In recent studies, convolutional neural networks (CNNs) outperformed dermatologists in distinguishing dermoscopic images of melanoma and nevi. In these studies, dermatologists and artificial ...intelligence were considered as opponents. However, the combination of classifiers frequently yields superior results, both in machine learning and among humans. In this study, we investigated the potential benefit of combining human and artificial intelligence for skin cancer classification.
Using 11,444 dermoscopic images, which were divided into five diagnostic categories, novel deep learning techniques were used to train a single CNN. Then, both 112 dermatologists of 13 German university hospitals and the trained CNN independently classified a set of 300 biopsy-verified skin lesions into those five classes. Taking into account the certainty of the decisions, the two independently determined diagnoses were combined to a new classifier with the help of a gradient boosting method. The primary end-point of the study was the correct classification of the images into five designated categories, whereas the secondary end-point was the correct classification of lesions as either benign or malignant (binary classification).
Regarding the multiclass task, the combination of man and machine achieved an accuracy of 82.95%. This was 1.36% higher than the best of the two individual classifiers (81.59% achieved by the CNN). Owing to the class imbalance in the binary problem, sensitivity, but not accuracy, was examined and demonstrated to be superior (89%) to the best individual classifier (CNN with 86.1%). The specificity in the combined classifier decreased from 89.2% to 84%. However, at an equal sensitivity of 89%, the CNN achieved a specificity of only 81.5%
Our findings indicate that the combination of human and artificial intelligence achieves superior results over the independent results of both of these systems.
•This article describes the first experiment on combining human and artificial intelligence for the classification of images suspicious of skin cancer.•The combination achieved a superior accuracy of 82.95% (compared to 81.59%/42.94% achieved by artificial/human intelligence alone).•The combination of human and artificial intelligence indicates superiority over a separated approach.
Recently, convolutional neural networks (CNNs) systematically outperformed dermatologists in distinguishing dermoscopic melanoma and nevi images. However, such a binary classification does not ...reflect the clinical reality of skin cancer screenings in which multiple diagnoses need to be taken into account.
Using 11,444 dermoscopic images, which covered dermatologic diagnoses comprising the majority of commonly pigmented skin lesions commonly faced in skin cancer screenings, a CNN was trained through novel deep learning techniques. A test set of 300 biopsy-verified images was used to compare the classifier's performance with that of 112 dermatologists from 13 German university hospitals. The primary end-point was the correct classification of the different lesions into benign and malignant. The secondary end-point was the correct classification of the images into one of the five diagnostic categories.
Sensitivity and specificity of dermatologists for the primary end-point were 74.4% (95% confidence interval CI: 67.0–81.8%) and 59.8% (95% CI: 49.8–69.8%), respectively. At equal sensitivity, the algorithm achieved a specificity of 91.3% (95% CI: 85.5–97.1%). For the secondary end-point, the mean sensitivity and specificity of the dermatologists were at 56.5% (95% CI: 42.8–70.2%) and 89.2% (95% CI: 85.0–93.3%), respectively. At equal sensitivity, the algorithm achieved a specificity of 98.8%. Two-sided McNemar tests revealed significance for the primary end-point (p < 0.001). For the secondary end-point, outperformance (p < 0.001) was achieved except for basal cell carcinoma (on-par performance).
Our findings show that automated classification of dermoscopic melanoma and nevi images is extendable to a multiclass classification problem, thus better reflecting clinical differential diagnoses, while still outperforming dermatologists at a significant level (p < 0.001).
•A convolutional neural network (CNN) received enhanced training with 11,444 dermoscopic images of >90% of lesions faced in a skin cancer screening setting.•The performance of 112 dermatologists from 13 university hospitals was then compared with the CNN on a test set of 300 biopsy-verified images.•The CNN achieved systematic outperformance of the dermatologists (p < 0.001) regardless of their clinical experience.
Human melanomas frequently harbor amplifications of EZH2. However, the contribution of EZH2 to melanoma formation has remained elusive. Taking advantage of murine melanoma models, we show that EZH2 ...drives tumorigenesis from benign BrafV600E- or NrasQ61K-expressing melanocytes by silencing of genes relevant for the integrity of the primary cilium, a signaling organelle projecting from the surface of vertebrate cells. Consequently, gain of EZH2 promotes loss of primary cilia in benign melanocytic lesions. In contrast, blockade of EZH2 activity evokes ciliogenesis and cilia-dependent growth inhibition in malignant melanoma. Finally, we demonstrate that loss of cilia enhances pro-tumorigenic WNT/β-catenin signaling, and is itself sufficient to drive metastatic melanoma in benign cells. Thus, primary cilia deconstruction is a key process in EZH2-driven melanomagenesis.
Display omitted
•EZH2 is an oncogene that drives metastatic BRAF and NRAS melanoma•EZH2 promotes primary cilium disassembly by suppressing ciliary genes•The primary cilium inhibits pro-tumorigenic WNT/β-catenin signaling in melanoma•Loss of cilia initiates metastatic melanoma
Zingg et al. show that EZH2 promotes melanomagenesis by silencing genes critical for primary cilium integrity, leading to loss of primary cilia and enhanced WNT signaling. Inhibition of EZH2 evokes cilia-dependent growth inhibition of melanoma, while loss of cilia is sufficient to drive melanoma formation.
Background
Data on patients with type 1 diabetes mellitus (T1DM) and severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infections are sparse. This study aimed to investigate the ...association between SARS‐CoV‐2 infection and T1DM.
Methods
Data from the Prospective Diabetes Follow‐up (DPV) Registry were analyzed for diabetes patients tested for SARS‐CoV‐2 by polymerase chain reaction (PCR) in Germany, Austria, Switzerland, and Luxembourg during January 2020–June 2021, using Wilcoxon rank‐sum and chi‐square tests for continuous and dichotomous variables, adjusted for multiple testing.
Results
Data analysis of 1855 pediatric T1DM patients revealed no differences between asymptomatic/symptomatic infected and SARS‐CoV‐2 negative/positive patients regarding age, new‐onset diabetes, diabetes duration, and body mass index. Glycated hemoglobin A1c (HbA1c) and diabetic ketoacidosis (DKA) rate were not elevated in SARS‐CoV‐2‐positive vs. ‐negative patients. The COVID‐19 manifestation index was 37.5% in individuals with known T1DM, but 57.1% in individuals with new‐onset diabetes. 68.8% of positively tested patients were managed as outpatients/telemedically. Data analysis of 240 adult T1MD patients revealed no differences between positively and negatively tested patients except lower HbA1c. Of these patients, 83.3% had symptomatic infections; 35.7% of positively tested patients were hospitalized.
Conclusions
Our results indicate low morbidity in SARS‐CoV‐2‐infected pediatric T1DM patients. Most patients with known T1DM and SARS‐CoV‐2 infections could be managed as outpatients. However, SARS‐CoV‐2 infection was usually symptomatic if it coincided with new‐onset diabetes. In adult patients, symptomatic SARS‐CoV‐2 infection and hospitalization were associated with age.
摘要
背景
关于1型糖尿病(T1DM)和严重急性呼吸综合征冠状病毒2型(SARS‐CoV‐2)感染患者的数据很少。本研究旨在探讨SARS‐CoV‐2感染与T1DM的关系。
方法
对德国, 奥地利, 瑞士和卢森堡在2020年1月至2021年6月期间进行SARS‐CoV‐2检测的糖尿病患者的前瞻性糖尿病随访(DPV)登记数据进行分析,采用Wilcoxon秩和检验和卡方检验,对连续变量和二分变量进行多重检验。
结果
1,855例儿童T1DM患者的数据分析显示,无症状/有症状感染和SARS‐CoV‐2阴性/阳性患者在年龄, 新发糖尿病, 糖尿病病程和体重指数方面没有差异。SARS‐CoV‐2阳性与阴性患者糖化血红蛋白(HbA1c)和糖尿病酮症酸中毒(DKA)发生率无明显差异。在已知的T1DM患者中,COVID‐19的表现指数为37.5%,而在新发糖尿病患者中为57.1%。68.8%的阳性患者以门诊/远程医疗的形式进行管理。对240例成人T1MD患者的数据分析显示,阳性和阴性患者之间除HbA1c降低外,其他均无差异。83.3%的患者有症状感染,35.7%的阳性患者住院治疗。
结论
SARS‐CoV‐2感染的儿童T1 DM患者的发病率较低。大多数已知的T1糖尿病和SARS‐CoV‐2感染的患者可以作为门诊患者进行管理。然而,如果SARS‐CoV‐2感染与新发糖尿病重合,通常是有症状的。在成人患者中,有症状的SARS‐CoV‐2感染和住院与年龄有关。
Highlights
Diabetic ketoacidosis rate and HbA1c were not elevated in pediatric patients with type 1 diabetes mellitus (T1DM) and severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection.
Pediatric patients with known T1DM did not show an elevated COVID‐19 manifestation index; only pediatric patients with coincident SARS‐CoV‐2 infection and new‐onset diabetes had an elevated rate of symptomatic infection.
Adult patients showed an age‐dependent increase in symptomatic SARS‐CoV‐2 infections and hospitalization rate.
Wrapping of nanoparticles by membranes Bahrami, Amir H.; Raatz, Michael; Agudo-Canalejo, Jaime ...
Advances in colloid and interface science,
06/2014, Letnik:
208
Journal Article
Recenzirano
How nanoparticles interact with biomembranes is central for understanding their bioactivity. Biomembranes wrap around nanoparticles if the adhesive interaction between the nanoparticles and membranes ...is sufficiently strong to compensate for the cost of membrane bending. In this article, we review recent results from theory and simulations that provide new insights on the interplay of bending and adhesion energies during the wrapping of nanoparticles by membranes. These results indicate that the interplay of bending and adhesion during wrapping is strongly affected by the interaction range of the particle–membrane adhesion potential, by the shape of the nanoparticles, and by shape changes of membrane vesicles during wrapping. The interaction range of the particle–membrane adhesion potential is crucial both for the wrapping process of single nanoparticles and the cooperative wrapping of nanoparticles by membrane tubules.
Display omitted
•The range of the particle-membrane interaction strongly affects membrane wrapping.•The wrapping process depends on the particle shape.•Shape changes of membrane vesicles during wrapping assist the wrapping process.
Progression-free survival (PFS) of patients with lower-risk myelodysplastic syndromes (MDS) treated with red blood cell transfusions is usually reduced, but it is unclear whether transfusion dose ...density is an independent prognostic factor. The European MDS Registry collects prospective data at 6-monthly intervals from newly diagnosed lower-risk myelodysplastic syndromes patients in 16 European countries and Israel. Data on the transfusion dose density - the cumulative dose received at the end of each interval divided by the time since the beginning of the interval in which the first transfusion was received - were analyzed using proportional hazards regression with time-varying co-variates, with death and progression to higher-risk MDS/acute myeloid leukemia as events. Of the 1,267 patients included in the analyses, 317 died without progression; in 162 patients the disease had progressed. PFS was significantly associated with age, EQ-5D index, baseline World Health Organization classification, bone marrow blast count, cytogenetic risk category, number of cytopenias, and country. Transfusion dose density was inversely associated with PFS (
<1×10
): dose density had an increasing effect on hazard until a dose density of 3 units/16 weeks. The transfusion dose density effect continued to increase beyond 8 units/16 weeks after correction for the impact of treatment with erythropoiesis-stimulating agents, lenalidomide and/or iron chelators. In conclusion, the negative effect of transfusion treatment on PFS already occurs at transfusion densities below 3 units/16 weeks. This indicates that transfusion dependency, even at relatively low dose densities, may be considered as an indicator of inferior PFS. This trial was registered at
Among low-risk patients with severe, symptomatic aortic stenosis who are eligible for both transcatheter aortic-valve implantation (TAVI) and surgical aortic-valve replacement (SAVR), data are ...lacking on the appropriate treatment strategy in routine clinical practice.
In this randomized noninferiority trial conducted at 38 sites in Germany, we assigned patients with severe aortic stenosis who were at low or intermediate surgical risk to undergo either TAVI or SAVR. Percutaneous- and surgical-valve prostheses were selected according to operator discretion. The primary outcome was a composite of death from any cause or fatal or nonfatal stroke at 1 year.
A total of 1414 patients underwent randomization (701 to the TAVI group and 713 to the SAVR group). The mean (±SD) age of the patients was 74±4 years; 57% were men, and the median Society of Thoracic Surgeons risk score was 1.8% (low surgical risk). The Kaplan-Meier estimate of the primary outcome at 1 year was 5.4% in the TAVI group and 10.0% in the SAVR group (hazard ratio for death or stroke, 0.53; 95% confidence interval CI, 0.35 to 0.79; P<0.001 for noninferiority). The incidence of death from any cause was 2.6% in the TAVI group and 6.2% in the SAVR group (hazard ratio, 0.43; 95% CI, 0.24 to 0.73); the incidence of stroke was 2.9% and 4.7%, respectively (hazard ratio, 0.61; 95% CI, 0.35 to 1.06). Procedural complications occurred in 1.5% and 1.0% of patients in the TAVI and SAVR groups, respectively.
Among patients with severe aortic stenosis at low or intermediate surgical risk, TAVI was noninferior to SAVR with respect to death from any cause or stroke at 1 year. (Funded by the German Center for Cardiovascular Research and the German Heart Foundation; DEDICATE-DZHK6 ClinicalTrials.gov number, NCT03112980.).
儿童和成人患者中的1型糖尿病和SARS‐CoV‐2——来自DPV网络的数据 Bastian Raphael Büttner; Sascha René Tittel; Clemens Kamrath ...
Journal of diabetes,
11/2022, Letnik:
14, Številka:
11
Journal Article
Recenzirano
Odprti dostop
Abstract Background Data on patients with type 1 diabetes mellitus (T1DM) and severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infections are sparse. This study aimed to investigate the ...association between SARS‐CoV‐2 infection and T1DM. Methods Data from the Prospective Diabetes Follow‐up (DPV) Registry were analyzed for diabetes patients tested for SARS‐CoV‐2 by polymerase chain reaction (PCR) in Germany, Austria, Switzerland, and Luxembourg during January 2020–June 2021, using Wilcoxon rank‐sum and chi‐square tests for continuous and dichotomous variables, adjusted for multiple testing. Results Data analysis of 1855 pediatric T1DM patients revealed no differences between asymptomatic/symptomatic infected and SARS‐CoV‐2 negative/positive patients regarding age, new‐onset diabetes, diabetes duration, and body mass index. Glycated hemoglobin A1c (HbA1c) and diabetic ketoacidosis (DKA) rate were not elevated in SARS‐CoV‐2‐positive vs. ‐negative patients. The COVID‐19 manifestation index was 37.5% in individuals with known T1DM, but 57.1% in individuals with new‐onset diabetes. 68.8% of positively tested patients were managed as outpatients/telemedically. Data analysis of 240 adult T1MD patients revealed no differences between positively and negatively tested patients except lower HbA1c. Of these patients, 83.3% had symptomatic infections; 35.7% of positively tested patients were hospitalized. Conclusions Our results indicate low morbidity in SARS‐CoV‐2‐infected pediatric T1DM patients. Most patients with known T1DM and SARS‐CoV‐2 infections could be managed as outpatients. However, SARS‐CoV‐2 infection was usually symptomatic if it coincided with new‐onset diabetes. In adult patients, symptomatic SARS‐CoV‐2 infection and hospitalization were associated with age.