Cascaded multi-scale convolutional encoder-decoders for breast mass segmentation in high-resolution mammograms.

This paper addresses breast mass segmentation from high-resolution mammograms. To cope with strong class imbalance, huge diversity of size, shape, texture and contour as well as limited receptive field, mass segmentation is achieved through a multi-scale cascade of deep convolutional encoder-decoders without any pre-detection scheme. Multi-scale information is integrated using auto-context to make long-range spatial context arising from lower scale impact training at higher resolution. The pipeline is trained end-to-end to benefit from simultaneous segmentation refinement performed at each level. It incorporates transfer learning and fine tuning from DDSM-CBIS to INbreast datasets to further improve mass delineations. The comprehensive evaluation provided for high-resolution INbreast images highlights promising model generalizability against standard encoder-decoder strategies.

Effects of positive end-expiratory pressure titration on intestinal oxygenation and perfusion in isoflurane anaesthetised horses.

REASONS FOR PERFORMING STUDY: High airway pressures, necessary to keep equine lungs open, can have a detrimental impact on central and peripheral perfusion. OBJECTIVE: The aim of this study was to assess the effects of stepwise increasing airway pressure recruitment on central and intestinal perfusion and oxygenation during isoflurane anaesthesia in horses. STUDY DESIGN: In vivo experimental study. METHODS: Ten anaesthetised horses were ventilated using intermittent positive pressure ventilation immediately after induction. After 90 min, end-expiratory pressure (PEEP) was increased by steps of 5 cmH2 O every 10 min up to a PEEP of 30 cmH2 O and decreased back to zero maintaining a constant airway pressure difference of 20 cmH2 O. Mean arterial blood pressure (MAP), heart rate, central venous pressure, pulmonary artery pressure, expiratory isoflurane concentration and cardiac output (thermodilution method) were measured. Cardiac index (CI) was calculated. Arterial blood gases were taken to measure arterial partial oxygen pressure (PaO2 ) and calculate arterial oxygen saturation (SaO2 ). Intestinal microperfusion and oxygenation were measured by laser Doppler flowmetry and white-light spectrophotometry. After ventral median laparotomy, a probe was placed on the stomach, jejunum and pelvic flexion of the colon. An ANOVA for repeated measurements and Tukey's post hoc test were used for statistical analysis (α = 5%). RESULTS: Recruitment of the lungs resulted in a significant increase in PaO2 from 201 ± 58 mmHg (baseline) to a maximum of 495 ± 75 mmHg. The CI and MAP decreased continuously with increasing airway pressures. When CI and MAP were 37 ± 9 ml/kg/min and 52 ± 8 mmHg (at PEEP of 25 cmH2 O), respectively, a sudden decrease in intestinal perfusion followed by a delayed decrease in oxygenation occurred. CONCLUSIONS: There was linear correlation between airway pressures and CI and MAP but not between central and gastrointestinal perfusion. Despite improvement of arterial oxygenation the decrease in CI and, therefore, in oxygen delivery PEEP resulted in a decrease in gastrointestinal oxygenation.