a data augmentation strategy to narrow the robustness gap between expert radiologists and deep learning classifiers.

Purpose: Successful performance of deep learning models for medical image analysis is highly dependent on the quality of the images being analysed. Factors like differences in imaging equipment and calibration, as well as patient-specific factors such as movements or biological variability (e.g.

Active aspiration versus simple compression to remove residual gas from the abdominal cavity after laparoscopic cholecystectomy: a randomized clinical trial.

Purpose: After laparoscopic surgical procedures, residual gas in the abdominal cavity can cause post-operative pain, which is commonly located in the shoulder region. Previous studies suggested that post-laparoscopy pain can be prevented by active suctioning of intraabdominal gas at the end of surgery.

Methods: This randomized controlled trial (registered at DRKS 00,023,286) compared active suctioning versus manual compression in their ability to reduce pain after laparoscopic cholecystectomy.

Suppression of polygalacturonase gene expression in the phytopathogenic fungus Ophiostoma novo-ulmi by RNA interference.

The fungal pathogen, Ophiostomo novo-ulmi, has been responsible for the rapid decline of American elm (Ulmus americana) across North America and remains a serious threat to surviving elm populations. The production of pectinolytic polygalacturonase enzymes has been implicated as a virulence factor for many fungal pathogens, including O. novo-ulmi.

Measuring surgeon performance of sentinel lymph node biopsy in breast cancer treatment by cumulative sum analysis.

Background: This study was performed to determine if surgeons' performance of sentinel lymph node biopsy (SLNB) for breast cancer varied with time and to devise a method to continuously evaluate that performance.

Methods: We retrospectively examined the SLNB experience of 13 community surgeons performing 765 SLNBs and 579 concomitant axillary dissections. False-negative rates (FNRs) were assessed for individuals and cohorts defined by caseload.

Response allocation to concurrent fixed-ratio reinforcement schedules with work requirements by adults with mental retardation and typical preschool children.

The present experiments examined the effect of work requirements in combination with reinforcement schedule on the choice behavior of adults with mental retardation and preschool children. The work requirements of age-appropriate tasks (i.e.