@article{SchneiderWeber, author = {Schneider, Diana and Weber, Karsten}, title = {AI for decision support: What are possible futures, social impacts, regulatory options, ethical conundrums and agency constellations?}, series = {TATuP - Zeitschrift f{\"u}r Technikfolgenabsch{\"a}tzung in Theorie und Praxis}, volume = {33}, journal = {TATuP - Zeitschrift f{\"u}r Technikfolgenabsch{\"a}tzung in Theorie und Praxis}, number = {1}, publisher = {oekom verlag}, address = {M{\"u}nchen}, issn = {2567-8833}, doi = {10.14512/tatup.33.1.08}, pages = {8 -- 13}, abstract = {Although artificial intelligence (AI) and automated decision-making systems have been around for some time, they have only recently gained in importance as they are now actually being used and are no longer just the subject of research. AI to support decision-making is thus affecting ever larger parts of society, creating technical, but above all ethical, legal, and societal challenges, as decisions can now be made by machines that were previously the responsibility of humans. This introduction provides an overview of attempts to regulate AI and addresses key challenges that arise when integrating AI systems into human decision-making. The Special topic brings together research articles that present societal challenges, ethical issues, stakeholders, and possible futures of AI use for decision support in healthcare, the legal system, and border control.}, language = {en} } @article{SchneiderWeber, author = {Schneider, Diana and Weber, Karsten}, title = {AI‑based decision support systems and society: An opening statement}, series = {TATuP - Zeitschrift f{\"u}r Technikfolgenabsch{\"a}tzung in Theorie und Praxis}, volume = {33}, journal = {TATuP - Zeitschrift f{\"u}r Technikfolgenabsch{\"a}tzung in Theorie und Praxis}, number = {1}, publisher = {Oekom}, doi = {10.14512/tatup.33.1.9}, pages = {9 -- 13}, abstract = {Although artificial intelligence (AI) and automated decision-making systems have been around for some time, they have only recently gained in importance as they are now actually being used and are no longer just the subject of research. AI to support decision-making is thus affecting ever larger parts of society, creating technical, but above all ethical, legal, and societal challenges, as decisions can now be made by machines that were previously the responsibility of humans. This introduction provides an overview of attempts to regulate AI and addresses key challenges that arise when integrating AI systems into human decision-making. The Special topic brings together research articles that present societal challenges, ethical issues, stakeholders, and possible futures of AI use for decision support in healthcare, the legal system, and border control.}, language = {en} } @inproceedings{SeebauerJahnMottok, author = {Seebauer, Stefan and Jahn, Sabrina and Mottok, J{\"u}rgen}, title = {Learning from Escape Rooms? A Study Design Concept Measuring the Effect of a Cryptography Educational Escape Room}, series = {2020 IEEE Global Engineering Education Conference (EDUCON): 27-30 April 2020, Porto, Portugal}, booktitle = {2020 IEEE Global Engineering Education Conference (EDUCON): 27-30 April 2020, Porto, Portugal}, doi = {10.1109/EDUCON45650.2020.9125333}, pages = {1684 -- 1685}, abstract = {Experts predict that more IT-Security specialists will be needed in the coming years, but in current higher education in engineering disciplines, this topic is hardly addressed. Newer learning methods such as game-based learning (GBL) are enjoying an increasing popularity as their improvement in the education of subject specific topics can be proven by a variety of studies. We chose Educational Escape Rooms (EduER) as a GBL-tool to impart IT-Security in higher education of engineers. In our Escape Room (ER), the students try to solve puzzles and riddles with learned knowledge on the emphasis of cryptography. This paper first deals with a brief introduction to GBL and EduERs, followed by the design of our ER concept, containing different tasks with focus on the topic of cryptography. The tasks cover different cryptographic methods and hash algorithms, e.g. AES, RSA, SHA3. Afterwards the experimental study is presented. The study of the EduER was carried out with students from the bachelor's program in Electrical Engineering and Information Technology at the OTH Regensburg. The participants were divided into three groups of 5 to 8 persons each. They received a briefing with important information, followed by the ER execution, a debriefing afterwards and an exam-like evaluation sheet to test their learned knowledge. Finally, first basic results are presented.}, language = {en} } @article{Seifert, author = {Seifert, Ruth}, title = {The "Human Rights Profession" and War: Humanitarian Help and Social Work in Armed Conflicts}, series = {Journal of Human Rights and Social Work}, volume = {9}, journal = {Journal of Human Rights and Social Work}, number = {3}, publisher = {Springer Nature}, doi = {10.1007/s41134-024-00349-5}, pages = {360 -- 371}, abstract = {Whereas, in the past, social work publications dealing with armed conflict were of interest to a relatively small community; beginning with the Ukraine conflict, this issue has moved into the center of social work discourses. This has raised issues concerning the positioning of social work in political and armed conflict again. Looking at the literature on social work and armed conflicts, it is generally assumed that the definition of the human rights profession also holds for social work in situations of violent collective conflicts. This, however, raises questions. This paper will argue that antinomies emerge for social work resulting from the definition as a human rights profession and the realities of humanitarian work in armed conflicts which have yet to be dealt with.}, language = {en} } @article{SellmerStanglBeyeretal., author = {Sellmer, Andreas and Stangl, Hubert and Beyer, Mandy and Gr{\"u}nstein, Elisabeth and Leonhardt, Michel and Pongratz, Herwig and Eichhorn, Emerich and Elz, Sigurd and Striegl, Birgit and Jenei-Lanzl, Zsuzsa and Dove, Stefan and Straub, Rainer H. and Kr{\"a}mer, Oliver H. and Mahboobi, Siavosh}, title = {Marbostat-100 Defines a New Class of Potent and Selective Antiinflammatory and Antirheumatic Histone Deacetylase 6 Inhibitors}, series = {Journal of medicinal chemistry}, volume = {61}, journal = {Journal of medicinal chemistry}, number = {8}, publisher = {ACS Publications}, doi = {10.1021/acs.jmedchem.7b01593}, pages = {3454 -- 3477}, abstract = {Epigenetic modifiers of the histone deacetylase (HDAC) family contribute to autoimmunity, cancer, HIV infection, inflammation, and neurodegeneration. Hence, histone deacetylase inhibitors (HDACi), which alter protein acetylation, gene expression patterns, and cell fate decisions, represent promising new drugs for the therapy of these diseases. Whereas pan-HDACi inhibit all 11 Zn2+-dependent histone deacetylases (HDACs) and cause a broad spectrum of side effects, specific inhibitors of histone deacetylase 6 (HDAC6i) are supposed to have less side effects. We present the synthesis and biological evaluation of Marbostats, novel HDAC6i that contain the hydroxamic acid moiety linked to tetrahydro-β-carboline derivatives. Our lead compound Marbostat-100 is a more potent and more selective HDAC6i than previously established well-characterized compounds in vitro as well as in cells. Moreover, Marbostat-100 is well tolerated by mice and effective against collagen type II induced arthritis. Thus, Marbostat-100 represents a most selective known HDAC6i and the possibility for clinical evaluation of a HDAC isoform-specific drug.}, language = {en} } @article{SinghRussTerzidis, author = {Singh, Max Diamond and Russ, Jochen and Terzidis, Orestis}, title = {The Impact of the ObamaCare Excise Tax on Innovation and Entrepreneurship - Early Empirical Findings}, series = {International Journal of Innovation Science}, volume = {7}, journal = {International Journal of Innovation Science}, number = {2}, publisher = {Emerald}, doi = {10.1260/1757-2223.7.2.75}, pages = {75 -- 90}, abstract = {This study addresses aspects of governmental influence on innovation by analyzing the impact of the ObamaCare excise tax on the medical device industry. We initially give an overview of common approaches to measuring innovativeness and entrepreneurship, empirically assess whether existing metrics are suitable for investigating the innovation performance of the U.S. medical device industry, and define a new measure (firm innovation activity) for entrepreneurship. Then we perform a quantitative analysis to explore the impact of the tax. We analyze more than 60,000 product clearances from 1996 to 2013, using the FDA database. We find a significant relationship between product counts and revenues for one segment. Contrary to the present criticism of the excise tax, we find hardly any noteworthy response in either firm innovation activity or number of products launched in the year after the tax was introduced. The 2013 reduction of new product submissions is well within the limits of typical annual fluctuations observed in previous years. This provides a first indication that the excise tax act did not have a strong impact on innovative activities through the present.}, language = {en} } @article{SinghTerzidis, author = {Singh, Max Diamond and Terzidis, O.}, title = {Introducing Innovation Phase Transition}, series = {International Journal of Innovation Science}, volume = {7}, journal = {International Journal of Innovation Science}, number = {4}, publisher = {Emerald}, issn = {1757-2223}, doi = {10.1108/IJIS-07-04-2015-B003}, pages = {249 -- 262}, abstract = {Innovation diffusion points toward how innovations spread into the market after launch. This paper investigates diffusion dynamics at market entry time and proposes a new evolution pattern at the intersection between inventions and innovations. With this in mind, we initially prove that patent filings correlate with new product introductions in the U.S. spine market. Then we test our new theory supposing that certain patent filing threshold numbers accelerate strong economic returns in terms of innovations. We find that firms hitting certain patent filing thresholds significantly increase their product launches in the mentioned market. Moreover, the results seem to indicate that economic returns of inventions may be measured substantially. Thus, this paper suggests a new research area by utilizing our proposed concept about an Innovation Outcome Trigger Value (IOTV). Furthermore, the implications may also be interesting for practitioners, since we empirically prove that inventive activities turn out to be worthwhile, indeed.}, language = {en} } @inproceedings{SkeltonWestner, author = {Skelton, David and Westner, Markus}, title = {International internships}, series = {Refereed Proceedings of the 3rd International Research Symposium on Cooperative and Work-Integrated Education, Stuttgart, Germany : 3rd - 27th of June, 2018, Stuttgart, Baden-Wurttemberg, Germany}, booktitle = {Refereed Proceedings of the 3rd International Research Symposium on Cooperative and Work-Integrated Education, Stuttgart, Germany : 3rd - 27th of June, 2018, Stuttgart, Baden-Wurttemberg, Germany}, editor = {Zegwaard, Karsten E. and Ford, Marty}, isbn = {978-0-473-46291-8}, pages = {193 -- 198}, abstract = {Students are increasingly considering expanding their cooperative experiences into the global environment. This paper investigates the experiences of students undertaking their internship capstone in other countries under the supervision of a partner institute and a participating industry sponsor. These experiences of German students interning in New Zealand (NZ), and NZ students interning in Germany offer an insight into the organisation required by the partner institutions and the comparative cultural and technical differences that these students have encountered. The international agreement between Ostbayerische Technische Hochschule (OTH) Regensburg, Germany, and the Eastern Institute of Technology (EIT), NZ, has already built a mutual framework of visiting lecturers teaching in each other's institutes, and a programme of hosting exchange students in each other's institutes. A natural progression of this partnership has now moved into the active recruitment and brokering of international internship opportunities for their respective final year Business and Information Technology (IT) students. The benefits are evident personally to the students, which are discussed in this paper, but there are also clear benefits for EIT and OTH in terms of marketing opportunities for prospective students and enriching the learning environments. This paper explores the challenges faced by the Schools of Business and Computing at EIT as it pioneers the first two years of establishing internship options for final year German exchange students within NZ, and matching final year EIT degree students to industry sponsors within Bavaria, a federal state of Germany. A qualitative approach was taken for this study with data acquired from internship students involved, the key organising lecturers from each institute, and industry sponsors from each country. A number of student, staff and industry representative interviews were conducted along with a cultural appraisal of relevant literature in the global education field. A further overview of the macro benefits for the partner institutes in terms of marketing, and programme enrichment is also part of the conclusion. The paper may be useful as a template for other institutes on developing international internship opportunities through global partners.}, subject = {Wirtschaftswissenschaftliches Studium}, language = {en} } @article{SommoggyGrepmeierApfelbacheretal., author = {Sommoggy, Julia and Grepmeier, Eva-Maria and Apfelbacher, Christian and Brandstetter, Susanne and Curbach, Janina}, title = {Pediatricians' experiences of managing outpatient care during the COVID-19 pandemic: A qualitative study in Germany}, series = {Frontiers in Pediatrics}, volume = {11}, journal = {Frontiers in Pediatrics}, publisher = {frontiers}, doi = {10.3389/fped.2023.1127238}, pages = {11}, abstract = {Background: Pediatricians are important sources of information for parents regarding their children's health. During the COVID-19 pandemic, pediatricians faced a variety of challenges regarding information uptake and transfer to patients, practice organization and consultations for families. This qualitative study aimed at shedding light on German pediatricians' experiences of providing outpatient care during the first year of the pandemic. Methods: We conducted 19 semi-structured, in-depth interviews with pediatricians in Germany from July 2020 to February 2021. All interviews were audio recorded, transcribed, pseudonymized, coded, and subjected to content analysis. Results: Pediatricians felt able to keep up to date regarding COVID-19 regulations. However, staying informed was time consuming and onerous. Informing the patients was perceived as strenuous, especially when political decisions had not been officially communicated to pediatricians or if the recommendations were not supported by the professional judgment of the interviewees. Some felt that they were not taken seriously or adequately involved in political decisions. Parents were reported to consider pediatric practices as sources of information also for non-medical inquiries. Answering these questions was time consuming for the practice personnel and involved non-billable hours. Practices had to adapt their set-up and organization immediately to the new circumstances of the pandemic, which proved costly and laborious as well. Some changes in the organization of routine care, such as the separation of appointments for patients with acute infection from preventive appointments, were perceived as positive and effective by some study participants. Telephone and online consultations were established at the beginning of the pandemic and considered helpful for some situations, whereas for others these methods were deemed insufficient (e.g. for examinations of sick children). All pediatricians reported reduced utilization mainly due to a decline in acute infections. However, preventive medical check-ups and immunization appointments were reported to be mostly attended. Conclusion: Positive experiences of reorganizing pediatric practice should be disseminated as "best practices" in order to improve future pediatric health services. Further research could show how some of these positive experiences in reorganizing care during the pandemic are to be maintained by pediatricians in the future.}, language = {en} } @article{SommoggyGrepmeierCurbach, author = {Sommoggy, Julia von and Grepmeier, Eva-Maria and Curbach, Janina}, title = {Health Literacy-Sensitive Counselling on Early Childhood Allergy Prevention}, series = {International Journal of Environmental Research and Public Health}, volume = {19}, journal = {International Journal of Environmental Research and Public Health}, number = {7}, publisher = {MDPI}, address = {Basel}, doi = {10.3390/ijerph19074182}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:355-epub-520757}, pages = {1 -- 16}, abstract = {In Germany, midwives are involved in extensive antenatal and postnatal care. As health professionals, they can play a key role in strengthening health literacy (HL) of parents on how to prevent chronic allergic diseases in their children. The objective of this study is to explore midwives' perspectives regarding HL-sensitive counselling in early childhood allergy prevention (ECAP). Twenty-four qualitative semi-structured interviews were conducted with midwives, and data were analyzed using qualitative content analysis. Only a small number of study participants were aware of HL as a concept. However, most of these use screening and counselling strategies which consider individual information needs and which support parental HL. HL sensitivity in counselling is largely based on the midwives' "gut feelings" and counselling experience, rather than on formal education. The midwives were largely aware of evidence-based ECAP recommendations; however, allergy prevention was not seen as a stand-alone topic but as part of their general counselling on infant feeding and hygiene. They found parents to be more open to receiving complex prevention information during antenatal counselling. In order to strengthen midwives' roles in HL-sensitive ECAP counselling, their formal education should provide them with explicit HL knowledge and counselling skills. ECAP should be an inherent part of antenatal care.}, language = {en} } @inproceedings{SouzaJrHookPapaetal., author = {Souza Jr., Luis Antonio de and Hook, Christian and Papa, Jo{\~a}o Paulo and Palm, Christoph}, title = {Barrett's Esophagus Analysis Using SURF Features}, series = {Bildverarbeitung f{\"u}r die Medizin 2017; Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 12. bis 14. M{\"a}rz 2017 in Heidelberg}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2017; Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 12. bis 14. M{\"a}rz 2017 in Heidelberg}, publisher = {Springer}, address = {Berlin}, doi = {10.1007/978-3-662-54345-0_34}, pages = {141 -- 146}, abstract = {The development of adenocarcinoma in Barrett's esophagus is difficult to detect by endoscopic surveillance of patients with signs of dysplasia. Computer assisted diagnosis of endoscopic images (CAD) could therefore be most helpful in the demarcation and classification of neoplastic lesions. In this study we tested the feasibility of a CAD method based on Speeded up Robust Feature Detection (SURF). A given database containing 100 images from 39 patients served as benchmark for feature based classification models. Half of the images had previously been diagnosed by five clinical experts as being "cancerous", the other half as "non-cancerous". Cancerous image regions had been visibly delineated (masked) by the clinicians. SURF features acquired from full images as well as from masked areas were utilized for the supervised training and testing of an SVM classifier. The predictive accuracy of the developed CAD system is illustrated by sensitivity and specificity values. The results based on full image matching where 0.78 (sensitivity) and 0.82 (specificity) were achieved, while the masked region approach generated results of 0.90 and 0.95, respectively.}, subject = {Speiser{\"o}hrenkrankheit}, language = {en} } @article{SouzaJrMendelStrasseretal., author = {Souza Jr., Luis Antonio de and Mendel, Robert and Strasser, Sophia and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Papa, Jo{\~a}o Paulo and Palm, Christoph}, title = {Convolutional Neural Networks for the evaluation of cancer in Barrett's esophagus: Explainable AI to lighten up the black-box}, series = {Computers in Biology and Medicine}, volume = {135}, journal = {Computers in Biology and Medicine}, publisher = {Elsevier}, issn = {0010-4825}, doi = {10.1016/j.compbiomed.2021.104578}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-20126}, pages = {1 -- 14}, abstract = {Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their level of accountability and transparency must be provided in such evaluations. The reliability related to machine learning predictions must be explained and interpreted, especially if diagnosis support is addressed. For this task, the black-box nature of deep learning techniques must be lightened up to transfer its promising results into clinical practice. Hence, we aim to investigate the use of explainable artificial intelligence techniques to quantitatively highlight discriminative regions during the classification of earlycancerous tissues in Barrett's esophagus-diagnosed patients. Four Convolutional Neural Network models (AlexNet, SqueezeNet, ResNet50, and VGG16) were analyzed using five different interpretation techniques (saliency, guided backpropagation, integrated gradients, input × gradients, and DeepLIFT) to compare their agreement with experts' previous annotations of cancerous tissue. We could show that saliency attributes match best with the manual experts' delineations. Moreover, there is moderate to high correlation between the sensitivity of a model and the human-and-computer agreement. The results also lightened that the higher the model's sensitivity, the stronger the correlation of human and computational segmentation agreement. We observed a relevant relation between computational learning and experts' insights, demonstrating how human knowledge may influence the correct computational learning.}, subject = {Deep Learning}, language = {en} } @article{SouzaJrPachecoPassosetal., author = {Souza Jr., Luis Antonio de and Pacheco, Andr{\´e} G.C. and Passos, Leandro A. and Santana, Marcos Cleison S. and Mendel, Robert and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {DeepCraftFuse: visual and deeply-learnable features work better together for esophageal cancer detection in patients with Barrett's esophagus}, series = {Neural Computing and Applications}, volume = {36}, journal = {Neural Computing and Applications}, publisher = {Springer}, address = {London}, doi = {10.1007/s00521-024-09615-z}, pages = {10445 -- 10459}, abstract = {Limitations in computer-assisted diagnosis include lack of labeled data and inability to model the relation between what experts see and what computers learn. Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their accountability and transparency level must be improved to transfer this success into clinical practice. The reliability of machine learning decisions must be explained and interpreted, especially for supporting the medical diagnosis. While deep learning techniques are broad so that unseen information might help learn patterns of interest, human insights to describe objects of interest help in decision-making. This paper proposes a novel approach, DeepCraftFuse, to address the challenge of combining information provided by deep networks with visual-based features to significantly enhance the correct identification of cancerous tissues in patients affected with Barrett's esophagus (BE). We demonstrate that DeepCraftFuse outperforms state-of-the-art techniques on private and public datasets, reaching results of around 95\% when distinguishing patients affected by BE that is either positive or negative to esophageal cancer.}, subject = {Deep Learning}, language = {en} } @article{SouzaJrPalmMendeletal., author = {Souza Jr., Luis Antonio de and Palm, Christoph and Mendel, Robert and Hook, Christian and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Weber, Silke A. T. and Papa, Jo{\~a}o Paulo}, title = {A survey on Barrett's esophagus analysis using machine learning}, series = {Computers in Biology and Medicine}, volume = {96}, journal = {Computers in Biology and Medicine}, publisher = {Elsevier}, doi = {10.1016/j.compbiomed.2018.03.014}, pages = {203 -- 213}, abstract = {This work presents a systematic review concerning recent studies and technologies of machine learning for Barrett's esophagus (BE) diagnosis and treatment. The use of artificial intelligence is a brand new and promising way to evaluate such disease. We compile some works published at some well-established databases, such as Science Direct, IEEEXplore, PubMed, Plos One, Multidisciplinary Digital Publishing Institute (MDPI), Association for Computing Machinery (ACM), Springer, and Hindawi Publishing Corporation. Each selected work has been analyzed to present its objective, methodology, and results. The BE progression to dysplasia or adenocarcinoma shows a complex pattern to be detected during endoscopic surveillance. Therefore, it is valuable to assist its diagnosis and automatic identification using computer analysis. The evaluation of the BE dysplasia can be performed through manual or automated segmentation through machine learning techniques. Finally, in this survey, we reviewed recent studies focused on the automatic detection of the neoplastic region for classification purposes using machine learning methods.}, subject = {Speiser{\"o}hrenkrankheit}, language = {en} } @inproceedings{SouzaJrPassosMendeletal., author = {Souza Jr., Luis Antonio de and Passos, Leandro A. and Mendel, Robert and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {Fine-tuning Generative Adversarial Networks using Metaheuristics}, series = {Bildverarbeitung f{\"u}r die Medizin 2021. Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, booktitle = {Bildverarbeitung f{\"u}r die Medizin 2021. Proceedings, German Workshop on Medical Image Computing, Regensburg, March 7-9, 2021}, publisher = {Springer Vieweg}, address = {Wiesbaden}, isbn = {978-3-658-33197-9}, doi = {10.1007/978-3-658-33198-6_50}, pages = {205 -- 210}, abstract = {Barrett's esophagus denotes a disorder in the digestive system that affects the esophagus' mucosal cells, causing reflux, and showing potential convergence to esophageal adenocarcinoma if not treated in initial stages. Thus, fast and reliable computer-aided diagnosis becomes considerably welcome. Nevertheless, such approaches usually suffer from imbalanced datasets, which can be addressed through Generative Adversarial Networks (GANs). Such techniques generate realistic images based on observed samples, even though at the cost of a proper selection of its hyperparameters. Many works employed a class of nature-inspired algorithms called metaheuristics to tackle the problem considering distinct deep learning approaches. Therefore, this paper's main contribution is to introduce metaheuristic techniques to fine-tune GANs in the context of Barrett's esophagus identification, as well as to investigate the feasibility of generating high-quality synthetic images for early-cancer assisted identification.}, subject = {Endoskopie}, language = {en} } @article{SouzaJrPassosSantanaetal., author = {Souza Jr., Luis Antonio de and Passos, Leandro A. and Santana, Marcos Cleison S. and Mendel, Robert and Rauber, David and Ebigbo, Alanna and Probst, Andreas and Messmann, Helmut and Papa, Jo{\~a}o Paulo and Palm, Christoph}, title = {Layer-selective deep representation to improve esophageal cancer classification}, series = {Medical \& Biological Engineering \& Computing}, volume = {62}, journal = {Medical \& Biological Engineering \& Computing}, publisher = {Springer Nature}, address = {Heidelberg}, doi = {10.1007/s11517-024-03142-8}, pages = {3355 -- 3372}, abstract = {Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their accountability and transparency level must be improved to transfer this success into clinical practice. The reliability of machine learning decisions must be explained and interpreted, especially for supporting the medical diagnosis.For this task, the deep learning techniques' black-box nature must somehow be lightened up to clarify its promising results. Hence, we aim to investigate the impact of the ResNet-50 deep convolutional design for Barrett's esophagus and adenocarcinoma classification. For such a task, and aiming at proposing a two-step learning technique, the output of each convolutional layer that composes the ResNet-50 architecture was trained and classified for further definition of layers that would provide more impact in the architecture. We showed that local information and high-dimensional features are essential to improve the classification for our task. Besides, we observed a significant improvement when the most discriminative layers expressed more impact in the training and classification of ResNet-50 for Barrett's esophagus and adenocarcinoma classification, demonstrating that both human knowledge and computational processing may influence the correct learning of such a problem.}, language = {en} } @inproceedings{SouzaPachecodeAngeloetal., author = {Souza, Luis A. and Pacheco, Andr{\´e} G.C. and de Angelo, Gabriel G. and Oliveira-Santos, Thiago and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {LiwTERM: A Lightweight Transformer-Based Model for Dermatological Multimodal Lesion Detection}, series = {2024 37th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), Manaus, Brazil, 9/30/2024 - 10/3/2024}, booktitle = {2024 37th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), Manaus, Brazil, 9/30/2024 - 10/3/2024}, publisher = {IEEE}, isbn = {979-8-3503-7603-6}, doi = {10.1109/SIBGRAPI62404.2024.10716324}, pages = {1 -- 6}, abstract = {Skin cancer is the most common type of cancer in the world, accounting for approximately 30\% of all diagnosed tumors. Early diagnosis reduces mortality rates and prevents disfiguring effects in different body regions. In recent years, machine learning techniques, particularly deep learning, have shown promising results in this task, presenting studies that have demonstrated that combining a patient's clinical information with images of the lesion is crucial for improving the classification of skin lesions. Despite that, meaningful use of clinical information with multiple images is mandatory, requiring further investigation. Thus, this project aims to contribute to developing multimodal machine learning-based models to cope with the skin lesion classification task employing a lightweight transformer model. As a main hypothesis, models can take multiple images from different sources as input, along with clinical information from the patient's history, leading to a more reliable diagnosis. Our model deals with the not-trivial task of combining images and clinical information (from anamneses) concerning the skin lesions in a lightweight transformer architecture that does not demand high computation resources but still presents competitive classification results.}, language = {en} } @article{SouzaPachecodeSouzaetal., author = {Souza, Luis A. and Pacheco, Andr{\´e} G.C. and de Souza, Alberto F. and Oliveira-Santos, Thiago and Badue, Claudine and Palm, Christoph and Papa, Jo{\~a}o Paulo}, title = {TransConv: a lightweight architecture based on transformers and convolutional neural networks for adenocarcinoma and Barrett's esophagus identification}, series = {Neural Computing and Applications}, journal = {Neural Computing and Applications}, number = {37}, publisher = {Springer}, doi = {10.1007/s00521-025-11299-y}, pages = {15535 -- 15546}, abstract = {Barrett's esophagus, also known as BE, is commonly associated with repeated exposure to stomach acid. If not treated properly, it may evolve into esophageal adenocarcinoma, aka esophageal cancer. This paper proposes TransConv, a hybrid architecture that benefits from features learned by pre-trained vision transformers (ViTs) and convolutional neural networks (CNNs), followed by a shallow neural network composed of three normalizations, ReLU activations, and fully connected layers, and a SoftMax head to distinguish between BE and esophageal cancer. TransConv is designed to be training-lightweight, and for the ViT and CNN backbone models, weights are kept frozen during training, i.e., the primary goal of TransConv is to learn the weights of the fully connected layer from both backbones only, avoiding the burden of updating their weights but still learning their final descriptions for the lightweight convolutional model. We report promising results with low computational training costs in two datasets, one public and another private. From our achievements, TransConv was able to deliver balanced accuracy results around 85\% and 86\% for each evaluated dataset, respectively, in a design that required only 50 epochs of model training, a very reduced number compared to state-of-the-art conducted studies in the same domain.}, language = {en} } @article{SpillerDegenKronewaldetal., author = {Spiller, Ralf and Degen, Matthias and Kronewald, Elke and Guertler, Katherine}, title = {Media watchblogs as an instrument of media accountability: An international survey}, series = {Journal of Applied Journalism \& Media Studies}, volume = {5}, journal = {Journal of Applied Journalism \& Media Studies}, number = {2}, publisher = {intellect}, issn = {2049-9531}, doi = {10.1386/ajms.5.2.151_1}, pages = {151 -- 176}, abstract = {Media watchdog weblogs can play an important role in media accountability, provided that the bloggers are regularly active and have an appropriate awareness of their role in the media landscape. The following article examines the presence and influence of media watchblogs in selected nations, as well as the operators' standards and perceptions of their roles. In semi-structured qualitative interviews, operators of watchblogs in Germany, Austria, Switzerland and the United Kingdom discussed their perceptions of their role, their ethical guidelines, their financing and any conflicts of interest, and their projections for the future. The interviews reveal that most media watchbloggers do not interpret their role to be a Fifth Estate with a mandate to monitor the mass media. The watchbloggers are reluctant to accept and uphold formal ethical and quality criteria, and limitations on time and money make them unable to guarantee ongoing critical media analysis. For these reasons, media watchblogs are determined to be a weak instrument of media accountability.}, language = {en} } @inproceedings{SpychalskiRodeRitthaleretal., author = {Spychalski, Dominik and Rode, Olaf and Ritthaler, Markus and Raptis, Georgios}, title = {Conceptual Design and Analysis of a Mobile Digital Identity for eHealth Applications}, series = {2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI): 27-30 July 2021, Athens, Greece}, booktitle = {2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI): 27-30 July 2021, Athens, Greece}, publisher = {IEEE}, issn = {2641-3604}, doi = {10.1109/BHI50953.2021.9508554}, pages = {1 -- 4}, abstract = {As mobile technology continues to improve, more and more professional services are being offered as mobile apps. This paradigm shift also affects eHealth applications. Digital identities in nation-wide eHealth infrastructures are often realized via smart cards, which however, do not support mobile applications well. In this paper we propose a concept of a mobile eID for eHealth based on smartphones with embedded secure hardware, an mobile authenticator app and an account manager as well as an Identity Provider (IdP) as backend services. The practical applicability of the concept is shown using the example of the German eHealth infrastructure. Our method generates a cryptographic key pair in secure hardware on the user's smartphone, registers it on the IdP and uses it to authenticate on the IdP. The security of the private key and the integrity of the smartphone is also validated and attestated. The user's established smartcard-based identity "Electronic Health Card" (EHC) forms the trust anchor. To authenticate against specialist eHealth apps the IdP issues standard-compliant OAuth2.0/OIDC tokens with a limited period of validity. Furthermore, in our security analysis we demonstrate that based on specific security requirements for smartphones and operating systems, at least the eIDAS security level "substantial" related to the technical security aspects of the system can be achieved. On the basis of this research German legislation was adjusted and "digital identities" supplementary to the smartcard-based EHC will be issued from 2023 in the German eHealth infrastructure.}, language = {en} }