ABSTRACT
BACKGROUND: The development of assistive technologies will become of increasing importance in the coming years and not only in surgery. The comprehensive perception of the actual situation is the basis of every autonomous action. Different sensor systems can be used for this purpose, of which video-based systems have a special potential. METHOD: Based on the available literature and on own research projects, central aspects of image-based support systems for surgery are presented. In this context, not only the potential but also the limitations of the methods are explained. RESULTS: An established application is the phase detection of surgical interventions, for which surgical videos are analyzed using neural networks. Through a time-based and transformative analysis the results of the prediction could only recently be significantly improved. Robotic camera guidance systems will also use image data to autonomously navigate laparoscopes in the near future. The reliability of the systems needs to be adapted to the high requirements in surgery by means of additional information. A comparable multimodal approach has already been implemented for navigation and localization during laparoscopic procedures. For this purpose, video data are analyzed using various methods and these data are fused with other sensor modalities. DISCUSSION: Image-based supportive methods are already available for various tasks and will become an important aspect for the surgery of the future; however, in order to be able to be reliably implemented for autonomous functions, they must be embedded in multimodal approaches in the future in order to provide the necessary security.
Subject(s)
Laparoscopes , Laparoscopy , Forecasting , Laparoscopy/methods , Neural Networks, Computer , Reproducibility of ResultsABSTRACT
Metabolic changes are linked to epigenetic reprogramming and play important roles in several tumor types. PGC-1α is a transcriptional coactivator controlling mitochondrial biogenesis and is linked to oxidative phosphorylation. We provide evidence that melanoma models with elevated PGC-1α levels are characteristic of the proliferative phenotype and are sensitive to bromodomain and extra-terminal domain (BET) inhibitor treatment. A super-enhancer region highly occupied by the BET family member BRD4 was identified for the PGC-1α gene. BET inhibitor treatment prevented this interaction, leading to a dramatic reduction of PGC-1α expression. Accordingly, BET inhibition diminished respiration and mitochondrial function in cells. In vivo, melanoma models with high PGC-1α expression strongly responded to BET inhibition by reduction of PGC-1α and impaired tumor growth. Altogether, our findings identify epigenetic regulatory elements that define a subset of melanomas with high sensitivity to BET inhibition, which opens up the opportunity to define melanoma patients most likely to respond to this treatment, depending on their tumor characteristics.