Development and deployment of a histopathology-based deep learning algorithm for patient prescreening in a clinical trial.

Accurate identification of genetic alterations in tumors, such as Fibroblast Growth Factor Receptor, is crucial for treating with targeted therapies; however, molecular testing can delay patient care due to the time and tissue required. Successful development, validation, and deployment of an AI-based, biomarker-detection algorithm could reduce screening cost and accelerate patient recruitment. Here, we develop a deep-learning algorithm using >3000 H&E-stained whole slide images from patients with advanced urothelial cancers, optimized for high sensitivity to avoid ruling out trial-eligible patients. The algorithm is validated on a dataset of 350 patients, achieving an area under the curve of 0.75, specificity of 31.8% at 88.7% sensitivity, and projected 28.7% reduction in molecular testing. We successfully deploy the system in a non-interventional study comprising 89 global study clinical sites and demonstrate its potential to prioritize/deprioritize molecular testing resources and provide substantial cost savings in the drug development and clinical settings.

Frontline Nurse Feedback During the Development of a System to Track Cleaning of Portable Medical Equipment.

As part of the development and testing of an innovative technology for tracking disinfection of portable medical equipment, end-user feedback was obtained during an initial trial on two acute care hospital units. The disinfection tracking device was installed on the computers-on-wheels and vital signs machines. Each device had the capability of detecting a cleaning event, reporting the event to an online database, and displaying the time since last cleaning event on a visual display. End-user feedback regarding functionality, usefulness of information provided, and impact on workflow was obtained by survey and facilitated group discussions. Seventeen frontline nurses completed the anonymous survey, and 22 participated in the facilitated group discussions. End users found the system functionally easy to use and the information about time since last cleaning useful and reported minimum disruption of workflow. Functionality of the system was confirmed by consistency between recorded and self-reported cleaning patterns. Managers found the data on cleaning of portable medical equipment helpful in validating compliance with hospital equipment cleaning policy. Frontline staff expressed appreciation for technology that helps them and improves outcomes but also discussed concerns about the potential for technology that creates extra work and disruption in the busy frontline nursing care delivery environment. Nurses were appreciative of opportunities to provide feedback and input into efforts to develop and introduce technology. Recorded cleaning events coincided with self-reported equipment cleaning patterns and illustrated that the device efficiently collects information deemed useful by the end user.

Number of steps and systolic blood pressure: Do work and leisure matter?

The number of steps per day influences blood pressure and health. However, the association between steps at work and leisure and blood pressure is unknown. Thus, we aimed to investigate the association between the domain-specific number of steps and systolic blood pressure. A thigh-worn accelerometer was used to measure the steps of 694 workers over 1-5 consecutive days, separated into work and leisure domains using a self-reported diary. We linearly regressed steps at work, leisure and total day against systolic blood pressure, adjusting for age, sex, body mass index, smoking, and antihypertensive medication. Additionally, we stratified the analyses on job type (blue-collar or white-collar). The results of this cross-sectional analysis indicated a beneficial association between the number of steps (per 2000-step interval) and systolic blood pressure for the total day (-0.5 mmHg; -1.0 to -0.8, 95% CI, p < 0.05) and work (-0.9 mmHg; -1.5 to -0.4, 95% CI, p < 0.05), but not for leisure (+0.1 mmHg; -0.7 to 0.9, 95% CI, p = 0.75). Blue-collar workers took almost twice as many steps at work (9143 ± SD3837) as white-collar workers (5863 ± SD3565) and, after stratification on job type, we observed a beneficial association between the number of steps at work and systolic blood pressure among blue-collar workers (-1.1 mmHg; -1.7 to -0.4, 95% CI, p < 0.05), but not for white-collar workers (-0.3 mmHg; -1.7 to 1.1, 95% CI, p = 0.7). These findings indicate that the number of steps at work, particularly among blue-collar workers, is beneficially associated with systolic blood pressure. Such findings support the potential of work (re)design to promote walking to improve blood pressure.

Comparison of physical behavior estimates from three different thigh-worn accelerometers brands: a proof-of-concept for the Prospective Physical Activity, Sitting, and Sleep consortium (ProPASS).

BACKGROUND: Pooling data from thigh-worn accelerometers across multiple studies has great potential to advance evidence on the health benefits of physical activity. This requires harmonization of information on body postures, physical activity types, volumes and time patterns across different brands of devices. The aim of this study is to compare the physical behavior estimates provided by three different brands of thigh-worn accelerometers. METHODS: Twenty participants volunteered for a 7-day free-living measurement. Three accelerometers - ActiGraph GT3X+, Axivity AX3 and ActivPAL Micro4 - were randomly placed in a vertical line on the midsection of the right thigh. Raw data from each accelerometer was processed and classified into 8 physical activities and postures using the Acti4 software. Absolute differences between estimates and the respective coefficient of variation (CV) were calculated. RESULTS: We observed very minor differences between physical behavior estimates from three different accelerometer brands. When averaged over 24 h (1,440 min), the absolute difference (CV) between accelerometers were: 1.2 mins (0.001) for lying/sitting, 3.4 mins (0.02) for standing, 3.5 mins (0.06) for moving, 1.9 mins (0.03) for walking, 0.1 mins (0.19) for running, 1.2 mins (0.19) for stair climbing, 1.9 mins (0.07) for cycling. Moreover, there was an average absolute difference of 282 steps (0.03) per 24 h. CONCLUSIONS: Physical behaviors were classified with negligible difference between the accelerometer brands. These results support harmonization of data from different thigh-worn accelerometers across multiple cohorts when analyzed in an identical manner.

Directed evolution of a non-heme-iron-dependent extradiol catechol dioxygenase: identification of mutants with intradiol oxidative cleavage activity.

The non-heme-iron(II)-dependent extradiol catechol dioxygenases catalyse the oxidative cleavage of substituted catechols found on bacterial aromatic degradation pathways. The reaction mechanism of the extradiol dioxygenases is believed to proceed through the same proximal hydroperoxide intermediate as the iron(III)-dependent intradiol catechol dioxygenases. Directed evolution was carried out on members of the class III extradiol catechol dioxygenases, by using 1) error-prone polymerase chain reaction, 2) a primer-based cross-over method; the mutant dioxygenases were then screened for their ability to process a range of substituted catechols. Several mutant enzymes were found to show higher activity towards certain substituted catechols, including 4-chlorocatechol, and higher affinity for the iron(II) cofactor. Two mutants isolated from error-prone PCR of Escherichia coli MhpB (mutants R215W and K273R) were found to produce a mixture of extradiol and intradiol cleavage products, as detected by GC-MS and 1H NMR spectroscopy. The residue corresponding to K273 in protocatechuate 4,5-dioxygenase (LigAB), Val244, is located approximately 12 A from the iron(II) centre, but close to the putative dioxygen channel; R215 is found on a sequence loop not present in LigB.