Search

Semantic segmentation of organs and tissue types is an important sub-problem in image based scene understanding for laparoscopic surgery and is a prerequisite for context-aware assistance and cognitive robotics. Deep Learning (DL) approaches are prominently applied to segmentation and tracking of laparoscopic instruments. This work compares different combinations of neural networks, loss functions, and training strategies in their application to semantic segmentation of different organs and tissue types in human laparoscopic images in order to investigate their applicability as components in cognitive systems. TernausNet-11 trained on Soft-Jaccard loss with a pretrained, trainable encoder performs best in regard to segmentation quality (78.31% mean Intersection over Union [IoU]) and inference time (28.07 ms) on a single GTX 1070 GPU.

The scope of this work is to study some aspects of the formation of the regional climate of the Caucasus (with a specific focus on Georgia) against the background of the impact of mineral aerosols using modelling (the RegCM interactively coupled with a dust module, WRF-Chem, and HYSPLIT models) and satellite data (MODIS, CALIPSO). The annual mean, as well as the error in summer and winter temperatures, standard deviation and correlation coefficient compared to the CRU data were calculated for 8 sub-regions with different orographic and climate properties. The calculation results showed that dust aerosol is an active player in the climatic system of the Caucasus (Georgia). Numerical results showed that the inclusion of dust radiative forcing in the RegCM numerical model brought the simulated summer temperature closer to the observed temperature values. The mean annual temperature increased throughout Georgia in simulations that took into account the direct impact of dust. Calculations using the WRF-Chem and HYSPLIT models revealed that during the study period, aeolian dust was brought into the territory of the South Caucasus (Georgia) equally not only from Africa and the Middle East, but also from Central (Western) Asia deserts, which was not noted earlier.

In this article for detection of gas accidental escape localization in the branched gas pipelines two mathematical models are suggested. The first model is indented for leak detection and localization in the horizontal branched pipeline and second one for an inclined section of the main gas pipeline. The algorithm of leak localization in the branched pipeline is not demand on knowledge of corresponding initial hydraulic parameters at entrance and ending points of each sections of pipeline. For detection of the damaged section and then leak localization in this section special functions and equations are constructed. Some results of calculations for horizontal pipelines having two, four and five sections are presented. Also a method and formula for the leak localization in the inclined section of the main gas pipeline are suggested. Some results of numerical calculations for the inclined pipeline are presented too. [ABSTRACT FROM AUTHOR]

Nowadays, when the emphasis is on alternative means of energy, natural gas is still used as an efficient and convenient fuel both in the home (for heating buildings and water, cooking, drying and lighting) and in industry together with electricity. In industrial terms, gas is one of the main sources of electricity generation in both developed and developing countries. Pipelines are the most popular means of transporting natural gas domestically and internationally. The main reasons for the constipation of gas pipelines are the formation of hydrates, freezing of water plugs, pollution, etc. It is an urgent task to take timely measures against the formation of hydrates in the pipeline. To stop gas hydrate formation in gas transporting pipelines, from existing methods the mathematical modelling with hydrodynamic method is more acceptable. In this paper the problem of prediction of possible points of hydrates origin in the main pipelines taking into consideration gas non-stationary flow and heat exchange with medium is studied. For solving the problem the system of partial differential equations governing gas non-stationary flow in main gas pipeline is investigated. The problem solution for gas adiabatic flow is presented.

Over the past two decades, Georgia has faced increasingly heavy rainfall, hail and flooding, which especially devastated Kakheti wine region in Southern Georgia, causing severe damage to hundreds of vineyards. Since 2015, 85 anti-hail missile systems have been installed to protect entire Kakheti region, however, for the effective use of a modern anti-hail system, it became necessary to timely forecast extreme weather events of a regional and local scale. Thus, this article aims to develop timely forecasting of strong convection, dangerous precipitation and hail using modern weather forecasting models and radar technologies in Georgia. For this reasons various combinations of the physics parameterization schemes of the WRF-ARW model, the ARL READY system and the data of the modern meteorological radar Meteor 735CDP10 are used to predict the thermodynamic state of the atmosphere and assess the possible level of development of convective processes. The analysis of the calculated results showed that the variants of the microphysics parametrization schemes of the WRF model lead to significant variability in precipitation forecasts on complex terrain. Meanwhile, the upper-air diagrams of the READY system clearly showed the instability of the atmosphere for the cases discussed. Some results of these calculations are presented and analysed in this paper.

The problem of the ongoing climate change resulting from natural and growing anthropogenic factors acquires a particular importance for the territory of the Caucasus. Dust aerosol represent one of the main pollutants on the territory of Georgia and impact on regional climate. In this study, the WRF Chemistry model with dust module is used to study transportation of dust to the territory of the South Caucasus from the Sahara and Sahel in Africa, Arabian and ar-Rub’ al-Khali deserts located in the Middle East, Kyzylkum, Karakum in the Central Asia. The results of calculations have shown the WRF model was able to simulate dust aerosols transportation to the Caucasus reliably in conditions of the complex topography and that dust aerosol is an important factor in the climate system of the South Caucasus.

Background: With Surgomics, we aim for personalized prediction of the patient's surgical outcome using machine-learning (ML) on multimodal intraoperative data to extract surgomic features as surgical process characteristics. As high-quality annotations by medical experts are crucial, but still a bottleneck, we prospectively investigate active learning (AL) to reduce annotation effort and present automatic recognition of surgomic features., Methods: To establish a process for development of surgomic features, ten video-based features related to bleeding, as highly relevant intraoperative complication, were chosen. They comprise the amount of blood and smoke in the surgical field, six instruments, and two anatomic structures. Annotation of selected frames from robot-assisted minimally invasive esophagectomies was performed by at least three independent medical experts. To test whether AL reduces annotation effort, we performed a prospective annotation study comparing AL with equidistant sampling (EQS) for frame selection. Multiple Bayesian ResNet18 architectures were trained on a multicentric dataset, consisting of 22 videos from two centers., Results: In total, 14,004 frames were tag annotated. A mean F1-score of 0.75 ± 0.16 was achieved for all features. The highest F1-score was achieved for the instruments (mean 0.80 ± 0.17). This result is also reflected in the inter-rater-agreement (1-rater-kappa > 0.82). Compared to EQS, AL showed better recognition results for the instruments with a significant difference in the McNemar test comparing correctness of predictions. Moreover, in contrast to EQS, AL selected more frames of the four less common instruments (1512 vs. 607 frames) and achieved higher F1-scores for common instruments while requiring less training frames., Conclusion: We presented ten surgomic features relevant for bleeding events in esophageal surgery automatically extracted from surgical video using ML. AL showed the potential to reduce annotation effort while keeping ML performance high for selected features. The source code and the trained models are published open source., (© 2023. The Author(s).)

Purpose: Endovascular interventions require intense practice to develop sufficient dexterity in catheter handling within the human body. Therefore, we present a modular training platform, featuring 3D-printed vessel phantoms with patient-specific anatomy and integrated piezoresistive impact force sensing of instrument interaction at clinically relevant locations for feedback-based skill training to detect and reduce damage to the delicate vascular wall., Methods: The platform was fabricated and then evaluated in a user study by medical ([Formula: see text]) and non-medical ([Formula: see text]) users. The users had to navigate a set of guidewire and catheter through a parkour of 3 modules including an aneurismatic abdominal aorta, while impact force and completion time were recorded. Eventually, a questionnaire was conducted., Results: The platform allowed to perform more than 100 runs in which it proved capable to distinguish between users of different experience levels. Medical experts in the fields of vascular and visceral surgery had a strong performance assessment on the platform. It could be shown, that medical students could improve runtime and impact over 5 runs. The platform was well received and rated as promising for medical education despite the experience of higher friction compared to real human vessels., Conclusion: We investigated an authentic patient-specific training platform with integrated sensor-based feedback functionality for individual skill training in endovascular surgery. The presented method for phantom manufacturing is easily applicable to arbitrary patient-individual imaging data. Further work shall address the implementation of smaller vessel branches, as well as real-time feedback and camera imaging for further improved training experience., (© 2023. The Author(s).)

Purpose: Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive warnings and semi-autonomous robotic assistance or improve training of surgeons via data-driven feedback. In surgical workflow analysis up to 91% average precision has been reported for phase recognition on an open data single-center video dataset. In this work we investigated the generalizability of phase recognition algorithms in a multicenter setting including more difficult recognition tasks such as surgical action and surgical skill., Methods: To achieve this goal, a dataset with 33 laparoscopic cholecystectomy videos from three surgical centers with a total operation time of 22 h was created. Labels included framewise annotation of seven surgical phases with 250 phase transitions, 5514 occurences of four surgical actions, 6980 occurences of 21 surgical instruments from seven instrument categories and 495 skill classifications in five skill dimensions. The dataset was used in the 2019 international Endoscopic Vision challenge, sub-challenge for surgical workflow and skill analysis. Here, 12 research teams trained and submitted their machine learning algorithms for recognition of phase, action, instrument and/or skill assessment., Results: F1-scores were achieved for phase recognition between 23.9% and 67.7% (n = 9 teams), for instrument presence detection between 38.5% and 63.8% (n = 8 teams), but for action recognition only between 21.8% and 23.3% (n = 5 teams). The average absolute error for skill assessment was 0.78 (n = 1 team)., Conclusion: Surgical workflow and skill analysis are promising technologies to support the surgical team, but there is still room for improvement, as shown by our comparison of machine learning algorithms. This novel HeiChole benchmark can be used for comparable evaluation and validation of future work. In future studies, it is of utmost importance to create more open, high-quality datasets in order to allow the development of artificial intelligence and cognitive robotics in surgery., Competing Interests: Declaration of Competing Interest M. Wagner, B.-P. Müller-Stich, S. Speidel and S. Bodenstedt worked with medical device manufacturer KARL STORZ SE & Co. KG in the projects “InnOPlan”, funded by the German Federal Ministry of Economic Affairs and Energy (grant number BMWI 01MD15002E) and “Surgomics”, funded by the German Federal Ministry of Health (grant number BMG 2520DAT82D and BMG 2520DAT82A). Lars Mündermann is an employee of KARL STORZ SE & Co. KG. A. Reinke works with the Helmholtz Imaging Platform (HIP), a platform of the Helmholtz Incubator on Information and Data Science. S. Kondo was an employee of Konica Minolta Inc. when this work was done. Wolfgang Reiter is an employee of Wintegral GmbH, a subsidiary of medical device manufacturer Richard Wolf GmbH. I. Twick, K. Kirtac, E. Hosgor, J. Lindström Bolmgren, M. Stenzel and B. von Siemens are employees of Caresyntax GmbH. Felix Nickel received travel support for conference participation as well as equipment provided for laparoscopic surgery courses by KARL STORZ SE & Co. KG, Johnson & Johnson, Intuitive Surgical, Cambridge Medical Robotics, and Medtronic. The other authors have no conflicts of interest., (Copyright © 2023. Published by Elsevier B.V.)

Image-based tracking of medical instruments is an integral part of surgical data science applications. Previous research has addressed the tasks of detecting, segmenting and tracking medical instruments based on laparoscopic video data. However, the proposed methods still tend to fail when applied to challenging images and do not generalize well to data they have not been trained on. This paper introduces the Heidelberg Colorectal (HeiCo) data set - the first publicly available data set enabling comprehensive benchmarking of medical instrument detection and segmentation algorithms with a specific emphasis on method robustness and generalization capabilities. Our data set comprises 30 laparoscopic videos and corresponding sensor data from medical devices in the operating room for three different types of laparoscopic surgery. Annotations include surgical phase labels for all video frames as well as information on instrument presence and corresponding instance-wise segmentation masks for surgical instruments (if any) in more than 10,000 individual frames. The data has successfully been used to organize international competitions within the Endoscopic Vision Challenges 2017 and 2019.

Objectives: There is a large spectrum of viral, bacterial, fungal, and prion pathogens that cause central nervous system (CNS) infections. As such, identification of the etiological agent requires multiple laboratory tests and accurate diagnosis requires clinical and epidemiological information. This hospital-based study aimed to determine the main causes of acute meningitis and encephalitis and enhance laboratory capacity for CNS infection diagnosis., Methods: Children and adults patients clinically diagnosed with meningitis or encephalitis were enrolled at four reference health centers. Cerebrospinal fluid (CSF) was collected for bacterial culture, and in-house and multiplex RT-PCR testing was conducted for herpes simplex virus (HSV) types 1 and 2, mumps virus, enterovirus, varicella zoster virus (VZV), Streptococcus pneumoniae, HiB and Neisseria meningitidis., Results: Out of 140 enrolled patients, the mean age was 23.9 years, and 58% were children. Bacterial or viral etiologies were determined in 51% of patients. Five Streptococcus pneumoniae cultures were isolated from CSF. Based on in-house PCR analysis, 25 patients were positive for S. pneumoniae, 6 for N. meningitidis, and 1 for H. influenzae. Viral multiplex PCR identified infections with enterovirus (n = 26), VZV (n = 4), and HSV-1 (n = 2). No patient was positive for mumps or HSV-2., Conclusions: Study findings indicate that S. pneumoniae and enteroviruses are the main etiologies in this patient cohort. The utility of molecular diagnostics for pathogen identification combined with the knowledge provided by the investigation may improve health outcomes of CNS infection cases in Georgia.