Aircraft Engine Performance Monitoring and Diagnostics Based on Deep Convolutional Neural Networks

The rapid advancement of machine-learning techniques has played a significant role in the evolution of engine health management technology. In the last decade, deep-learning methods have received a great deal of attention in many application domains, including object recognition and computer vision. Recently, there has been a rapid rise in the use of convolutional neural networks for rotating machinery diagnostics inspired by their powerful feature learning and classification capability. However, the application in the field of gas turbine diagnostics is still limited. This paper presents a gas turbine fault detection and isolation method using modular convolutional neural networks preceded by a physics-driven performance-trend-monitoring system. The trend-monitoring system was employed to capture performance changes due to degradation, establish a new baseline when it is needed, and generatefault signatures. The fault detection and isolation system was trained to step-by-step detect and classify gas path faults to the component level using fault signatures obtained from the physics part. The performance of the method proposed was evaluated based on different fault scenarios for a three-shaft turbofan engine, under significant measurement noise to ensure model robustness. Two comparative assessments were also carried out: with a single convolutional-neural-network-architecture-based fault classification method and with a deep long short-term memory-assisted fault detection and isolation method. The results obtained revealed the performance of the proposed method to detect and isolate multiple gas path faults with over 96% accuracy. Moreover, sharing diagnostic tasks with modular architectures is seen as relevant to significantly enhance diagnostic accuracy.

Nowcasting for Real-Time COVID-19 Tracking in New York City: An Evaluation Using Reportable Disease Data From Early in the Pandemic (Preprint)

BACKGROUND Nowcasting approaches enhance the utility of reportable disease data for trend monitoring by correcting for delays, but implementation details affect accuracy. OBJECTIVE To support real-time COVID-19 situational awareness, the New York City Department of Health and Mental Hygiene used nowcasting to account for testing and reporting delays. We conducted an evaluation to determine which implementation details would yield the most accurate estimated case counts. METHODS A time-correlated Bayesian approach called Nowcasting by Bayesian Smoothing (NobBS) was applied in real time to line lists of reportable disease surveillance data, accounting for the delay from diagnosis to reporting and the shape of the epidemic curve. We retrospectively evaluated nowcasting performance for confirmed case counts among residents diagnosed during the period from March to May 2020, a period when the median reporting delay was 2 days. RESULTS Nowcasts with a 2-week moving window and a negative binomial distribution had lower mean absolute error, lower relative root mean square error, and higher 95% prediction interval coverage than nowcasts conducted with a 3-week moving window or with a Poisson distribution. Nowcasts conducted toward the end of the week outperformed nowcasts performed earlier in the week, given fewer patients diagnosed on weekends and lack of day-of-week adjustments. When estimating case counts for weekdays only, metrics were similar across days when the nowcasts were conducted, with Mondays having the lowest mean absolute error of 183 cases in the context of an average daily weekday case count of 2914. CONCLUSIONS Nowcasting using NobBS can effectively support COVID-19 trend monitoring. Accounting for overdispersion, shortening the moving window, and suppressing diagnoses on weekends—when fewer patients submitted specimens for testing—improved the accuracy of estimated case counts. Nowcasting ensured that recent decreases in observed case counts were not overinterpreted as true declines and supported officials in anticipating the magnitude and timing of hospitalizations and deaths and allocating resources geographically.

Pharmaco-vigilance of oral MethylErgometrine prescriptions for delayed and secondary postpartum haemorrhage

ObjectivesThe objectives of this study were to quantify the prescription of oral methergin tablets in a busy Women’s Hospital, assess the stated indications for such prescription and highlight the issues and safety profile of Methergin use especially in the postpartum patient.MethodsReview of prescription data for oral Methergin and the corresponding annual figures on primary and secondary postpartum hemorrhage.ResultsOver a period of 5 years, oral Methergin prescriptions for delayed and secondary postpartum hemorrhage constituted less than 1% of the overall prescription in Obstetrics and Gynaecology, which ranged between 1214 and 2085 per year. The numbers were too few to ascertain any relationship with both types of postpartum hemorrhage. Although stated on the relevant Patient Information leaflet, no local or regional guideline on its use exist.ConclusionsSpecific and random trend monitoring of medications for continuing safety profile, risk benefit issues, or unapproved indication, may help in identifying, preventing and mitigating any medication safety matters. Clinical pharmacists in collaboration with physicians are well placed in conducting such pharmacovigilance activities to improve medication safety.

Multicenter comparison of three intraoperative hemoglobin trend monitoring methods

Transfusion decisions are guided by clinical factors and measured hemoglobin (Hb). Time required for blood sampling and analysis may cause Hb measurement to lag clinical conditions, thus continuous intraoperative Hb trend monitoring may provide useful information. This multicenter study was designed to compare three methods of determining intraoperative Hb changes (trend accuracy) to laboratory determined Hb changes. Adult surgical patients with planned arterial catheterization were studied. With each blood gas analysis performed, pulse cooximetry hemoglobin (SpHb) was recorded, and arterial blood Hb was measured by hematology (tHb), arterial blood gas cooximetry (ABGHb), and point of care (aHQHb) analyzers. Hb change was calculated and trend accuracy assessed by modified Bland–Altman analysis. Secondary measures included Hb measurement change direction agreement. Trend accuracy mean bias (95% limits of agreement; g/dl) for SpHb was 0.10 (− 1.14 to 1.35); for ABGHb was − 0.02 (− 1.06 to 1.02); and for aHQHb was 0.003 (− 0.95 to 0.95). Changes more than ± 0.5 g/dl agreed with tHb changes more than ± 0.25 g/dl in 94.2% (88.9–97.0%) SpHb changes, 98.9% (96.1–99.7%) ABGHb changes and 99.0% (96.4–99.7%) aHQHb changes. Sequential changes in SpHb, ABGHb and aHQHb exceeding ± 0.5 g/dl have similar agreement to the direction but not necessarily the magnitude of sequential tHb change. While Hb blood tests should continue to be used to inform transfusion decisions, intraoperative continuous noninvasive SpHb decreases more than − 0.5 g/dl could be a good indicator of the need to measure tHb.