Real-time instance segmentation of surgical instruments using attention and multi-scale feature fusion.

Precise instrument segmentation aids surgeons to navigate the body more easily and increases patient safety. While accurate tracking of surgical instruments in real-time plays a crucial role in minimally invasive computer-assisted surgeries, it is a challenging task to achieve, mainly due to: (1) a complex surgical environment, and (2) model design trade-off in terms of both optimal accuracy and speed. Deep learning gives us the opportunity to learn complex environment from large surgery scene environments and placements of these instruments in real world scenarios. The Robust Medical Instrument Segmentation 2019 challenge (ROBUST-MIS) provides more than 10,000 frames with surgical tools in different clinical settings. In this paper, we propose a light-weight single stage instance segmentation model complemented with a convolutional block attention module for achieving both faster and accurate inference. We further improve accuracy through data augmentation and optimal anchor localization strategies. To our knowledge, this is the first work that explicitly focuses on both real-time performance and improved accuracy. Our approach out-performed top team performances in the most recent edition of ROBUST-MIS challenge with over 44% improvement on area-based multi-instance dice metric MI_DSC and 39% on distance-based multi-instance normalized surface dice MI_NSD. We also demonstrate real-time performance (>60 frames-per-second) with different but competitive variants of our final approach.

Assessing YOLACT++ for real time and robust instance segmentation of medical instruments in endoscopic procedures.

Image-based tracking of laparoscopic instruments plays a fundamental role in computer and robotic-assisted surgeries by aiding surgeons and increasing patient safety. Computer vision contests, such as the Robust Medical Instrument Segmentation (ROBUST-MIS) Challenge, seek to encourage the development of robust models for such purposes, providing large, diverse, and high-quality datasets. To date, most of the existing models for instance segmentation of medical instruments were based on two-stage detectors, which provide robust results but are nowhere near to the real-time, running at 5 frames-per-second (fps) at most. However, for the method to be clinically applicable, a real-time capability is utmost required along with high accuracy. In this paper, we propose the addition of attention mechanisms to the YOLACT architecture to allow real-time instance segmentation of instruments with improved accuracy on the ROBUST-MIS dataset. Our proposed approach achieves competitive performance compared to the winner of the 2019 ROBUST-MIS challenge in terms of robustness scores, obtaining 0.313 ML_DSC and 0.338 MLNSD while reaching real-time performance at >45 fps.

U-isotopes and (226)Ra as tracers of hydrogeochemical processes in carbonated karst aquifers from arid areas.

Sierra de Gádor is a karst macrosystem with a highly complex geometry, located in southeastern Spain. In this arid environment, the main economic activities, agriculture and tourism, are supported by water resources from the Sierra de Gádor aquifer system. The aim of this work was to study the levels and behaviour of some of the most significant natural radionuclides in order to improve the knowledge of the hydrogeochemical processes involved in this groundwater system. For this study, 28 groundwater and 7 surface water samples were collected, and the activity concentrations of the natural U-isotopes ((238)U, (235)U and (234)U) and (226)Ra by alpha spectrometry were determined. The activity concentration of (238)U presented a large variation from around 1.1 to 65 mBq L(-1). Elevated groundwater U concentrations were the result of oxidising conditions that likely promoted U dissolution. The PHREEQC modelling code showed that dissolved U mainly existed as uranyl carbonate complexes. The (234)U/(238)U activity ratios were higher than unity for all samples (1.1-3.8). Additionally, these ratios were in greater disequilibrium in groundwater than surface water samples, the likely result of greater water-rock contact time. (226)Ra presented a wide range of activity concentrations, (0.8 up to about 4 × 10(2) mBq L(-1)); greatest concentrations were detected in the thermal area of Alhama. Most of the samples showed (226)Ra/(234)U activity ratios lower than unity (median = 0.3), likely the result of the greater mobility of U than Ra in the aquifer system. The natural U-isotopes concentrations were strongly correlated with dissolution of sulphate evaporites (mainly gypsum). (226)Ra had a more complex behaviour, showing a strong correlation with water salinity, which was particularly evident in locations where thermal anomalies were detected. The most saline samples showed the lowest (234)U/(238)U activity ratios, probably due to fast uniform bulk mineral dissolution, which would minimize the impact of solubility-controlled fractionation processes. Furthermore, the high bulk dissolution rates promoted greater groundwater (226)Ra/(234)U ratios because the Ra has a comparatively much greater mobility than U in saline conditions.

Groundwater contamination evolution in the Guadiamar and Agrio aquifers after the Aznalcóllar spill: assessment and environmental implications.

In 1998, the Agrio and Guadiamar rivers underwent an enormous environmental disaster caused by the rupture of the Aznalcóllar tailings dam and the release of 6 hm(3) of pyrite sludge and acidic water. Both rivers run over recent alluvial materials which form a small-sized aquifer which is however important because underground water feeds the flow of the rivers. This work analyzes the state of groundwater 10 years after the spill. Before the dam failure, this aquifer was already contaminated in the zone nearest to the mine, to which the impact of the spill was added. Contamination levels in the alluvial aquifer of the Agrio River have decreased remarkably. However, they are still important, with acidic pH values and high concentrations of toxic elements (maximum values of 16 mg/L of Zn and 15 mg/L of Al). There are also important levels of contamination in the Guadiamar alluvial area closest to the mine, as well as in specific zones located further south. The concentration of toxic elements is mainly controlled by pH. The evolution of contaminant levels show a sharp decrease after the first years following the spill, followed by a subsequent stabilization. It is necessary to take measures for the recovery of the aquifer because, otherwise, groundwater will continue contributing contaminants into the Agrio and Guadiamar rivers.