Search

OBJECTIVES: To differentiate invasive lepidic predominant adenocarcinoma (iLPA) from adenocarcinoma in situ (AIS)/minimally invasive adenocarcinoma (MIA) of lung utilizing visual semantic and computer-aided detection (CAD)-based texture features on subjects initially diagnosed as AIS or MIA with CT-guided biopsy. MATERIALS AND METHODS: From 2011 to 2017, all patients with CT-guided biopsy results of AIS or MIA who subsequently underwent resection were identified. CT scan before the biopsy was used to assess visual semantic and CAD texture features, totaling 23 semantic and 95 CAD-based quantitative texture variables. The least absolute shrinkage and selection operator (LASSO) method or forward selection was used to select the most predictive feature and combination of semantic and texture features for detection of invasive lung adenocarcinoma. RESULTS: Among the 33 core needle-biopsied patients with AIS/MIA pathology, 24 (72.7%) had invasive LPA and 9 (27.3%) had AIS/MIA on resection. On CT, visual semantic features included 21 (63.6%) part-solid, 5 (15.2%) pure ground glass, and 7 (21.2%) solid nodules. LASSO selected seven variables for the model, but all were not statistically significant. Volume was found to be statistically significant when assessing the correlation between independent variables using the backward selection technique. The LASSO selected tumor_Perc95, nodule surround, small cyst-like spaces, and volume when assessing the correlation between independent variables. CONCLUSIONS: Lung biopsy results showing noninvasive LPA underestimate invasiveness. Although statistically non-significant, some semantic features showed potential for predicting invasiveness, with septal stretching absent in all noninvasive cases, and solid consistency present in a significant portion of invasive cases.

Coronavirus disease 2019 (COVID-19), is an ongoing issue in certain populations, presenting rapidly worsening pneumonia and persistent symptoms. This study aimed to test the predictability of rapid progression using radiographic scores and laboratory markers and present longitudinal changes. This retrospective study included 218 COVID-19 pneumonia patients admitted at the Chungnam National University Hospital. Rapid progression was defined as respiratory failure requiring mechanical ventilation within one week of hospitalization. Quantitative COVID (QCOVID) scores were derived from high-resolution computed tomography (CT) analyses: (1) ground glass opacity (QGGO), (2) mixed diseases (QMD), and (3) consolidation (QCON), and the sum, quantitative total lung diseases (QTLD). Laboratory data, including inflammatory markers, were obtained from electronic medical records. Rapid progression was observed in 9.6% of patients. All QCOVID scores predicted rapid progression, with QMD showing the best predictability (AUC = 0.813). In multivariate analyses, the QMD score and interleukin(IL)-6 level were important predictors for rapid progression (AUC = 0.864). With >2 months follow-up CT, remained lung lesions were observed in 21 subjects, even after several weeks of negative reverse transcription polymerase chain reaction test. AI-driven quantitative CT scores in conjugation with laboratory markers can be useful in predicting the rapid progression and monitoring of COVID-19.

PURPOSE: To integrate and evaluate an artificial intelligence (AI) system that assists in checking endotracheal tube (ETT) placement on chest x-rays (CXRs) in clinical practice. APPROACH: In clinical use over 17 months, 214 CXR images were ordered to check ETT placement with AI assistance by intensive care unit (ICU) physicians. The system was built on the SimpleMind Cognitive AI platform and integrated into a clinical workflow. It automatically identified the ETT and checked its placement relative to the trachea and carina. The ETT overlay and misplacement alert messages generated by the AI system were compared with radiology reports as the reference. A survey study was also conducted to evaluate usefulness of the AI system in clinical practice. RESULTS: The alert messages indicating that either the ETT was misplaced or not detected had a positive predictive value of 42% (21/50) and negative predictive value of 98% (161/164) based on the radiology reports. In the survey, radiologist and ICU physician users indicated that they agreed with the AI outputs and that they were useful. CONCLUSIONS: The AI system performance in real-world clinical use was comparable to that seen in previous experiments. Based on this and physician survey results, the system can be deployed more widely at our institution, using insights gained from this evaluation to make further algorithm improvements and quality assurance of the AI system.

In-situ marine cloud droplet number concentrations (CDNCs), cloud condensation nuclei (CCN), and CCN proxies, based on particle sizes and optical properties, are accumulated from seven field campaigns: ACTIVATE; NAAMES; CAMP2EX; ORACLES; SOCRATES; MARCUS; and CAPRICORN2. Each campaign involves aircraft measurements, ship-based measurements, or both. Measurements collected over the North and Central Atlantic, Indo-Pacific, and Southern Oceans, represent a range of clean to polluted conditions in various climate regimes. With the extensive range of environmental conditions sampled, this data collection is ideal for testing satellite remote detection methods of CDNC and CCN in marine environments. Remote measurement methods are vital to expanding the available data in these difficult-to-reach regions of the Earth and improving our understanding of aerosol-cloud interactions. The data collection includes particle composition and continental tracers to identify potential contributing CCN sources. Several of these campaigns include High Spectral Resolution Lidar (HSRL) and polarimetric imaging measurements and retrievals that will be the basis for the next generation of space-based remote sensors and, thus, can be utilized as satellite surrogates.

To translate artificial intelligence (AI) algorithms into clinical practice requires generalizability of models to real-world data. One of the main obstacles to generalizability is data shift, a data distribution mismatch between model training and real environments. Explainable AI techniques offer tools to detect and mitigate the data shift problem and develop reliable AI for clinical practice. Most medical AI is trained with datasets gathered from limited environments, such as restricted disease populations and center-dependent acquisition conditions. The data shift that commonly exists in the limited training set often causes a significant performance decrease in the deployment environment. To develop a medical application, it is important to detect potential data shift and its impact on clinical translation. During AI training stages, from premodel analysis to in-model and post hoc explanations, explainability can play a key role in detecting model susceptibility to data shift, which is otherwise hidden because the test data have the same biased distribution as the training data. Performance-based model assessments cannot effectively distinguish the model overfitting to training data bias without enriched test sets from external environments. In the absence of such external data, explainability techniques can aid in translating AI to clinical practice as a tool to detect and mitigate potential failures due to data shift. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

BackgroundIdiopathic pulmonary fibrosis (IPF) is a progressive, irreversible, and usually fatal lung disease of unknown reasons, generally affecting the elderly population. Early diagnosis of IPF is crucial for triaging patients' treatment planning into anti-fibrotic treatment or treatments for other causes of pulmonary fibrosis. However, current IPF diagnosis workflow is complicated and time-consuming, which involves collaborative efforts from radiologists, pathologists, and clinicians and it is largely subject to inter-observer variability.PurposeThe purpose of this work is to develop a deep learning-based automated system that can diagnose subjects with IPF among subjects with interstitial lung disease (ILD) using an axial chest computed tomography (CT) scan. This work can potentially enable timely diagnosis decisions and reduce inter-observer variability.MethodsOur dataset contains CT scans from 349 IPF patients and 529 non-IPF ILD patients. We used 80% of the dataset for training and validation purposes and 20% as the holdout test set. We proposed a two-stage model: at stage one, we built a multi-scale, domain knowledge-guided attention model (MSGA) that encouraged the model to focus on specific areas of interest to enhance model explainability, including both high- and medium-resolution attentions; at stage two, we collected the output from MSGA and constructed a random forest (RF) classifier for patient-level diagnosis, to further boost model accuracy. RF classifier is utilized as a final decision stage since it is interpretable, computationally fast, and can handle correlated variables. Model utility was examined by (1) accuracy, represented by the area under the receiver operating characteristic curve (AUC) with standard deviation (SD), and (2) explainability, illustrated by the visual examination of the estimated attention maps which showed the important areas for model diagnostics.ResultsDuring the training and validation stage, we observe that when we provide no guidance from domain knowledge, the IPF diagnosis model reaches acceptable performance (AUC±SD = 0.93±0.07), but lacks explainability; when including only guided high- or medium-resolution attention, the learned attention maps are not satisfactory; when including both high- and medium-resolution attention, under certain hyperparameter settings, the model reaches the highest AUC among all experiments (AUC±SD = 0.99±0.01) and the estimated attention maps concentrate on the regions of interests for this task. Three best-performing hyperparameter selections according to MSGA were applied to the holdout test set and reached comparable model performance to that of the validation set.ConclusionsOur results suggest that, for a task with only scan-level labels available, MSGA+RF can utilize the population-level domain knowledge to guide the training of the network, which increases both model accuracy and explainability.

BackgroundTo evaluate the anatomic site(s) of initial disease progression in patients with castration resistant metastatic prostate cancer (mCRPC) in the presence or absence of pre-treatment visceral metastases while on systemic therapy.MethodsThis is a retrospective cohort study of mCRPC patients who have baseline and at least one follow up bone scan and CT chest, abdomen and pelvis (CAP). Disease progression was determined by RECIST and/or ≥ 30% increase in automated bone scan lesion area score. Kaplan-Meier plot was used to estimate the median progression free survival and log-rank tests were used to compare anatomic sites.ResultsOf 203 patients, 61 (30%) had pre-treatment visceral metastases. Patients with baseline visceral disease were 1.5 times more likely to develop disease progression (HR = 1.53; 95% CI, 1.03-2.26). Disease progression was a result of worsening bone scan disease (42% (16/38)) versus visceral (32% (12/38)) or lymph node disease (3% (1/38)) by CT or a combination thereof (23% (9/38)). Median time to progression (TTP) did not differ by anatomic location of initial progression (p = 0.86). Development of new lesions occurred in 50% of those visceral patients with soft tissue only progression and was associated with a significantly longer TTP (3.1 months (2.8-4.3 months) than those with worsening of pre-existing lesions (1.8 months (1.6-2.7 months); p = 0.04.ConclusionsPatients with pre-treatment visceral metastases in mCRPC are more likely to experience disease progression of bone disease with the initial anatomic site of progression similar to those without baseline visceral involvement.

Imaging response assessment is a cornerstone of patient care and drug development in oncology. Clinicians/clinical researchers rely on tumor imaging to estimate the impact of new treatments and guide decision making for patients and candidate therapies. This is important in brain cancer, where associations between tumor size/growth and emerging neurological deficits are strong. Accurately measuring the impact of a new therapy on tumor growth early in clinical development, where patient numbers are small, would be valuable for decision making regarding late-stage development activation. Current attempts to measure the impact of a new therapy have limited influence on clinical development, as determination of progression, stability or response does not currently account for individual tumor growth kinetics prior to the initiation of experimental therapies. Therefore, we posit that imaging-based response assessment, often used as a tool for estimating clinical effect, is incomplete as it does not adequately account for growth trajectories or biological characteristics of tumors prior to the introduction of an investigational agent. Here, we propose modifications to the existing framework for evaluating imaging assessment in primary brain tumors that will provide a more reliable understanding of treatment effects. Measuring tumor growth trajectories prior to a given intervention may allow us to more confidently conclude whether there is an anti-tumor effect. This updated approach to imaging-based tumor response assessment is intended to improve our ability to select candidate therapies for later-stage development, including those that may not meet currently sought thresholds for "response" and ultimately lead to identification of effective treatments.

This study aims to examine the current experiences of high school students in computer science (CS) courses and the factors that motivated them to continue their future enrollment. The participants were 603 high school students in grades 9 through 12 in Indiana, all of whom enrolled in at least one CS course during the 2020-2021 academic year. This research revealed that fun and meaningful CS pedagogy, knowledgeable CS teachers, and relevance to their lives and future careers enabled high school students to hold positive experiences in their CS classes. These experiences impacted students to take additional CS courses. In addition to these positive experiences, gender and early exposure to CS emerge as predictors to pursue CS courses. The findings will carry significance for policymakers and educators offering insights to enhance and broaden students' participation and engagement in the CS course.

RATIONALE: The long-acting β2-agonist/long-acting muscarinic antagonist combination indacaterol/glycopyrronium (IND/GLY) elicits bronchodilation, improves symptoms, and reduces exacerbations in COPD. Magnetic resonance imaging (MRI) of the lung with hyperpolarized gas and gadolinium contrast enhancement enables assessment of whole lung functional responses to IND/GLY. OBJECTIVES: The primary objective was assessment of effect of IND/GLY on global ventilated lung volume (%VV) versus placebo in COPD. Lung function, regional ventilation and perfusion in response to IND/GLY were also measured. METHODS: This double-blind, randomized, placebo-controlled, crossover study assessed %VV and pulmonary perfusion in patients with moderate-to-severe COPD after 8 days of once-daily IND/GLY treatment (110/50 µg) followed by 8 days of placebo, or vice versa, using inhaled hyperpolarized 3He gas and gadolinium contrast-enhanced MRI, respectively. Lung function measures including spirometry were performed for each treatment after 8 days. MEASUREMENTS AND MAIN RESULTS: Of 31 patients randomized, 29 completed both treatment periods. IND/GLY increased global %VV versus placebo (61.73% vs. 56.73%, respectively, least squares means treatment difference: 5.00% [90% CI 1.40 to 8.60]; P = 0.025). IND/GLY improved whole lung index of ventilation volume to perfusion volume (V/Q) ratio versus placebo; 94% (90% CI 83 to 105) versus 86% (90% CI 75 to 97; P = 0.047), respectively. IND/GLY showed a trend to improve diffusing capacity for carbon monoxide (DLCO) (+ 0.66 mL/min/mmHg; P = 0.082). By Day 8, forced expiratory volume in 1 s (FEV1) was increased by 0.32 L versus placebo (90% CI 0.26 to 0.38; P

Atmospheric marine particle concentrations impact cloud properties, which strongly impact the amount of solar radiation reflected back into space or absorbed by the ocean surface. While satellites can provide a snapshot of current conditions at the overpass time, models are necessary to simulate temporal variations in both particle and cloud properties. However, poor model accuracy limits the reliability with which these tools can be used to predict future climate. Here, we leverage the comprehensive ocean ecosystem and atmospheric aerosol–cloud dataset obtained during the third deployment of the North Atlantic Aerosols and Marine Ecosystems Study (NAAMES3). Airborne and ship-based measurements were collected in and around a cold-air outbreak during a 3 d (where d stands for day) intensive operations period from 17–19 September 2017. Cold-air outbreaks are of keen interest for model validation because they are challenging to accurately simulate, which is due, in part, to the numerous feedbacks and sub-grid-scale processes that influence aerosol and cloud evolution. The NAAMES observations are particularly valuable because the flight plans were tailored to lie along Lagrangian trajectories, making it possible to spatiotemporally connect upwind and downwind measurements with the state-of-the-art FLEXible PARTicle (FLEXPART) Lagrangian particle dispersion model and then calculate a rate of change in particle properties. Initial aerosol conditions spanning an east–west, closed-cell-to-clear-air transition region of the cold-air outbreak indicate similar particle concentrations and properties. However, despite the similarities in the aerosol fields, the cloud properties downwind of each region evolved quite differently. One trajectory carried particles through a cold-air outbreak, resulting in a decrease in accumulation mode particle concentration (−42 %) and cloud droplet concentrations, while the other remained outside of the cold-air outbreak and experienced an increase in accumulation mode particle concentrations (+62 %). The variable meteorological conditions between these two adjacent trajectories result from differences in the local sea surface temperature in the Labrador Current and surrounding waters, altering the stability of the marine atmospheric boundary layer. Further comparisons of historical satellite observations indicate that the observed pattern occurs annually in the region, making it an ideal location for future airborne Lagrangian studies tracking the evolution of aerosols and clouds over time under cold-air outbreak conditions.

Engineering ethics education typically focuses on decisions by individual engineers and case studies of disasters. This does not reflect the everyday decisions that practicing engineers must make, and neglects the fact that most engineers work on teams rather than alone. The focus on safety and disaster prevention leaves little time for discussing pervasive social impacts of engineering and technology. Our research seeks to fill these gaps by determining how ethical decision making occurs in team settings, how it can be influenced by ethics-focused team members, and to what extent a social context influence consideration of social impacts. Over three years, we observed ethics discussions among teams of engineering students. First, we observed undergraduates during their Capstone Design project. We contrasted typical teams and teams with an additional team member trained and educated in engineering ethics. Second, we observed university research laboratory groups composed of undergraduate students, graduate students and postdocs in their spontaneous and regular conversations. Data analysis suggests that engineering students have a narrow understanding of engineering ethics, and that their explicit and implicit understanding can be in conflict. We also observed that an engineering expert can improve the breadth and depth of conversations. We also observe that engineering students rarely--if ever--discuss ethical or social implications of their work during routine activities. This also points to the benefits of facilitated discussions.