Data-efficient Bayesian learning for radial dynamic MR reconstruction.

BACKGROUND: Cardiac MRI has become the gold-standard imaging technique for assessing cardiovascular morphology and function. In spite of this, its slow data acquisition process presents imaging challenges due to the motion from heartbeats, respiration, and blood flow. In recent studies, deep learning (DL) algorithms have shown promising results for the task of image reconstruction. However, there have been instances where they have introduced artifacts that may be misinterpreted as pathologies or may obscure the detection of pathologies. Therefore, it is important to obtain a metric, such as the uncertainty of the network output, that identifies such artifacts. However, this can be quite challenging for large-scale image reconstruction problems such as dynamic multi-coil non-Cartesian MRI. PURPOSE: To efficiently quantify uncertainties of a physics-informed DL-based image reconstruction method for a large-scale accelerated 2D multi-coil dynamic radial MRI reconstruction problem, and demonstrate the benefits of physics-informed DL over model-agnostic DL in reducing uncertainties while at the same time improving image quality. METHODS: We extended a recently proposed physics-informed 2D U-Net that learns spatio-temporal slices (named XT-YT U-Net), and employed it for the task of uncertainty quantification (UQ) by using Monte Carlo dropout and a Gaussian negative log-likelihood loss function. Our data comprised 2D dynamic MR images acquired with a radial balanced steady-state free precession sequence. The XT-YT U-Net, which allows for training with a limited amount of data, was trained and validated on a dataset of 15 healthy volunteers, and further tested on data from four patients. An extensive comparison between physics-informed and model-agnostic neural networks (NNs) concerning the obtained image quality and uncertainty estimates was performed. Further, we employed calibration plots to assess the quality of the UQ. RESULTS: The inclusion of the MR-physics model of data acquisition as a building block in the NN architecture led to higher image quality (NRMSE: - 33 ± 8.2 % $-33 \pm 8.2 \%$ , PSNR: 6.3 ± 1.3 % $6.3 \pm 1.3 \%$ , and SSIM: 1.9 ± 0.96 % $1.9 \pm 0.96 \%$ ), lower uncertainties ( - 46 ± 8.7 % $-46 \pm 8.7 \%$ ), and, based on the calibration plots, an improved UQ compared to its model-agnostic counterpart. Furthermore, the UQ information can be used to differentiate between anatomical structures (e.g., coronary arteries, ventricle boundaries) and artifacts. CONCLUSIONS: Using an XT-YT U-Net, we were able to quantify uncertainties of a physics-informed NN for a high-dimensional and computationally demanding 2D multi-coil dynamic MR imaging problem. In addition to improving the image quality, embedding the acquisition model in the network architecture decreased the reconstruction uncertainties as well as quantitatively improved the UQ. The UQ provides additional information to assess the performance of different network approaches.

Direct cost analysis of intensive care unit stay in four European countries: applying a standardized costing methodology.

OBJECTIVES: The objective of the present study was to measure and compare the direct costs of intensive care unit (ICU) days at seven ICU departments in Germany, Italy, the Netherlands, and the United Kingdom by means of a standardized costing methodology. METHODS: A retrospective cost analysis of ICU patients was performed from the hospital's perspective. The standardized costing methodology was developed on the basis of the availability of data at the seven ICU departments. It entailed the application of the bottom-up approach for "hotel and nutrition" and the top-down approach for "diagnostics," "consumables," and "labor." RESULTS: Direct costs per ICU day ranged from €1168 to €2025. Even though the distribution of costs varied by cost component, labor was the most important cost driver at all departments. The costs for "labor" amounted to €1629 at department G but were fairly similar at the other departments (€711 ± 115). CONCLUSIONS: Direct costs of ICU days vary widely between the seven departments. Our standardized costing methodology could serve as a valuable instrument to compare actual cost differences, such as those resulting from differences in patient case-mix.

The effect of hospital size and surgical service on case cancellation in elective surgery: results from a prospective multicenter study.

BACKGROUND: Short-term case cancellation causes frustration for anesthesiologists, surgeons, and patients and leads to suboptimal use of operating room (OR) resources. In many facilities, >10% of all cases are cancelled on the day of surgery, thereby causing major problems for OR management and anesthesia departments. The effect of hospital type and service type on case cancellation rate is unclear. METHODS: In 25 hospitals of different types (university hospitals, large community hospitals, and mid- to small-size community hospitals) we studied all elective surgical cases of the following subspecialties over a period of 2 weeks: general surgery, trauma/orthopedics, urology, and gynecology. Case cancellation was defined as any patient who had been scheduled to be operated on the next day, but cancelled after the finalization of the OR plan on the day before surgery. A list of possible cancellation reasons was provided for standardized documentation. RESULTS: A total of 6009 anesthesia cases of 82 different anesthesia services were recorded during the study period. Services in university hospitals had cancellation rates 2.23 (95% confidence interval [CI] = 1.49 to 3.34) times higher than mid- to small-size community hospitals 12.4% (95% CI = 11.0% to 13.8%) versus 5.0% (95% CI = 4.0% to 6.2%). Of the surgical services, general surgical services had a significantly (1.78, 95% CI = 1.25 to 2.53) higher cancellation rate than did gynecology services-11.0% (95% CI = 9.7% to 12.5%) versus 6.6% (95% CI = 5.1% to 8.4%). CONCLUSIONS: When benchmarking cancellation rates among hospitals, comparisons should control for academic institutions having higher incidences of case cancellation than nonacademic hospitals and general surgery services having higher incidences than other services.