconstruction. Patients undergoing facial MMS between December 2017 and September 2019 were included. Patients received or did not receive a mirror after MMS to view the surgical defect before closing the defect. Patients were asked to complete the Dutch FACE-Q Skin Cancer before, one-week, three-months, and one-year after MMS. A total of 113 patients where included. One-hundred-eight (95.6%), 113 (100%), and 93 (82.3%) questionnaires were completed, one-week, three-months, and one-year follow-up, respectively. Satisfaction with facial appearance and appraisal of scars significantly improved over time for all patients, no such improvement was seen for appearance-related distress. Female patients who looked in the mirror had higher satisfaction with facial appearance than female patients who did not look in the mirror. Also, lower appearance-related distress scores were seen in patients who looked in the mirror prior to a flap reconstruction. Showing the defect in the mirror prior to the reconstruction may result in higher patient satisfaction in female patients and patients before undergoing a flap reconstruction.

5.
J Am Acad Dermatol ; 2020 Oct 20.
Artigo em Inglês | MEDLINE | ID: mdl-33096134

RESUMO

BACKGROUND: There are no well-defined clinical factors to predict the risk of occult invasion in melanoma of the lentigo maligna type (LM) prior to complete histopathologic analysis. OBJECTIVE: To evaluate whether clinical size was a predictor of invasion in LM and subclinical extension. METHODS: Consecutive cases of LM were recorded in a prospectively maintained database from 2006 to 2019. Patient and tumor data were recorded during initial evaluation. 'LM clinical area' was calculated in square millimeters (length x width). All patients were treated with staged excision. RESULTS: We included 600 patients. Mean age was 65.9 years (SD 12.3; range 27 - 95 years); 62.8% (n=377) were males. The mean LM clinical area was 128.32 mm2 for in situ lesions vs 200.14 mm for invasive lesions (p=0.1). Based on quantile regression, the median margin required for complete removal increased with LM clinical area. LIMITATIONS: study performed in a tertiary cancer center with possible referral bias and more complex cases. CONCLUSIONS: LM can present with variable clinical size which may correlate with subclinical extension; however, the presence of invasion is not well estimated by LM clinical area.

7.
J Cutan Pathol ; 2020 Sep 28.
Artigo em Inglês | MEDLINE | ID: mdl-32989842

RESUMO

BACKGROUND: Accurate basal cell carcinoma (BCC) subtyping is requisite for appropriate management, but non-representative sampling occurs in 18% to 25% of biopsies. By enabling non-invasive diagnosis and more comprehensive sampling, integrated reflectance confocal microscopy-optical coherence tomography (RCM-OCT) may improve the accuracy of BCC subtyping and subsequent management. We evaluated RCM-OCT images and histopathology slides for the presence of two key features, angulation and small nests and cords, and calculated (a) sensitivity and specificity of these features, combined and individually, for identifying an infiltrative BCC subtype and (b) agreement across modalities. METHODS: Thirty-three RCM-OCT-imaged, histopathologically-proven BCCs (17 superficial and/or nodular; 16 containing an infiltrative component) were evaluated. RESULTS: The presence of angulation or small nests and cords was sufficient to identify infiltrative BCC on RCM-OCT with 100% sensitivity and 82% specificity, similar to histopathology (100% sensitivity, 88% specificity, kappa = 0.82). When both features were present, the sensitivity for identifying infiltrative BCC was 100% using either modality and specificity was 88% on RCM-OCT vs 94% on histopathology, indicating near-perfect agreement between non-invasive and invasive diagnostic modalities (kappa = 0.94). CONCLUSIONS: RCM-OCT can non-invasively identify key histopathologic features of infiltrative BCC offering a possible alternative to traditional invasive biopsy.

8.
JAMA Dermatol ; 2020 Sep 16.
Artigo em Inglês | MEDLINE | ID: mdl-32936222

RESUMO

Importance: The development of new primary cutaneous melanoma (CM) after starting immune checkpoint inhibitor (ICI) therapy is poorly characterized. Objective: To determine the incidence of new CM in patients treated with ipilimumab, nivolumab, and/or pembrolizumab for metastatic melanoma. Design, Setting, and Participants: Single-center, retrospective, observational cohort study using an institutional database to identify patients diagnosed with melanoma at a tertiary care cancer hospital in New York, New York. Exposures: Ipilimumab, nivolumab, and/or pembrolizumab treatment for metastatic melanoma. Main Outcomes and Measures: Primary outcomes were the incidence proportion, the incidence rate, and the 5-year cause-specific cumulative risk. Results: A total of 2251 patients were included in the study; mean (SD) age at the time of ICI start was 62.8 (14.4) years. The majority were male (63.8%, n = 1437), White (92.7%, n = 2086), and non-Hispanic (92.1%, n = 2073). Forty-two of 2251 patients who received ipilimumab, nivolumab, and/or pembrolizumab were diagnosed with 48 new CMs at a median (range) of 397.5 (39-2409) days after ICI initiation. The median age of affected patients at the time of ICI first dose was 66.5 years. The majority were male (66.7%, n = 28), White (92.9%, n = 39), and non-Hispanic (100.0%, n = 42). There were no differences in age, sex, race, and ethnicity among patients who did and did not develop a new CM. Patients who developed a new CM were more likely to have a family history of melanoma (23.8% vs 16.3%, P = .02). Most new CMs (n = 30, 62.5%) were diagnosed after the last date of ICI administration. Twenty-seven (56.3%) new CMs were in situ and 21 (43.8%) were invasive. Of the invasive CMs with a reported Breslow thickness (n = 20), the median (range) thickness was 0.4 (0.1-8.4) mm. The overall incidence proportion of new CM was 1.9% (95% CI, 1.4%-2.5%) and the incidence rate was 1103 cases per 100 000 person-years (95% CI, 815-1492). The 5-year cumulative cause-specific risk of new CM was 4.9% (95% CI, 3.3%-7.4%). Conclusions and Relevance: Patients treated with ICI therapy for metastatic melanoma remain at risk for the development of new CM.

9.
JAMA Dermatol ; 156(9): 953-962, 2020 Sep 01.
Artigo em Inglês | MEDLINE | ID: mdl-32745161

RESUMO

Importance: The performance of prognostic gene expression profile (GEP) tests for cutaneous melanoma is poorly characterized. Objective: To systematically assess the performance of commercially available GEP tests in patients with American Joint Committee on Cancer (AJCC) stage I or stage II disease. Data Sources: For this systematic review and meta-analysis, comprehensive searches of PubMed/MEDLINE, Embase, and Web of Science were conducted on December 12, 2019, for English-language studies of humans without date restrictions. Study Selection: Two reviewers identified GEP external validation studies of patients with localized melanoma. After exclusion criteria were applied, 7 studies (8%; 5 assessing DecisionDx-Melanoma and 2 assessing MelaGenix) were included. Data Extraction and Synthesis: Data were extracted using an adaptation of the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies (CHARMS-PF). When feasible, meta-analysis using random-effects models was performed. Risk of bias and level of evidence were assessed with the Quality in Prognosis Studies tool and an adaptation of Grading of Recommendations Assessment, Development, and Evaluation. Main Outcomes and Measures: Proportion of patients with or without melanoma recurrence correctly classified by the GEP test as being at high or low risk. Results: In the 7 included studies, a total of 1450 study participants contributed data (age and sex unknown). The performance of both GEP tests varied by AJCC stage. Of patients tested with DecisionDx-Melanoma, 623 had stage I disease (6 true-positive [TP], 15 false-negative, 61 false-positive, and 541 true-negative [TN] results) and 212 had stage II disease (59 TP, 13 FN, 78 FP, and 62 TN results). Among patients with recurrence, DecisionDx-Melanoma correctly classified 29% with stage I disease and 82% with stage II disease. Among patients without recurrence, the test correctly classified 90% with stage I disease and 44% with stage II disease. Of patients tested with MelaGenix, 88 had stage I disease (7 TP, 15 FN, 15 FP, and 51 TN results) and 245 had stage II disease (59 TP, 19 FN, 95 FP, and 72 TN results). Among patients with recurrence, MelaGenix correctly classified 32% with stage I disease and 76% with stage II disease. Among patients without recurrence, the test correctly classified 77% with stage I disease and 43% with stage II disease. Conclusions and Relevance: The prognostic ability of GEP tests among patients with localized melanoma varied by AJCC stage and appeared to be poor at correctly identifying recurrence in patients with stage I disease, suggesting limited potential for clinical utility in these patients.

10.
JAMA Dermatol ; 156(9): 963-972, 2020 Sep 01.
Artigo em Inglês | MEDLINE | ID: mdl-32756880

RESUMO

Importance: Persistent radiation-induced alopecia (pRIA) and its management have not been systematically described. Objective: To characterize pRIA in patients with primary central nervous system (CNS) tumors or head and neck sarcoma. Design, Setting, and Participants: A retrospective cohort study of patients from January 1, 2011, to January 30, 2019, was conducted at 2 large tertiary care hospitals and comprehensive cancer centers. Seventy-one children and adults diagnosed with primary CNS tumors or head and neck sarcomas were evaluated for pRIA. Main Outcomes and Measures: The clinical and trichoscopic features, scalp radiation dose-response relationship, and response to topical minoxidil were assessed using standardized clinical photographs of the scalp, trichoscopic images, and radiotherapy treatment plans. Results: Of the 71 patients included (median [range] age, 27 [4-75] years; 51 female [72%]), 64 (90%) had a CNS tumor and 7 (10%) had head and neck sarcoma. Alopecia severity was grade 1 in 40 of 70 patients (56%), with localized (29 of 54 [54%]), diffuse (13 of 54 [24%]), or mixed (12 of 54 [22%]) patterns. The median (range) estimated scalp radiation dose was 39.6 (15.1-50.0) Gy; higher dose (odds ratio [OR], 1.15; 95% CI, 1.04-1.28) and proton irradiation (OR, 5.7; 95% CI, 1.05-30.8) were associated with greater alopecia severity (P < .001), and the dose at which 50% of patients were estimated to have severe (grade 2) alopecia was 36.1 Gy (95% CI, 33.7-39.6 Gy). Predominant trichoscopic features included white patches (16 of 28 [57%]); in 15 patients, hair-shaft caliber negatively correlated with scalp dose (correlation coefficient, -0.624; P = .01). The association between hair density and scalp radiation dose was not statistically significant (-0.381; P = .16). Twenty-eight of 34 patients (82%) responded to topical minoxidil, 5% (median follow-up, 61 [interquartile range, 21-105] weeks); 4 of 25 (16%) topical minoxidil recipients with clinical images improved in severity grade. Two patients responded to hair transplantation and 1 patient responded to plastic surgical reconstruction. Conclusions and Relevance: Persistent radiation-induced alopecia among patients with primary CNS tumors or head and neck sarcomas represents a dose-dependent phenomenon that has distinctive clinical and trichoscopic features. The findings of this study suggest that topical minoxidil and procedural interventions may have benefit in the treatment of pRIA.

12.
J Am Acad Dermatol ; 83(3): 780-787, 2020 Sep.
Artigo em Inglês | MEDLINE | ID: mdl-32360723

RESUMO

BACKGROUND: The number needed to biopsy (NNB) ratio for melanoma diagnosis is calculated by dividing the total number of biopsies by the number of biopsied melanomas. It is the inverse of positive predictive value (PPV), which is calculated by dividing the number of biopsied melanomas by the total number of biopsies. NNB is increasingly used as a metric to compare the diagnostic accuracy of health care practitioners. OBJECTIVE: To investigate the association of NNB with the standard statistical measures of sensitivity and specificity. METHODS: We extracted published diagnostic accuracy data from 5 cross-sectional skin cancer reader studies (median [min-max] readers/study was 29 [8-511]). Because NNB is a ratio, we converted it to PPV. RESULTS: Four studies showed no association and 1 showed a negative association between PPV and sensitivity. All 5 studies showed a positive association between PPV and specificity. LIMITATIONS: Reader study data. CONCLUSIONS: An individual health care practitioner with a lower NNB is likely to have a higher specificity than one with a higher NNB, assuming they practice under similar conditions; no conclusions can be made about their relative sensitivities. We advocate for additional research to define quality metrics for melanoma detection and caution when interpreting NNB.

13.
JAMA Dermatol ; 156(8): 882-890, 2020 Aug 01.
Artigo em Inglês | MEDLINE | ID: mdl-32459294

RESUMO

Importance: Basal cell carcinoma (BCC) is the most common skin cancer. Dermoscopic imaging has improved diagnostic accuracy; however, diagnosis of nonpigmented BCC remains limited to arborizing vessels, ulceration, and shiny white structures. Objective: To assess multiple aggregated yellow-white (MAY) globules as a diagnostic feature for BCC. Design, Setting, and Participants: In this retrospective, single-center, case-control study, nonpigmented skin tumors, determined clinically, were identified from a database of lesions consecutively biopsied during a 7-year period (January 1, 2009, to December 31, 2015). A subset of tumors was prospectively diagnosed, and reflectance confocal microscopy, optical coherence tomography, and histopathologic correlation were performed. Data analysis was conducted from July 1 to September 31, 2019. Exposures: Investigators evaluated for the presence or absence of known dermoscopic criteria. MAY globules were defined as aggregated, white-yellow structures visualized in polarized and nonpolarized light. Main Outcomes and Measures: The primary outcome was the diagnostic accuracy of MAY globules for the diagnosis of BCC. Secondary objectives included the association with BCC location and subtype. Interrater agreement was estimated. Results: A total of 656 nonpigmented lesions from 643 patients (mean [SD] age, 63.1 [14.9] years; 381 [58.1%] male) were included. In all, 194 lesions (29.6%) were located on the head and neck. A total of 291 (44.4%) were BCCs. MAY globules were seen in 61 of 291 BCC cases (21.0%) and in 3 of 365 other diagnoses (0.8%) (P < .001). The odds ratio for diagnosis of BCC was 32.0 (96% CI, 9.9-103.2). The presence of MAY globules was associated with a diagnosis of histologic high-risk BCC (odds ratio, 6.5; 95% CI, 3.1-14.3). The structure was never seen in cases of superficial BCCs. Conclusions and Relevance: The findings suggest that MAY globules may have utility as a new BCC dermoscopic criterion with a high specificity. MAY globules were negatively associated with superficial BCC and positively associated with deeper-seated, histologic, higher-grade tumor subtypes.

14.
J Am Acad Dermatol ; 83(2): 463-468, 2020 Aug.
Artigo em Inglês | MEDLINE | ID: mdl-32320764

RESUMO

BACKGROUND: Functional status assessment may help estimate which patients ≥85 years of age will benefit from surgical treatment for keratinocyte carcinoma (KC), but predictive value for short-term survival in this population has not been determined. OBJECTIVE: We sought to assess the predictive value of functional status for short-term survival in patients ≥85 years of age who have KC. METHODS: This was a retrospective cohort review of 238 patients ≥85 years of age who presented for the management of KC between 2010 and 2015. Functional status was assessed with the Karnofsky Performance Scale (KPS) and Katz Activities of Daily Living (ADL) index. Overall survival was determined. RESULTS: Lower functional status scores of KPS ≤40 and Katz ADL ≤4 were associated with 37% and 53% survival at 2 years, respectively. LIMITATIONS: Retrospective design and single-center study. CONCLUSION: In this study, KPS and Katz ADL predicted short-term survival. Patients with low functional status scores had significantly decreased survival at 2 years, with double the death rate of patients with high functional status. Functional status should be considered during shared decision-making for elderly individuals who are seeking treatment for KC.

15.
J Am Acad Dermatol ; 2020 Mar 12.
Artigo em Inglês | MEDLINE | ID: mdl-32171811

RESUMO

BACKGROUND: Severe cutaneous adverse reactions (SCARs) are associated with high morbidity and mortality in patients with cancer. Early identification and treatment of SCARs may improve outcomes. OBJECTIVE: To identify biomarkers to predict outcomes in hospitalized patients with cancer who developed SCARs. METHODS: Retrospective review of 144 hospitalized patients with cancer with a morbilliform rash, recorded testing for serum cytokines (interleukin [IL]-6, IL-10, and tumor necrosis factor [TNF]-α) or elafin, and a dermatology consultation. Rashes were categorized as simple morbilliform rash without systemic involvement or complex morbilliform rash with systemic involvement. RESULTS: Fifty-four of 144 (37.5%) patients died during follow-up. Elevated levels of IL-6, IL-10, and TNF-α were associated with decreased survival. Overall survivals in patients with elevated levels of IL-6, IL-10, and TNF-α were 53.7%, 56.6%, 53.6%, respectively, compared with 85.7%, 82.5% and 83.6%, respectively, in those with lower levels. Patients with increased levels of both IL-6 and TNF-α had a nearly 6-fold increase in mortality (hazard ratio, 5.82) compared with patients with lower levels. LIMITATIONS: Retrospective design, limited sample size, and high-risk population. CONCLUSIONS: Hospitalized patients with cancer with rash and elevated IL-6 and TNF-α were nearly 6 times more likely to die over the course of follow-up. These biomarkers may serve as prognostic biomarkers and therapeutic targets for this high-risk population.

16.
J Am Acad Dermatol ; 82(4): 962-968, 2020 Apr.
Artigo em Inglês | MEDLINE | ID: mdl-31634517

RESUMO

BACKGROUND: Initial biopsy of basal cell carcinoma (BCC) may fail to show aggressive histologic subtypes. Additionality, the clinical evaluation of BCC before surgery can miss subclinical extension. Reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) are emerging tools that can help in the presurgical evaluation of BCCs. OBJECTIVE: To assess the feasibility of a combined RCM-OCT imaging modality for presurgical evaluation of biopsy-proven BCCs for residual tumor, margin status, and depth. METHODS: Thirty-eight BCCs in 35 patients referred to a tertiary cancer center for Mohs micrographic surgery (MMS) were imaged with combined RCM-OCT. Images were correlated to MMS frozen sections. RESULTS: Thirty-eight BCCs were analyzed. The mean age of patients was 67.34 years (range, 36-84 years), and 20 patients were female (57.14%). Twenty four BCCs were located on the head (63.16%) , and the mean size was 8.58 mm (range, 3-30 mm). RCM-OCT showed an overall agreement of 91.1% with MMS frozen sections. A sensitivity of 82.6% (95% confidence interval [CI], 69%-92%), specificity of 93.8% (95% CI, 88%-97%), and receiver operating characteristic curve of 0.88 (95% CI, 0.82-0.94) was found. OCT depth was highly correlated with MMS depth (r2 = 0.9). LIMITATIONS: Small sample size and difficulty evaluating certain challenging anatomic sites. CONCLUSIONS: Combined RCM-OCT may emerge as a useful tool for presurgical evaluation of BCCs.


Assuntos
Carcinoma Basocelular/diagnóstico , Cuidados Pré-Operatórios/métodos , Neoplasias Cutâneas/diagnóstico , Pele/diagnóstico por imagem , Adulto , Idoso , Idoso de 80 Anos ou mais , Biópsia , Carcinoma Basocelular/patologia , Carcinoma Basocelular/cirurgia , Estudos de Viabilidade , Feminino , Humanos , Masculino , Margens de Excisão , Pessoa de Meia-Idade , Cirurgia de Mohs , Imagem Multimodal/métodos , Neoplasia Residual , Estudos Prospectivos , Sensibilidade e Especificidade , Pele/patologia , Neoplasias Cutâneas/patologia , Neoplasias Cutâneas/cirurgia , Tomografia de Coerência Óptica
17.
Dermatol Surg ; 46(4): 514-518, 2020 04.
Artigo em Inglês | MEDLINE | ID: mdl-31592922

RESUMO

BACKGROUND: Postoperative concerns after Mohs micrographic surgery (MMS) are not well characterized. OBJECTIVE: To better define patient concerns and contributing characteristics in the immediate postoperative period after MMS. MATERIALS AND METHODS: A standardized telephone encounter template was implemented to better assess patient concerns in the 24-hour postoperative period. A review was then performed of patients undergoing MMS from October 2016 to July 2017 to assess for the most common patient concerns and association with clinical characteristics. RESULTS: We included 307 patients. Overall, 60.6% of patients reported a concern. Fifty-four percent of patients reported pain. Most pain was characterized as "a little" (67.7%). On univariate analysis, flap repairs, location on the upper extremities, and swelling were associated with higher mean postoperative pain. Tumor type was not associated with increased pain. On multivariate analyses, patients with larger defects and associated edema were statistically significantly associated with higher degrees of pain. CONCLUSION: Over half of patients experience postoperative concerns after MMS, most commonly pain. The immediate postoperative period may be an optimal time to identify patient concerns allowing for reassurance or early intervention when necessary.


Assuntos
Assistência ao Convalescente/métodos , Cirurgia de Mohs/efeitos adversos , Dor Pós-Operatória/epidemiologia , Cuidados Pós-Operatórios/métodos , Neoplasias Cutâneas/cirurgia , Adulto , Assistência ao Convalescente/normas , Assistência ao Convalescente/estatística & dados numéricos , Idoso , Idoso de 80 Anos ou mais , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Medição da Dor , Dor Pós-Operatória/diagnóstico , Dor Pós-Operatória/etiologia , Cuidados Pós-Operatórios/normas , Cuidados Pós-Operatórios/estatística & dados numéricos , Período Pós-Operatório , Estudos Retrospectivos , Inquéritos e Questionários/normas , Inquéritos e Questionários/estatística & dados numéricos , Telefone
20.
J Am Acad Dermatol ; 82(3): 622-627, 2020 Mar.
Artigo em Inglês | MEDLINE | ID: mdl-31306724

RESUMO

BACKGROUND: Computer vision has promise in image-based cutaneous melanoma diagnosis but clinical utility is uncertain. OBJECTIVE: To determine if computer algorithms from an international melanoma detection challenge can improve dermatologists' accuracy in diagnosing melanoma. METHODS: In this cross-sectional study, we used 150 dermoscopy images (50 melanomas, 50 nevi, 50 seborrheic keratoses) from the test dataset of a melanoma detection challenge, along with algorithm results from 23 teams. Eight dermatologists and 9 dermatology residents classified dermoscopic lesion images in an online reader study and provided their confidence level. RESULTS: The top-ranked computer algorithm had an area under the receiver operating characteristic curve of 0.87, which was higher than that of the dermatologists (0.74) and residents (0.66) (P < .001 for all comparisons). At the dermatologists' overall sensitivity in classification of 76.0%, the algorithm had a superior specificity (85.0% vs. 72.6%, P = .001). Imputation of computer algorithm classifications into dermatologist evaluations with low confidence ratings (26.6% of evaluations) increased dermatologist sensitivity from 76.0% to 80.8% and specificity from 72.6% to 72.8%. LIMITATIONS: Artificial study setting lacking the full spectrum of skin lesions as well as clinical metadata. CONCLUSION: Accumulating evidence suggests that deep neural networks can classify skin images of melanoma and its benign mimickers with high accuracy and potentially improve human performance.


Assuntos
Aprendizado Profundo , Dermoscopia/métodos , Interpretação de Imagem Assistida por Computador/métodos , Melanoma/diagnóstico , Neoplasias Cutâneas/diagnóstico , Colômbia , Estudos Transversais , Dermatologistas/estatística & dados numéricos , Dermoscopia/estatística & dados numéricos , Diagnóstico Diferencial , Humanos , Cooperação Internacional , Internato e Residência/estatística & dados numéricos , Israel , Ceratose Seborreica/diagnóstico , Melanoma/patologia , Nevo/diagnóstico , Curva ROC , Pele/diagnóstico por imagem , Pele/patologia , Neoplasias Cutâneas/patologia , Espanha , Estados Unidos
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA
...