Survivors of critical illness demonstrate skeletal muscle wasting with associated functional impairment. OBJECTIVE To perform a comprehensive prospective characterization of skeletal muscle wasting, defining the pathogenic roles of altered protein synthesis and breakdown. DESIGN, SETTING, AND PARTICIPANTS Sixty-three critically ill patients (59% male; mean age: 54.7 years [95% CI, 50.0-59.6 years]) with an Acute Physiology and Chronic Health Evaluation II score of 23.5 (95% CI, 21.9-25.2) were prospectively recruited within 24 hours following intensive care unit (ICU) admission from August 2009 to April 2011 at a university teaching and a community hospital in England. Patients were recruited if older than 18 years and were anticipated to be intubated for longer than 48 hours, to spend more than 7 days in critical care, and to survive ICU stay. MAIN OUTCOMES AND MEASURES Muscle loss was determined through serial ultrasound measurement of the rectus femoris cross-sectional area (CSA) on days 1, 3, 7, and 10. In a subset of patients, the fiber CSA area was quantified along with the ratio of protein to DNA on days 1 and 7. Histopathological analysis was performed. In addition, muscle protein synthesis, breakdown rates, and respective signaling pathways were characterized. RESULTS There were significant reductions in the rectus femoris CSA observed at day 10 (−17.7% [95% CI, −20.9% to −4.8%]; P < .001). In the 28 patients assessed by all 3 measurement methods on days 1 and 7, the rectus femoris CSA decreased by 10.3% (95% CI, 6.1% to 14.5%), the fiber CSA by 17.5% (95% CI, 5.8% to 29.3%), and the ratio of protein to DNA by 29.5% (95% CI, 13.4% to 45.6%). Decrease in the rectus femoris CSA was greater in patients who experienced multiorgan failure compared with single organ failure by day 7 (−15.7% [95% CI, −19.1% to −12.4%] vs −3.0% [95% CI, −10.5% to 4.6%], P < .001), even by day 3 (−8.7% [95% CI, −13.7% to −3.6%] vs −1.8% [95% CI, −7.3% to 3.8%], respectively; P = .03). Myofiber necrosis occurred in 20 of 37 patients (54.1%). Protein synthesis measured by the muscle protein fractional synthetic rate was depressed in patients on day 1

No abstract

The volume and complexity of diagnostic imaging is increasing at a pace faster than the availability of human expertise to interpret it. Artificial intelligence has shown great promise in classifying two-dimensional photographs of some common diseases and typically relies on databases of millions of annotated images. Until now, the challenge of reaching the performance of expert clinicians in a real-world clinical pathway with three-dimensional diagnostic scans has remained unsolved. Here, we apply a novel deep learning architecture to a clinically heterogeneous set of three-dimensional optical coherence tomography scans from patients referred to a major eye hospital. We demonstrate performance in making a referral recommendation that reaches or exceeds that of experts on a range of sight-threatening retinal diseases after training on only 14,884 scans. Moreover, we demonstrate that the tissue segmentations produced by our architecture act as a device-independent representation; referral accuracy is maintained when using tissue segmentations from a different type of device. Our work removes previous barriers to wider clinical use without prohibitive training data requirements across multiple pathologies in a real-world setting.

No abstract