Article
Dermatology
George A. Zakhem, Joseph W. Fakhoury, Catherine C. Motosko, Roger S. Ho
Summary: This study characterized the evolution of AI in skin cancer assessment and found that articles with dermatologists included as authors described algorithms built with more images. Dermatologists' greater involvement in issues related to data collection, biases in data sets, and technology applications is crucial for the development of AI models in skin cancer assessment.
JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY
(2021)
Review
Oncology
Katherine Sanchez, Kanika Kamal, Priya Manjaly, Sophia Ly, Arash Mostaghimi
Summary: The development and implementation of artificial intelligence (AI) in dermatology is impacting patient care. AI has mainly been applied in clinical settings for melanoma, but it is also critically important for non-melanoma skin cancers such as basal cell and squamous cell cancers. AI can help improve diagnosis time and accuracy, predict therapeutic response, and aid in designing new therapies. However, realistic expectations and transparent guidelines are necessary to promote confidence in AI systems. Dermatologists play a crucial role in curating diverse and high-quality datasets for training AI algorithms. AI should be seen as a tool to complement dermatologists' expertise rather than a replacement.
CURRENT TREATMENT OPTIONS IN ONCOLOGY
(2023)
Article
Medicine, General & Internal
Helen Marsden, Caroline Morgan, Stephanie Austin, Claudia Degiovanni, Marcello Venzi, Polychronis Kemos, Jack Greenhalgh, Dan Mullarkey, Ioulios Palamaras
Summary: Identification of skin cancer using AI-based digital health technology can improve the triage and management of suspicious skin lesions. The DERM-003 study demonstrated the effectiveness of an AI as a medical device in identifying SCC, BCC, pre-malignant, and benign lesions. Dermoscopic images taken with smartphone cameras were assessed by the AI and compared to histopathology results.
FRONTIERS IN MEDICINE
(2023)
Article
Dermatology
A. Nikolas MacLellan, Emma L. Price, Pamela Publicover-Brouwer, Kara Matheson, Thai Yen Ly, Sylvia Pasternak, Noreen M. Walsh, Christopher J. Gallant, Amanda Oakley, Peter R. Hull, Richard G. Langley
Summary: This study compared the diagnostic accuracy of clinical examination, teledermatology, and noninvasive imaging techniques in detecting melanoma, with FotoFinder Moleanalyzer Pro showing the highest sensitivity and specificity. However, using pathology as the gold standard for comparison has limitations.
JOURNAL OF THE AMERICAN ACADEMY OF DERMATOLOGY
(2021)
Article
Chemistry, Analytical
Giansalvo Cirrincione, Sergio Cannata, Giovanni Cicceri, Francesco Prinzi, Tiziana Currieri, Marta Lovino, Carmelo Militello, Eros Pasero, Salvatore Vitabile
Summary: This paper presents a ViT-based architecture for classifying melanoma versus non-cancerous lesions. The proposed model is trained and tested on public skin cancer data from the ISIC challenge, and the results obtained are highly promising. Different classifier configurations are analyzed to find the most discriminating one, with the best accuracy of 0.948, sensitivity of 0.928, specificity of 0.967, and AUROC of 0.948.
Review
Microbiology
Te Sun, Xueli Niu, Qing He, Fujun Chen, Rui-Qun Qi
Summary: Microorganisms are closely linked to skin diseases, and imbalances or invasions of external pathogens can cause various skin diseases. The development and prognosis of these diseases are closely related to the type and composition of microorganisms present. Therefore, detecting the characteristics and changes in microorganisms can greatly improve the diagnosis and prediction of skin diseases.
FRONTIERS IN MICROBIOLOGY
(2023)
Review
Medicine, General & Internal
Ana Maria Malciu, Mihai Lupu, Vlad Mihai Voiculescu
Summary: Reflectance confocal microscopy (RCM) is a non-invasive imaging method designed for identifying various skin diseases, but the diagnosis process may be subjective. Deep learning technologies and machine learning algorithms developed in recent years provide a more objective approach to RCM image analysis, reducing artifacts and evaluation times.
JOURNAL OF CLINICAL MEDICINE
(2022)
Review
Health Care Sciences & Services
Abdulrahman Takiddin, Jens Schneider, Yin Yang, Alaa Abd-Alrazaq, Mowafa Househ
Summary: AI tools are being used to detect and classify skin cancer efficiently, with shallow and deep machine learning techniques being utilized. This study aimed to identify and group different AI-based technologies used for skin cancer diagnosis, analyzing the correlation between data set size, diagnostic classes, and model performance. Results showed that studies with smaller data sets tended to report higher accuracy, but direct comparisons between methods were hindered by varied evaluation metrics and image types.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Dermatology
Konstantinos Liopyris, Stamatios Gregoriou, Julia Dias, Alexandros J. Stratigos
Summary: Artificial intelligence (AI) has great potential in dermatology, particularly in the early diagnosis and evaluation of skin cancer. Studies have shown that machine learning and convolutional neural networks can outperform or at least perform as well as clinicians in recognizing skin lesions. However, the application of AI algorithms in everyday clinical practice still faces challenges.
DERMATOLOGY AND THERAPY
(2022)
Article
Medicine, General & Internal
Victoriya Andreeva, Evgeniia Aksamentova, Andrey Muhachev, Alexey Solovey, Igor Litvinov, Alexey Gusarov, Natalia N. Shevtsova, Dmitry Kushkin, Karina Litvinova
Summary: The diagnosis and treatment of non-melanoma skin cancer is a pressing issue. Fluorescence spectroscopy has shown the ability to detect abnormal cells, but distinguishing changes in overall spectrum and molecular basis is challenging due to overlapping spectra. By applying deep learning algorithms, we evaluated the ability of fluorescence spectroscopy to differentiate between pathologies and normal skin, achieving a high rate of correct prediction.
Article
Multidisciplinary Sciences
Jane Scheetz, Philip Rothschild, Myra McGuinness, Xavier Hadoux, H. Peter Soyer, Monika Janda, James J. J. Condon, Luke Oakden-Rayner, Lyle J. Palmer, Stuart Keel, Peter van Wijngaarden
Summary: The survey revealed that the majority of medical specialists believe that artificial intelligence will improve healthcare outcomes and impact workforce demands in the next decade. Key benefits of artificial intelligence include improved disease screening and streamlining of monotonous tasks, while concerns revolve around medical liability and the involvement of technology companies in healthcare.
SCIENTIFIC REPORTS
(2021)
Review
Oncology
Qianwei Liu, Jie Zhang, Yanping Bai
Summary: This study employs visual bibliometric analysis to explore the evolution and deployment of AI in the context of skin cancer diagnosis. The results show that the application of AI in skin cancer research is rapidly expanding but still at the feasibility study stage, with challenges in algorithm stability, data quality, and result interpretability.
FRONTIERS IN ONCOLOGY
(2023)
Review
Oncology
Haleigh Stafford, Jane Buell, Elizabeth Chiang, Uma Ramesh, Michael Migden, Priyadharsini Nagarajan, Moran Amit, Dan Yaniv
Summary: Non-melanoma skin cancer is a common and increasing cancer diagnosis globally. Deep learning, a type of artificial intelligence, has shown promise in aiding in the diagnosis of non-melanoma skin cancer. This review discusses the evidence, available technologies and remote applications, and attitudes and ethical considerations related to deep learning in skin cancer screening and diagnosis.
Letter
Dermatology
Ilya Klabukov, Denis Baranovskii
Summary: Artificial intelligence has potential applications in dermatology to reduce misdiagnosis rates, with multispectral imaging being particularly helpful in identifying skin malignancies.
CLINICAL AND EXPERIMENTAL DERMATOLOGY
(2023)
Article
Health Care Sciences & Services
Dina Nur Anggraini Ningrum, Sheng-Po Yuan, Woon-Man Kung, Chieh-Chen Wu, I-Shiang Tzeng, Chu-Ya Huang, Jack Yu-Chuan Li, Yao-Chin Wang
Summary: Skin cancer, especially melanoma, poses a global burden with increasing incidence each year. Automated detection using deep convolutional neural networks (CNN) shows promise, but requires high computational resources. This study successfully combines patient metadata with dermoscopic images, leading to improved accuracy in melanoma detection, suitable for low-resource healthcare settings.
JOURNAL OF MULTIDISCIPLINARY HEALTHCARE
(2021)
Article
Oncology
Max Jung, Michael Rose, Ruth Knuechel, Chiara Loeffler, Hannah Muti, Jakob Nikolas Kather, Nadine T. Gaisa
Summary: Quantitative analysis of immune phenotypes in squamous bladder cancer (sq-BLCA) revealed that hot tumor-immune phenotypes with high PD-L1 expression might be a promising strategy for ICI therapy in sq-BLCA.
Article
Oncology
Georg C. Lodde, Philipp Jansen, Rudolf Herbst, Patrick Terheyden, Jochen Utikal, Claudia Pfoehler, Jens Ulrich, Alexander Kreuter, Peter Mohr, Ralf Gutzmer, Friedegund Meier, Edgar Dippel, Michael Weichenthal, Antje Sucker, Jan-Malte Placke, Anne Zaremba, Lea Jessica Albrecht, Bernd Kowall, Wolfgang Galetzka, Jürgen C. Becker, Alpaslan Tasdogan, Lisa Zimmer, Elisabeth Livingstone, Eva Hadaschik, Dirk Schadendorf, Selma Ugurel, Klaus Griewank
Summary: RAC1 mutations are rare in melanoma, mostly originating from the skin or of unknown origin, and often coexist with MAP kinase mutations. Immune checkpoint inhibitors have shown to be effective in treating advanced disease.
EUROPEAN JOURNAL OF CANCER
(2023)
Article
Oncology
Theresa Steeb, Julia Bruetting, Lydia Reinhardt, Julia Hoffmann, Nina Weiler, Markus Heppt, Michael Erdmann, Astrid Doppler, Christiane Weber, Dirk Schadendorf, Friedegund Meier, Carola Berking
Summary: Skin cancer patients are increasingly using the internet to find information about their disease. However, online information can be misleading. SKINFO is a new website designed specifically for German-speaking skin cancer patients, providing reliable and verified information. Usability testing of the website showed that while patients appreciated the content and design, improvements could be made in terms of navigation and providing more specific information for patients' relatives.
JOURNAL OF CANCER EDUCATION
(2023)
Article
Dermatology
Dana Westphal, Matthias Meinhardt, Konrad Gruetzmann, Lisa Schoene, Julian Steininger, Lena T. Neuhaus, Miriam Wiegel, Daniel Schrimpf, Daniela E. Aust, Evelin Schroeck, Gustavo B. Baretton, Stefan Beissert, Tareq A. Juratli, Gabriele G. Schackert, Jan Gravemeyer, Juergen C. Becker, Andreas von Deimling, Christian Koelsche, Barbara Klink, Friedegund Meier, Alexander Schulz, Michael H. Muders, Michael Seifert
Summary: This study integrated omics data analysis to evaluate the methylomes and transcriptomes of matched melanoma metastases, identifying 38 candidate genes with distinct promoter methylation and gene expression changes in intracranial compared with extracranial metastases. The protein expression of the 11 most promising genes was validated using immunohistochemistry, and significant differences were observed in intracranial metastases. Knockdown of PRKCZ or GRB10 altered protein kinase B expression and decreased the viability of a brain-specific melanoma cell line. These findings provide insights into the molecular mechanisms that differentiate brain metastases and could be targeted for therapy.
JOURNAL OF INVESTIGATIVE DERMATOLOGY
(2023)
Article
Dermatology
Katharina Schumann, Cornelia Mauch, Kai-Christian Klespe, Carmen Loquai, Ulrike Nikfarjam, Max Schlaak, Larissa Akcetin, Peter Koelblinger, Magdalena Hoellwerth, Markus Meissner, Guelcin Mengi, Andreas Dominik Braun, Miriam Mengoni, Reinhard Dummer, Joanna Mangana, Mihaela-Anca Sindrilaru, Dan Radmann, Christine Hafner, Johann Freund, Klemens Rappersberger, Felix Weihsengruber, Frank Meiss, Lydia Reinhardt, Friedegund Meier, Barbara Rainer, Erika Richtig, Julia Maria Ressler, Christoph Hoeller, Thomas Eigentler, Teresa Amaral, Wiebke K. Peitsch, Uwe Hillen, Wolfgang Harth, Fabian Ziller, Kerstin Schatton, Thilo Gambichler, Laura Susok, Lara Valeska Maul, Heinz Laubli, Dirk Debus, Carsten Weishaupt, Sevil Boerger, Katharina Sievers, Sebastian Haferkamp, Veronika Zenderowski, Van Anh Nguyen, Marina Wanner, Ralf Gutzmer, Patrick Terheyden, Katharina Kaehler, Steffen Emmert, Alexander Thiem, Michael Sachse, Silke Gercken-Riedel, Kjell Matthias Kaune, Kai-Martin Thoms, Lucie Heinzerling, Markus Vincent Heppt, Sabine Tratzmiller, Wolfram Hoetzenecker, Angela Oellinger, Andreas Steiner, Tobias Peinhaupt, Maurizio Podda, Sabine Schmid, Uwe Wollina, Tilo Biedermann, Christian Posch
Summary: This real-world study examines the clinical outcomes of adjuvant melanoma treatment using PD-1 antibodies and BRAF + MEK inhibitors in specialized skin cancer centers in Germany, Austria, and Switzerland. The results show that patients treated with PD-1 antibodies have a lower 12-month recurrence rate compared to patients treated with BRAF + MEK inhibitors. Additionally, PD-1 antibody treatment is not affected by BRAF mutation status.
JOURNAL OF THE EUROPEAN ACADEMY OF DERMATOLOGY AND VENEREOLOGY
(2023)
Letter
Dermatology
Farzan Solimani, Amrei Dilling, Franziska C. Ghoreschi, Alexander Nast, Kamran Ghoreschi, Katharina Meier
JOURNAL OF THE EUROPEAN ACADEMY OF DERMATOLOGY AND VENEREOLOGY
(2023)
Article
Otorhinolaryngology
Tanja Jutzi, Eva I. Krieghoff-Henning, Titus J. Brinker
Summary: The incidence of malignant melanoma is increasing worldwide and early detection is crucial for effective treatment. Although skin cancer early detection has improved, visual detection of early melanomas remains challenging due to morphological overlaps with nevi. Therefore, there is a high medical need to develop methods for reliable early skin cancer detection.
LARYNGO-RHINO-OTOLOGIE
(2023)
Article
Biochemistry & Molecular Biology
Sebastian Foersch, Christina Glasner, Ann-Christin Woerl, Markus Eckstein, Daniel-Christoph Wagner, Stefan Schulz, Franziska Kellers, Aurelie Fernandez, Konstantina Tserea, Michael Kloth, Arndt Hartmann, Achim Heintz, Wilko Weichert, Wilfried Roth, Carol Geppert, Jakob Nikolas Kather, Moritz Jesinghaus
Summary: Researchers developed a multistain deep learning model utilizing artificial intelligence to determine the immunoscore and predict response to neoadjuvant therapy in colorectal cancer patients. The model outperformed other clinical, molecular, and immune cell-based parameters and can serve as a valuable decision-making tool for clinicians.
Article
Multidisciplinary Sciences
Anna M. Sigmund, Markus Winkler, Sophia Engelmayer, Daniela Kugelmann, Desalegn T. Egu, Letyfee S. Steinert, Michael Fuchs, Matthias Hiermair, Mariya Y. Radeva, Franziska C. Bayerbach, Elisabeth Butz, Stefan Kotschi, Christoph Hudemann, Michael Hertl, Sunil Yeruva, Enno Schmidt, Amir S. Yazdi, Kamran Ghoreschi, Franziska Vielmuth, Jens Waschke
Summary: Pemphigus vulgaris is a life-threatening blistering skin disease caused by autoantibodies. Apremilast, a phosphodiesterase 4 inhibitor used in psoriasis, was found to prevent skin blistering in pemphigus vulgaris. It stabilizes keratinocyte adhesion and induces phosphorylation of plakoglobin, leading to the assembly of desmosomal plaques.
NATURE COMMUNICATIONS
(2023)
Article
Multidisciplinary Sciences
Theresa Kraft, Konrad Gruetzmann, Matthias Meinhardt, Friedegund Meier, Dana Westphal, Michael Seifert
Summary: Melanomas can metastasize to distant organs, including the brain, posing a clinical challenge. This study reveals that the DNA methylation patterns in intra- and extracranial melanoma metastases are more similar to each other within individual patients than to metastases in the same tissue from other patients. The analysis identifies specific DNA methylation alterations, signaling pathway genes, and gene expression changes that distinguish intra- from extracranial metastases. The findings contribute to a better understanding of the DNA methylation differences in melanoma metastases and may guide future therapeutic approaches.
SCIENTIFIC REPORTS
(2023)
Article
Medicine, General & Internal
Georgios Kokolakis, Kamran Ghoreschi
Summary: The understanding of immunopathogenesis of psoriasis has led to the development of targeted therapies, including IL-17 antibodies which have been established for the treatment of psoriasis, psoriatic arthritis, and axial spondyloarthritis. This study presents two patients with psoriasis who lost response to previous treatments, but responded well to bimekizumab, a new antibody that neutralizes both IL-17A and IL-17F. The fast response and lack of side effects suggest the important role of IL-17F and the potential benefit of bimekizumab in non-responders to other anti-IL-17 therapies.
JOURNAL OF CLINICAL MEDICINE
(2023)
Article
Dermatology
Anna Balato, Emanuele Scala, Kilian Eyerich, Nicole Costantino Brembilla, Andrea Chiricozzi, Robert Sabat, Kamran Ghoreschi
Summary: Modern treatments for psoriasis target the immune pathways associated with the disease, but the clinical evidence for increased infection risk is confounded by comorbidities in patients. In an era with a growing infection risk, it is important to stay updated on this risk. This mini-review discusses the immunopathogenesis of psoriasis, the risk of infections linked to the disease and systemic therapy, and provides an overview of infection prevention and management.
DERMATOLOGY PRACTICAL & CONCEPTUAL
(2023)
Letter
Biochemistry & Molecular Biology
Daniel Truhn, Jorge S. Reis-Filho, Jakob Nikolas Kather
Article
Cell Biology
Alexander A. Wurm, Silke Brilloff, Sofia Kolovich, Silvia Schaefer, Elahe Rahimian, Vida Kufrin, Marius Bill, Zunamys I. Carrero, Stephan Drukewitz, Alexander Krueger, Melanie Huether, Sebastian Uhrig, Sandra Oster, Dana Westphal, Friedegund Meier, Katrin Pfutze, Daniel Huebschmann, Peter Horak, Simon Kreutzfeldt, Daniela Richter, Evelin Schroeck, Gustavo Baretton, Christoph Heining, Lino Moehrmann, Stefan Froehling, Claudia R. Ball, Hanno Glimm
Summary: In this study, we investigate the impact of pathway activation on miRNA expression patterns using small RNA sequencing. We discover that miRNAs capable of inhibiting key members of activated pathways are frequently diminished. Based on this observation, we develop an approach to identify druggable target genes in cancer by integrating a low-miRNA-expression signature. We validate our approach in colorectal cancer cells and diverse cancer models and demonstrate its additional value in supporting drug prediction strategies.
CELL REPORTS MEDICINE
(2023)
Article
Oncology
Philipp Jansen, Jean Le 'Clerc Arrastia, Daniel Otero Baguer, Maximilian Schmidt, Jennifer Landsberg, Joerg Wenzel, Michael Emberger, Dirk Schadendorf, Eva Hadaschik, Peter Maass, Klaus Georg Griewank
Summary: This study highlights the enormous potential of artificial intelligence in pathology, showing that it can aid in the identification of rare cutaneous adnexal tumors and potentially become a standard tool in routine diagnostics.
EUROPEAN JOURNAL OF CANCER
(2024)