Uncertainty Estimation in Medical Image Classification: Systematic Review.

BACKGROUND: Deep neural networks are showing impressive results in different medical image classification tasks. However, for real-world applications, there is a need to estimate the network's uncertainty together with its prediction. OBJECTIVE: In this review, we investigate in what form uncertainty estimation has been applied to the task of medical image classification. We also investigate which metrics are used to describe the effectiveness of the applied uncertainty estimation. METHODS: Google Scholar, PubMed, IEEE Xplore, and ScienceDirect were screened for peer-reviewed studies, published between 2016 and 2021, that deal with uncertainty estimation in medical image classification. The search terms "uncertainty," "uncertainty estimation," "network calibration," and "out-of-distribution detection" were used in combination with the terms "medical images," "medical image analysis," and "medical image classification." RESULTS: A total of 22 papers were chosen for detailed analysis through the systematic review process. This paper provides a table for a systematic comparison of the included works with respect to the applied method for estimating the uncertainty. CONCLUSIONS: The applied methods for estimating uncertainties are diverse, but the sampling-based methods Monte-Carlo Dropout and Deep Ensembles are used most frequently. We concluded that future works can investigate the benefits of uncertainty estimation in collaborative settings of artificial intelligence systems and human experts. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/11936.

Digital Natives' Preferences on Mobile Artificial Intelligence Apps for Skin Cancer Diagnostics: Survey Study.

BACKGROUND: Artificial intelligence (AI) has shown potential to improve diagnostics of various diseases, especially for early detection of skin cancer. Studies have yet to investigate the clear application of AI technology in clinical practice or determine the added value for younger user groups. Translation of AI-based diagnostic tools can only be successful if they are accepted by potential users. Young adults as digital natives may offer the greatest potential for successful implementation of AI into clinical practice, while at the same time, representing the future generation of skin cancer screening participants. OBJECTIVE: We conducted an anonymous online survey to examine how and to what extent individuals are willing to accept AI-based mobile apps for skin cancer diagnostics. We evaluated preferences and relative influences of concerns, with a focus on younger age groups. METHODS: We recruited participants below 35 years of age using three social media channels-Facebook, LinkedIn, and Xing. Descriptive analysis and statistical tests were performed to evaluate participants' attitudes toward mobile apps for skin examination. We integrated an adaptive choice-based conjoint to assess participants' preferences. We evaluated potential concerns using maximum difference scaling. RESULTS: We included 728 participants in the analysis. The majority of participants (66.5%, 484/728; 95% CI 0.631-0.699) expressed a positive attitude toward the use of AI-based apps. In particular, participants residing in big cities or small towns (P=.02) and individuals that were familiar with the use of health or fitness apps (P=.02) were significantly more open to mobile diagnostic systems. Hierarchical Bayes estimation of the preferences of participants with a positive attitude (n=484) revealed that the use of mobile apps as an assistance system was preferred. Participants ruled out app versions with an accuracy of ≤65%, apps using data storage without encryption, and systems that did not provide background information about the decision-making process. However, participants did not mind their data being used anonymously for research purposes, nor did they object to the inclusion of clinical patient information in the decision-making process. Maximum difference scaling analysis for the negative-minded participant group (n=244) showed that data security, insufficient trust in the app, and lack of personal interaction represented the dominant concerns with respect to app use. CONCLUSIONS: The majority of potential future users below 35 years of age were ready to accept AI-based diagnostic solutions for early detection of skin cancer. However, for translation into clinical practice, the participants' demands for increased transparency and explainability of AI-based tools seem to be critical. Altogether, digital natives between 18 and 24 years and between 25 and 34 years of age expressed similar preferences and concerns when compared both to each other and to results obtained by previous studies that included other age groups.

Logistics Workers Are a Key Factor for SARS-CoV-2 Spread in Brazilian Small Towns: Case-Control Study.

BACKGROUND: Data on how SARS-CoV-2 enters and spreads in a population are essential for guiding public policies. OBJECTIVE: This study seeks to understand the transmission dynamics of SARS-CoV-2 in small Brazilian towns during the early phase of the epidemic and to identify core groups that can serve as the initial source of infection as well as factors associated with a higher risk of COVID-19. METHODS: Two population-based seroprevalence studies, one household survey, and a case-control study were conducted in two small towns in southeastern Brazil between May and June 2020. In the population-based studies, 400 people were evaluated in each town; there were 40 homes in the household survey, and 95 cases and 393 controls in the case-control study. SARS-CoV-2 serology testing was performed on participants, and a questionnaire was applied. Prevalence, household secondary infection rate, and factors associated with infection were assessed. Odds ratios (ORs) were calculated by logistic regression. Logistics worker was defined as an individual with an occupation focused on the transportation of people or goods and whose job involves traveling outside the town of residence at least once a week. RESULTS: Higher seroprevalence of SARS-CoV-2 was observed in the town with a greater proportion of logistics workers. The secondary household infection rate was 49.1% (55/112), and it was observed that in most households (28/40, 70%) the index case was a logistics worker. The case-control study revealed that being a logistics worker (OR 18.0, 95% CI 8.4-38.7) or living with one (OR 6.9, 95% CI 3.3-14.5) increases the risk of infection. In addition, having close contact with a confirmed case (OR 13.4, 95% CI 6.6-27.3) and living with more than four people (OR 2.7, 95% CI 1.1-7.1) were also risk factors. CONCLUSIONS: Our study shows a strong association between logistics workers and the risk of SARS-CoV-2 infection and highlights the key role of these workers in the viral spread in small towns. These findings indicate the need to focus on this population to determine COVID-19 prevention and control strategies, including vaccination and sentinel genomic surveillance.

Integrating Patient Data Into Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review.

BACKGROUND: Recent years have been witnessing a substantial improvement in the accuracy of skin cancer classification using convolutional neural networks (CNNs). CNNs perform on par with or better than dermatologists with respect to the classification tasks of single images. However, in clinical practice, dermatologists also use other patient data beyond the visual aspects present in a digitized image, further increasing their diagnostic accuracy. Several pilot studies have recently investigated the effects of integrating different subtypes of patient data into CNN-based skin cancer classifiers. OBJECTIVE: This systematic review focuses on the current research investigating the impact of merging information from image features and patient data on the performance of CNN-based skin cancer image classification. This study aims to explore the potential in this field of research by evaluating the types of patient data used, the ways in which the nonimage data are encoded and merged with the image features, and the impact of the integration on the classifier performance. METHODS: Google Scholar, PubMed, MEDLINE, and ScienceDirect were screened for peer-reviewed studies published in English that dealt with the integration of patient data within a CNN-based skin cancer classification. The search terms skin cancer classification, convolutional neural network(s), deep learning, lesions, melanoma, metadata, clinical information, and patient data were combined. RESULTS: A total of 11 publications fulfilled the inclusion criteria. All of them reported an overall improvement in different skin lesion classification tasks with patient data integration. The most commonly used patient data were age, sex, and lesion location. The patient data were mostly one-hot encoded. There were differences in the complexity that the encoded patient data were processed with regarding deep learning methods before and after fusing them with the image features for a combined classifier. CONCLUSIONS: This study indicates the potential benefits of integrating patient data into CNN-based diagnostic algorithms. However, how exactly the individual patient data enhance classification performance, especially in the case of multiclass classification problems, is still unclear. Moreover, a substantial fraction of patient data used by dermatologists remains to be analyzed in the context of CNN-based skin cancer classification. Further exploratory analyses in this promising field may optimize patient data integration into CNN-based skin cancer diagnostics for patients' benefits.

Hidden Variables in Deep Learning Digital Pathology and Their Potential to Cause Batch Effects: Prediction Model Study.

BACKGROUND: An increasing number of studies within digital pathology show the potential of artificial intelligence (AI) to diagnose cancer using histological whole slide images, which requires large and diverse data sets. While diversification may result in more generalizable AI-based systems, it can also introduce hidden variables. If neural networks are able to distinguish/learn hidden variables, these variables can introduce batch effects that compromise the accuracy of classification systems. OBJECTIVE: The objective of the study was to analyze the learnability of an exemplary selection of hidden variables (patient age, slide preparation date, slide origin, and scanner type) that are commonly found in whole slide image data sets in digital pathology and could create batch effects. METHODS: We trained four separate convolutional neural networks (CNNs) to learn four variables using a data set of digitized whole slide melanoma images from five different institutes. For robustness, each CNN training and evaluation run was repeated multiple times, and a variable was only considered learnable if the lower bound of the 95% confidence interval of its mean balanced accuracy was above 50.0%. RESULTS: A mean balanced accuracy above 50.0% was achieved for all four tasks, even when considering the lower bound of the 95% confidence interval. Performance between tasks showed wide variation, ranging from 56.1% (slide preparation date) to 100% (slide origin). CONCLUSIONS: Because all of the analyzed hidden variables are learnable, they have the potential to create batch effects in dermatopathology data sets, which negatively affect AI-based classification systems. Practitioners should be aware of these and similar pitfalls when developing and evaluating such systems and address these and potentially other batch effect variables in their data sets through sufficient data set stratification.

Deep learning in cancer pathology: a new generation of clinical biomarkers.

Clinical workflows in oncology rely on predictive and prognostic molecular biomarkers. However, the growing number of these complex biomarkers tends to increase the cost and time for decision-making in routine daily oncology practice; furthermore, biomarkers often require tumour tissue on top of routine diagnostic material. Nevertheless, routinely available tumour tissue contains an abundance of clinically relevant information that is currently not fully exploited. Advances in deep learning (DL), an artificial intelligence (AI) technology, have enabled the extraction of previously hidden information directly from routine histology images of cancer, providing potentially clinically useful information. Here, we outline emerging concepts of how DL can extract biomarkers directly from histology images and summarise studies of basic and advanced image analysis for cancer histology. Basic image analysis tasks include detection, grading and subtyping of tumour tissue in histology images; they are aimed at automating pathology workflows and consequently do not immediately translate into clinical decisions. Exceeding such basic approaches, DL has also been used for advanced image analysis tasks, which have the potential of directly affecting clinical decision-making processes. These advanced approaches include inference of molecular features, prediction of survival and end-to-end prediction of therapy response. Predictions made by such DL systems could simplify and enrich clinical decision-making, but require rigorous external validation in clinical settings.

Patient Perception of Mobile Phone Apps for the Care and Prevention of Sexually Transmitted Diseases: Cross-Sectional Study.

BACKGROUND: In the emerging era of digitalization and electronic health, various health-related apps have been launched, including apps for sexually transmitted diseases. Until now, little has been known about how patients perceive the value of such apps. OBJECTIVE: To investigate patient's attitudes and awareness toward sexually transmitted disease-related apps in an outpatient sexually transmitted disease clinic setting. METHODS: A cross-sectional study was conducted at a dermatovenereological outpatient unit between April and July 2019. Patients completed a self-administered questionnaire on their perceptions of the popularity and usefulness of sexually transmitted disease-related apps. Descriptive analysis was performed with expression of categorical variables as frequencies and percentages. For continuous variables, the median, range, and interquartile range were indicated. Contingency tables and chi-square tests were used to investigate associations between sociodemographic data and items of the questionnaire. RESULTS: A total of 226 patients were surveyed (heterosexual: 137/193, 71.0%; homosexual: 44/193, 22.8%; bisexual: 12/193, 6.2%); 11.9% (27/225) had previously used health-related apps. Nearly half of the patients (97/214, 45.3%) specifically considered sexually transmitted disease-related apps useful, 47.8% (100/209) voted that they could supplement or support the consultation of a physician. Interestingly, only 35.1% (74/211) preferred a printed patient brochure on sexually transmitted diseases over downloading and using an app, but 64.0% (134/209) would download a sexually transmitted disease-related app recommended by their physician. General information regarding sexually transmitted diseases (93/167, 55.7%), evaluation of skin diseases based on photos or videos (78/167, 53.3%), information on the prevention of sexually transmitted diseases (76/167, 45.5%), mediation of nearby contact points or test sites (74/167, 44.3%), anonymous medical advice (69/167, 41.3%), and calculation of the risk of having a sexually transmitted disease (63/167, 37.3%) were rated as the most important features. Men were more likely than women to find sexually transmitted disease-related apps useful in general (P=.04; χ2=6.28) and to pay for such apps (P=.01; χ2=9.19). Patients aged <40 years would rather download an app recommended by their physician (P=.03; χ2=7.23), whereas patients aged >40 years preferred reading a patient brochure on sexually transmitted diseases (P=.02; χ2=8.14). CONCLUSIONS: This study demonstrated high general interest in the use of sexually transmitted disease-related apps in this sample of dermatovenereological outpatients. In particular, young age and male sex were significantly associated with a positive perception, underlining the high potential of apps in the prevention and early recognition of sexually transmitted diseases in this group. Future studies are warranted to validate these findings in other populations.

Patient Attitudes and Their Awareness Towards Skin Cancer-Related Apps: Cross-Sectional Survey.

BACKGROUND: In the emerging era of digitalization and electronic health, skin cancer-related apps represent useful tools to support dermatologic consultation and examination. Yet, little is known about how patients perceive the value of such apps. OBJECTIVE: The aim of this study was to investigate patient attitudes and their awareness toward skin cancer-related apps. METHODS: A cross-sectional study including 200 patients from the oncological outpatient unit was conducted at the University Hospital (LMU Munich, Germany) between September and December 2018. Patients were asked to complete a self-administered questionnaire on the popularity and usefulness of health-related and skin cancer-related apps. A descriptive analysis was performed with the expression of categorical variables as frequencies and percentages. For continuous variables, the median and range were indicated. Contingency tables and chi-square tests were performed to investigate associations between sociodemographic data and selected items of the questionnaire. RESULTS: A total of 98.9% (195/197) of patients had never used skin cancer-related apps or could not remember. In 49.7% (93/187) of cases, patients were unsure about the usefulness of skin cancer apps, whereas 42.6% (78/183) thought that skin cancer apps could supplement or support the professional skin examination performed by a physician. However, 47.9% (90/188) were interested in acquiring more information by their dermatologists about skin cancer apps. Young age (P=.002), male gender (P=.02), a previous history of melanoma (P=.004), and higher educational level (P=.002) were significantly associated with a positive attitude. Nevertheless, 55.9% (105/188) preferred a printed patient brochure on skin cancer to downloading and using an app. CONCLUSIONS: The experience and knowledge of skin cancer-related apps was surprisingly low in this population, although there was a high general interest in more information about such apps. Printed patient brochures were the preferred information source.

EAT-Brazil Award for Tobacco Control: a brief description of its first edition.

Smoking is a major global risk factor for preventable death and disability. EAT is an acronym for Education Against Tobacco, a multinational network of physicians and medical students that aims to improve tobacco control by means of school-based prevention targeted at adolescents through counseling, use of software and support materials. The first EAT-Brazil Award, launched in March 2018, was a competition designed to encourage the proposal of objective solutions for tobacco control in Brasil, and identify new talents in the area. Brazilian undergraduate students from any field of study could submit a one-page essay on the subject, competing for the amount of R$ 1000.00 (one thousand reais). There were a total of 39 applicants (20 women and 19 men) from 9 Brazilian states and 18 undergraduate programs, with a mean age of 22.5 years (SD = 3.7). Data from an online anonymous questionnaire answered after the submission of their essays revealed that most applicants were students of institutions from in the state of Minas Gerais (n = 26/39; 66.6%), studied medicine (n = 20/39, 51.3%), and had no prior knowledge of the EAT-Brazil Network (n = 27/39, 69.2%). The winner of the award was Lucas Guimarães de Azevedo, a fourth-year medical student at Federal University of Western Bahia. The next editions of the award should focus on increasing the number of applicants and diversifying their geographical distribution.

EAT-Brazil Award for Tobacco Control: a brief description of its first edition

SUMMARY Smoking is a major global risk factor for preventable death and disability. EAT is an acronym for Education Against Tobacco, a multinational network of physicians and medical students that aims to improve tobacco control by means of school-based prevention targeted at adolescents through counseling, use of software and support materials. The first EAT-Brazil Award, launched in March 2018, was a competition designed to encourage the proposal of objective solutions for tobacco control in Brasil, and identify new talents in the area. Brazilian undergraduate students from any field of study could submit a one-page essay on the subject, competing for the amount of R$ 1000.00 (one thousand reais). There were a total of 39 applicants (20 women and 19 men) from 9 Brazilian states and 18 undergraduate programs, with a mean age of 22.5 years (SD = 3.7). Data from an online anonymous questionnaire answered after the submission of their essays revealed that most applicants were students of institutions from in the state of Minas Gerais (n = 26/39; 66.6%), studied medicine (n = 20/39, 51.3%), and had no prior knowledge of the EAT-Brazil Network (n = 27/39, 69.2%). The winner of the award was Lucas Guimarães de Azevedo, a fourth-year medical student at Federal University of Western Bahia. The next editions of the award should focus on increasing the number of applicants and diversifying their geographical distribution.

RESUMO O tabagismo é um dos principais fatores de risco globais para morte e incapacidade evitáveis. EAT é a sigla em inglês para Educação contra o Tabaco (Education Against Tobacco), uma rede mundial formada por médicos e estudantes de medicina cuja missão é atuar no combate ao tabagismo por meio da prevenção da iniciação ao tabagismo em adolescentes escolares mediante aconselhamento, uso de aplicativos móveis e de materiais de apoio. O primeiro Prêmio EAT-Brazil, lançado em março de 2018, foi um concurso destinado a encorajar a proposição de soluções objetivas para o avanço do controle do tabagismo no país e a identificação de novos talentos para a área. Estudantes de graduação brasileiros de qualquer curso submeteram um texto de uma página sobre o tema, concorrendo à quantia de R$ 1.000. Houve um total de 39 trabalhos inscritos (20 por mulheres e 19 por homens) de nove estados brasileiros e 18 cursos de graduação, com idade média de 22,5 anos (DP=3,7). Dados de um questionário anônimo on-line respondido pelos inscritos revelou que a maioria era composta por graduandos de alguma instituição do estado de Minas Gerais (n=26/39; 66,6%), que estudavam medicina (n=20/39; 51,3%) e não tinham conhecimento prévio sobre a Rede EAT-Brazil (n=27/39; 69,2%). O ganhador do prêmio foi Lucas Guimarães de Azevedo, aluno do oitavo período de medicina da Universidade Federal do Oeste da Bahia. As próximas edições do Prêmio devem focar o aumento do número de inscritos e a diversificação de sua distribuição geográfica.

Pathologist-level classification of histopathological melanoma images with deep neural networks.

BACKGROUND: The diagnosis of most cancers is made by a board-certified pathologist based on a tissue biopsy under the microscope. Recent research reveals a high discordance between individual pathologists. For melanoma, the literature reports 25-26% of discordance for classifying a benign nevus versus malignant melanoma. Deep learning was successfully implemented to enhance the precision of lung and breast cancer diagnoses. The aim of this study is to illustrate the potential of deep learning to assist human assessment for a histopathologic melanoma diagnosis. METHODS: Six hundred ninety-five lesions were classified by an expert histopathologist in accordance with current guidelines (350 nevi and 345 melanomas). Only the haematoxylin and eosin stained (H&E) slides of these lesions were digitalised using a slide scanner and then randomly cropped. Five hundred ninety-five of the resulting images were used for the training of a convolutional neural network (CNN). The additional 100 H&E image sections were used to test the results of the CNN in comparison with the original class labels. FINDINGS: The total discordance with the histopathologist was 18% for melanoma (95% confidence interval [CI]: 7.4-28.6%), 20% for nevi (95% CI: 8.9-31.1%) and 19% for the full set of images (95% CI: 11.3-26.7%). INTERPRETATION: Even in the worst case, the discordance of the CNN was about the same compared with the discordance between human pathologists as reported in the literature. Despite the vastly reduced amount of data, time necessary for diagnosis and cost compared with the pathologist, our CNN archived on-par performance. Conclusively, CNNs indicate to be a valuable tool to assist human melanoma diagnoses.

Process Evaluation of a Medical Student-Delivered Smoking Prevention Program for Secondary Schools: Protocol for the Education Against Tobacco Cluster Randomized Trial.

BACKGROUND: Most smokers start smoking during their early adolescence under the impression that smoking entails positive attributes. Given the addictive nature of cigarettes, however, many of them might end up as long-term smokers and suffering from tobacco-related diseases. To prevent tobacco use among adolescents, the large international medical students' network Education Against Tobacco (EAT) educates more than 40,000 secondary school students per year in the classroom setting, using evidence-based self-developed apps and strategies. OBJECTIVE: This study aimed to evaluate the long-term effectiveness of the school-based EAT intervention in reducing smoking prevalence among seventh-grade students in Germany. Additionally, we aimed to improve the intervention by drawing conclusions from our process evaluation. METHODS: We conduct a cluster-randomized controlled trial with measurements at baseline and 9, 16, and 24 months postintervention via paper-and-pencil questionnaires administered by teachers. The study groups consist of randomized schools receiving the 2016 EAT curriculum and control schools with comparable baseline data (no intervention). The primary outcome is the difference of change in smoking prevalence between the intervention and control groups at the 24-month follow-up. Secondary outcomes are between-group differences of changes in smoking-related attitudes and the number of new smokers, quitters, and never-smokers. RESULTS: A total of 11,268 students of both sexes, with an average age of 12.32 years, in seventh grade of 144 secondary schools in Germany were included at baseline. The prevalence of cigarette smoking in our sample was 2.6%. The process evaluation surveys were filled out by 324 medical student volunteers, 63 medical student supervisors, 4896 students, and 141 teachers. CONCLUSIONS: The EAT cluster randomized trial is the largest school-based tobacco-prevention study in Germany conducted to date. Its results will provide important insights with regards to the effectiveness of medical student-delivered smoking prevention programs at school. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/13508.

A Smoking Prevention Program Delivered by Medical Students to Secondary Schools in Brazil Called "Education Against Tobacco": Randomized Controlled Trial.

BACKGROUND: Smoking is the largest preventable cause of mortality in Brazil. Education Against Tobacco (EAT) is a network of more than 3500 medical students and physicians across 14 countries who volunteer for school-based smoking prevention programs. EAT educates 50,000 adolescents per year in the classroom setting. A recent quasi-experimental study conducted in Germany showed that EAT had significant short-term smoking cessation effects among adolescents aged 11 to 15 years. OBJECTIVE: The aim is to measure the long-term effectiveness of the most recent version of the EAT curriculum in Brazil. METHODS: A randomized controlled trial was conducted among 2348 adolescents aged 12 to 21 years (grades 7-11) at public secondary schools in Brazil. The prospective experimental design included measurements at baseline and at 6 and 12 months postintervention. The study groups comprised randomized classes receiving the standardized EAT intervention (90 minutes of mentoring in a classroom setting) and control classes in the same schools (no intervention). Data were collected on smoking status, gender, social aspects, and predictors of smoking. The primary endpoint was the difference in the change in smoking prevalence between the intervention group and the control group at 12-month follow-up. RESULTS: From baseline to 12 months, the smoking prevalence increased from 11.0% to 20.9% in the control group and from 14.1% to 15.6% in the intervention group. This difference was statistically significant (P<.01). The effects were smaller for females (control 12.4% to 18.8% vs intervention 13.1% to 14.6%) than for males (control 9.1% to 23.6% vs intervention 15.3% to 16.8%). Increased quitting rates and prevented onset were responsible for the intervention effects. The differences in change in smoking prevalence from baseline to 12 months between the intervention and control groups were increased in students with low school performance. CONCLUSIONS: To our knowledge, this is the first randomized trial on school-based tobacco prevention in Brazil that shows significant long-term favorable effects. The EAT program encourages quitting and prevents smoking onset, especially among males and students with low educational background. TRIAL REGISTRATION: ClinicalTrials.gov NCT02725021; https://clinicaltrials.gov/ct2/show/NCT02725021. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/resprot.7134.

[The global medical network Education Against Tobacco-voluntary tobacco prevention made in Germany]. / Das weltweite Medizinernetzwerk Aufklärung gegen Tabak ­ Ehrenamtliche Prävention made in Germany.

Smoking is the leading preventable cause of premature death in Germany. The network "Education Against Tobacco" (EAT) is an initiative that was founded in Germany in 2012, in which more than 3500 medical students and physicians engage in volunteer work in about 80 medical faculties in 14 countries. In this article, the concept, activities, objectives and associated research studies oft he EAT initiative are introduced.On the school level, the initiative addresses 10- to 15-year-old secondary school students. In addition to a multimodal approach, school visits use modern media such as facemorphing apps, which are not only used by students (45,000 per year in 14 countries), but by a total of over 500,000 other people as well. The effectiveness of the school-based intervention is currently being investigated in randomised long-term studies with 20,000 adolescents in Germany. A first long-term study demonstrated evidence of a protective effect regarding the onset of smoking, especially among female students, students having a low level of education and students with a migratory background.The programme educates several hundred prospective physicians at 13 (of 28 participating) German medical faculties each year in science-based elective courses for the well-established smoking cessation counselling of patients and sensitises them to the tobacco epidemic. The approved members engage in dialogue with local members of the German house of representatives as "Ärzteverband Tabakprävention".EAT motivates the prospective generation of physicians, initially through prevention in school settings, to face the challenge of national tobacco control at the university and federal level.

Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review.

BACKGROUND: State-of-the-art classifiers based on convolutional neural networks (CNNs) were shown to classify images of skin cancer on par with dermatologists and could enable lifesaving and fast diagnoses, even outside the hospital via installation of apps on mobile devices. To our knowledge, at present there is no review of the current work in this research area. OBJECTIVE: This study presents the first systematic review of the state-of-the-art research on classifying skin lesions with CNNs. We limit our review to skin lesion classifiers. In particular, methods that apply a CNN only for segmentation or for the classification of dermoscopic patterns are not considered here. Furthermore, this study discusses why the comparability of the presented procedures is very difficult and which challenges must be addressed in the future. METHODS: We searched the Google Scholar, PubMed, Medline, ScienceDirect, and Web of Science databases for systematic reviews and original research articles published in English. Only papers that reported sufficient scientific proceedings are included in this review. RESULTS: We found 13 papers that classified skin lesions using CNNs. In principle, classification methods can be differentiated according to three principles. Approaches that use a CNN already trained by means of another large dataset and then optimize its parameters to the classification of skin lesions are the most common ones used and they display the best performance with the currently available limited datasets. CONCLUSIONS: CNNs display a high performance as state-of-the-art skin lesion classifiers. Unfortunately, it is difficult to compare different classification methods because some approaches use nonpublic datasets for training and/or testing, thereby making reproducibility difficult. Future publications should use publicly available benchmarks and fully disclose methods used for training to allow comparability.

Teledermatology: Comparison of Store-and-Forward Versus Live Interactive Video Conferencing.

A decreasing number of dermatologists and an increasing number of patients in Western countries have led to a relative lack of clinicians providing expert dermatologic care. This, in turn, has prolonged wait times for patients to be examined, putting them at risk. Store-and-forward teledermatology improves patient access to dermatologists through asynchronous consultations, reducing wait times to obtain a consultation. However, live video conferencing as a synchronous service is also frequently used by practitioners because it allows immediate interaction between patient and physician. This raises the question of which of the two approaches is superior in terms of quality of care and convenience. There are pros and cons for each in terms of technical requirements and features. This viewpoint compares the two techniques based on a literature review and a clinical perspective to help dermatologists assess the value of teledermatology and determine which techniques would be valuable in their practice.

A Face-Aging App for Smoking Cessation in a Waiting Room Setting: Pilot Study in an HIV Outpatient Clinic.

BACKGROUND: There is strong evidence for the effectiveness of addressing tobacco use in health care settings. However, few smokers receive cessation advice when visiting a hospital. Implementing smoking cessation technology in outpatient waiting rooms could be an effective strategy for change, with the potential to expose almost all patients visiting a health care provider without preluding physician action needed. OBJECTIVE: The objective of this study was to develop an intervention for smoking cessation that would make use of the time patients spend in a waiting room by passively exposing them to a face-aging, public morphing, tablet-based app, to pilot the intervention in a waiting room of an HIV outpatient clinic, and to measure the perceptions of this intervention among smoking and nonsmoking HIV patients. METHODS: We developed a kiosk version of our 3-dimensional face-aging app Smokerface, which shows the user how their face would look with or without cigarette smoking 1 to 15 years in the future. We placed a tablet with the app running on a table in the middle of the waiting room of our HIV outpatient clinic, connected to a large monitor attached to the opposite wall. A researcher noted all the patients who were using the waiting room. If a patient did not initiate app use within 30 seconds of waiting time, the researcher encouraged him or her to do so. Those using the app were asked to complete a questionnaire. RESULTS: During a 19-day period, 464 patients visited the waiting room, of whom 187 (40.3%) tried the app and 179 (38.6%) completed the questionnaire. Of those who completed the questionnaire, 139 of 176 (79.0%) were men and 84 of 179 (46.9%) were smokers. Of the smokers, 55 of 81 (68%) said the intervention motivated them to quit (men: 45, 68%; women: 10, 67%); 41 (51%) said that it motivated them to discuss quitting with their doctor (men: 32, 49%; women: 9, 60%); and 72 (91%) perceived the intervention as fun (men: 57, 90%; women: 15, 94%). Of the nonsmokers, 92 (98%) said that it motivated them never to take up smoking (men: 72, 99%; women: 20, 95%). Among all patients, 102 (22.0%) watched another patient try the app without trying it themselves; thus, a total of 289 (62.3%) of the 464 patients were exposed to the intervention (average waiting time 21 minutes). CONCLUSIONS: A face-aging app implemented in a waiting room provides a novel opportunity to motivate patients visiting a health care provider to quit smoking, to address quitting at their subsequent appointment and thereby encourage physician-delivered smoking cessation, or not to take up smoking.

Implementation of cell-free tumor DNA sequencing from the cerebrospinal fluid to guide treatment in a patient with primary leptomeningeal melanoma: A case report.

Primary leptomeningeal melanoma (PLM) is a rare type of cancer that represents a major clinical and molecular diagnostic challenge. A definitive diagnosis requires consistent magnetic resonance imaging findings and cerebrospinal fluid (CSF) cytology. Due to the small number of malignant cells in the CSF, routine testing for mutations in the BRAF gene is difficult, which prevents the stratification of these patients to potentially beneficial therapies. We herein present the case of a 62-year old man with CSF cytology indicating PLM, where BRAF mutation testing, from cell-free (cf) tumor DNA isolated from the CSF and plasma was implemented to guide clinical decision making. Testing for BRAFV600E mutation from the CSF and plasma was technically feasible, yielded concordant results, and guided the treatment for this patient. This case suggests that mutation testing of cfDNA isolated from the CSF is technically feasible and may guide therapy in cases where a tissue diagnosis is not possible for PLM and other malignancies with defined oncogenic driver mutations.

Facial-Aging Mobile Apps for Smoking Prevention in Secondary Schools in Brazil: Appearance-Focused Interventional Study.

BACKGROUND: Most smokers start smoking during their early adolescence, often with the idea that smoking is glamorous. Interventions that harness the broad availability of mobile phones as well as adolescents' interest in their appearance may be a novel way to improve school-based prevention. A recent study conducted in Germany showed promising results. However, the transfer to other cultural contexts, effects on different genders, and implementability remains unknown. OBJECTIVE: In this observational study, we aimed to test the perception and implementability of facial-aging apps to prevent smoking in secondary schools in Brazil in accordance with the theory of planned behavior and with respect to different genders. METHODS: We used a free facial-aging mobile phone app ("Smokerface") in three Brazilian secondary schools via a novel method called mirroring. The students' altered three-dimensional selfies on mobile phones or tablets and images were "mirrored" via a projector in front of their whole grade. Using an anonymous questionnaire, we then measured on a 5-point Likert scale the perceptions of the intervention among 306 Brazilian secondary school students of both genders in the seventh grade (average age 12.97 years). A second questionnaire captured perceptions of medical students who conducted the intervention and its conduction per protocol. RESULTS: The majority of students perceived the intervention as fun (304/306, 99.3%), claimed the intervention motivated them not to smoke (289/306, 94.4%), and stated that they learned new benefits of not smoking (300/306, 98.0%). Only a minority of students disagreed or fully disagreed that they learned new benefits of nonsmoking (4/306, 1.3%) or that they themselves were motivated not to smoke (5/306, 1.6%). All of the protocol was delivered by volunteer medical students. CONCLUSIONS: Our data indicate the potential for facial-aging interventions to reduce smoking prevalence in Brazilian secondary schools in accordance with the theory of planned behavior. Volunteer medical students enjoyed the intervention and are capable of complete implementation per protocol.

Patient-Centered Mobile Health Data Management Solution for the German Health Care System (The DataBox Project).

This article describes the DataBox project which offers a perspective of a new health data management solution in Germany. DataBox was initially conceptualized as a repository of individual lung cancer patient data (structured and unstructured). The patient is the owner of the data and is able to share his or her data with different stakeholders. Data is transferred, displayed, and stored online, but not archived. In the long run, the project aims at replacing the conventional method of paper- and storage-device-based handling of data for all patients in Germany, leading to better organization and availability of data which reduces duplicate diagnostic procedures, treatment errors, and enables the training as well as usage of artificial intelligence algorithms on large datasets.