Automated intracranial vessel segmentation of 4D flow MRI data in patients with atherosclerotic stenosis using a convolutional neural network.

Introduction: Intracranial 4D flow MRI enables quantitative assessment of hemodynamics in patients with intracranial atherosclerotic disease (ICAD). However, quantitative assessments are still challenging due to the time-consuming vessel segmentation, especially in the presence of stenoses, which can often result in user variability. To improve the reproducibility and robustness as well as to accelerate data analysis, we developed an accurate, fully automated segmentation for stenosed intracranial vessels using deep learning. Methods: 154 dual-VENC 4D flow MRI scans (68 ICAD patients with stenosis, 86 healthy controls) were retrospectively selected. Manual segmentations were used as ground truth for training. For automated segmentation, deep learning was performed using a 3D U-Net. 20 randomly selected cases (10 controls, 10 patients) were separated and solely used for testing. Cross-sectional areas and flow parameters were determined in the Circle of Willis (CoW) and the sinuses. Furthermore, the flow conservation error was calculated. For statistical comparisons, Dice scores (DS), Hausdorff distance (HD), average symmetrical surface distance (ASSD), Bland-Altman analyses, and interclass correlations were computed using the manual segmentations from two independent observers as reference. Finally, three stenosis cases were analyzed in more detail by comparing the 4D flow-based segmentations with segmentations from black blood vessel wall imaging (VWI). Results: Training of the network took approximately 10 h and the average automated segmentation time was 2.2 ± 1.0 s. No significant differences in segmentation performance relative to two independent observers were observed. For the controls, mean DS was 0.85 ± 0.03 for the CoW and 0.86 ± 0.06 for the sinuses. Mean HD was 7.2 ± 1.5 mm (CoW) and 6.6 ± 3.7 mm (sinuses). Mean ASSD was 0.15 ± 0.04 mm (CoW) and 0.22 ± 0.17 mm (sinuses). For the patients, the mean DS was 0.85 ± 0.04 (CoW) and 0.82 ± 0.07 (sinuses), the HD was 8.4 ± 3.1 mm (CoW) and 5.7 ± 1.9 mm (sinuses) and the mean ASSD was 0.22 ± 0.10 mm (CoW) and 0.22 ± 0.11 mm (sinuses). Small bias and limits of agreement were observed in both cohorts for the flow parameters. The assessment of the cross-sectional lumen areas in stenosed vessels revealed very good agreement (ICC: 0.93) with the VWI segmentation but a consistent overestimation (bias ± LOA: 28.1 ± 13.9%). Discussion: Deep learning was successfully applied for fully automated segmentation of stenosed intracranial vasculatures using 4D flow MRI data. The statistical analysis of segmentation and flow metrics demonstrated very good agreement between the CNN and manual segmentation and good performance in stenosed vessels. To further improve the performance and generalization, more ICAD segmentations as well as other intracranial vascular pathologies will be considered in the future.

Nutritional needs, resources, and barriers among unhoused adults cared for by a street medicine organization in Chicago, Illinois: a cross-sectional study.

BACKGROUND: Those experiencing houselessness rely on obtaining food from community organizers and donations. Simultaneously, the houseless face disproportionally high rates of medical conditions that may be affected by diet including diabetes, hypertension, and hyperlipidemia. There is limited literature on the resources and barriers of the houseless community regarding optimal nutrition from an actionable perspective. Further, less data is available on how street medicine organizations may best impact the nutrition of the unhoused they serve. Elucidating this information will inform how organizational efforts may best support the nutrition of the houseless community. METHODS: In partnership with the medical student-run organization, Chicago Street Medicine, at Northwestern University Feinberg School of Medicine, twenty adults experiencing houselessness in Chicago, Illinois participated in the cross-sectional study. A 10-item survey was verbally administered to characterize the participants' daily food intake, food sources, barriers, resources, and nutritional preferences and needs. All data was directly transcribed into REDCap. Descriptive statistics were generated. RESULTS: Individuals consumed a median of 2 snacks and meals per day (IQR: 1-3). No participant consumed adequate servings of every food group, with only one participant meeting the dietary intake requirements for one food group. Participants most often received their food from donations (n = 15), purchasing themselves (n = 11), food pantries (n = 4), and shelters (n = 3). Eleven of nineteen participants endorsed dental concerns as a major barrier to consuming certain foods. Twelve participants had access to a can opener and twelve could heat their meals on a stove or microwave. Seven had access to kitchen facilities where they may prepare a meal. Approximately half of participants had been counseled by a physician to maintain a particular diet, with most related to reducing sugar intake. CONCLUSION: Most houseless participants were unable to acquire a balanced diet and often relied on organizational efforts to eat. Organizations should consider the chronic health conditions, dentition needs, and physical resources and barriers to optimal nutrition when obtaining food to distribute to the unhoused.

Quantitative electroencephalogram in term neonates under different sleep states.

Electroencephalogram (EEG) can be used to assess depth of consciousness, but interpreting EEG can be challenging, especially in neonates whose EEG undergo rapid changes during the perinatal course. EEG can be processed into quantitative EEG (QEEG), but limited data exist on the range of QEEG for normal term neonates during wakefulness and sleep, baseline information that would be useful to determine changes during sedation or anesthesia. We aimed to determine the range of QEEG in neonates during awake, active sleep and quiet sleep states, and identified the ones best at discriminating between the three states. Normal neonatal EEG from 37 to 46 weeks were analyzed and classified as awake, quiet sleep, or active sleep. After processing and artifact removal, total power, power ratio, coherence, entropy, and spectral edge frequency (SEF) 50 and 90 were calculated. Descriptive statistics were used to summarize the QEEG in each of the three states. Receiver operating characteristic (ROC) curves were used to assess discriminatory ability of QEEG. 30 neonates were analyzed. QEEG were different between awake vs asleep states, but similar between active vs quiet sleep states. Entropy beta, delta2 power %, coherence delta2, and SEF50 were best at discriminating awake vs active sleep. Entropy beta had the highest AUC-ROC ≥ 0.84. Entropy beta, entropy delta1, theta power %, and SEF50 were best at discriminating awake vs quiet sleep. All had AUC-ROC ≥ 0.78. In active sleep vs quiet sleep, theta power % had highest AUC-ROC > 0.69, lower than the other comparisons. We determined the QEEG range in healthy neonates in different states of consciousness. Entropy beta and SEF50 were best at discriminating between awake and sleep states. QEEG were not as good at discriminating between quiet and active sleep. In the future, QEEG with high discriminatory power can be combined to further improve ability to differentiate between states of consciousness.

Reducing the environmental impact of mask inductions in children: A quality improvement report.

BACKGROUND: Inhalational anesthetic agents are potent greenhouse gases with global warming potential that far exceed that of carbon dioxide. Traditionally, pediatric inhalation inductions are achieved with a volatile anesthetic delivered to the patient in oxygen and nitrous oxide at high fresh gas flows. While contemporary volatile anesthetics and anesthesia machines allow for a more environmentally conscious induction, practice has not changed. We aimed to reduce the environmental impact of our inhalation inductions by decreasing the use of nitrous oxide and fresh gas flows. METHODS: Through a series of four plan-do-study-act cycles, the improvement team used content experts to demonstrate the environmental impact of the current inductions and to provide practical ways to reduce this, by focusing on nitrous oxide use and fresh gas flows, with visual reminders introduced at point of delivery. The primary measures were the percentage of inhalation inductions that used nitrous oxide and the maximum fresh gas flows/kg during the induction period. Statistical process control charts were used to measure improvement over time. RESULTS: 33 285 inhalation inductions were included over a 20-month period. nitrous oxide use decreased from 80% to <20% and maximum fresh gas flows/kg decreased from a rate of 0.53 L/min/kg to 0.38 L/min/kg, an overall reduction of 28%. Reduction in fresh gas flows was greatest in the lightest weight groups. Induction times and behaviors remained unchanged over the duration of this project. CONCLUSIONS: Our quality improvement group decreased the environmental impact of inhalation inductions and created cultural change within our department to sustain change and foster the pursuit of future environmental efforts.

Comprehensive Anaemia Programme and Personalized Therapies (CAPPT): protocol for a cluster-randomised controlled trial testing the effect women's groups, home counselling and iron supplementation on haemoglobin in pregnancy in southern Nepal.

BACKGROUND: Anaemia in pregnancy remains prevalent in Nepal and causes severe adverse health outcomes. METHODS: This non-blinded cluster-randomised controlled trial in the plains of Nepal has two study arms: (1) Control: routine antenatal care (ANC); (2) Home visiting, iron supplementation, Participatory Learning and Action (PLA) groups, plus routine ANC. Participants, including women in 54 non-contiguous clusters (mean 2582; range 1299-4865 population) in Southern Kapilbastu district, are eligible if they consent to menstrual monitoring, are resident, married, aged 13-49 years and able to respond to questions. After 1-2 missed menses and a positive pregnancy test, consenting women < 20 weeks' gestation, who plan to reside locally for most of the pregnancy, enrol into trial follow-up. Interventions comprise two home-counselling visits (at 12-21 and 22-26 weeks' gestation) with iron folic acid (IFA) supplement dosage tailored to women's haemoglobin concentration, plus monthly PLA women's group meetings using a dialogical problem-solving approach to engage pregnant women and their families. Home visits and PLA meetings will be facilitated by auxiliary nurse midwives. The hypothesis is as follows: Haemoglobin of women at 30 ± 2 weeks' gestation is ≥ 0.4 g/dL higher in the intervention arm than in the control. A sample of 842 women (421 per arm, average 15.6 per cluster) will provide 88% power, assuming SD 1.2, ICC 0.09 and CV of cluster size 0.27. Outcomes are captured at 30 ± 2 weeks gestation. Primary outcome is haemoglobin concentration (g/dL). Secondary outcomes are as follows: anaemia prevalence (%), mid-upper arm circumference (cm), mean probability of micronutrient adequacy (MPA) and number of ANC visits at a health facility. Indicators to assess pathways to impact include number of IFA tablets consumed during pregnancy, intake of energy (kcal/day) and dietary iron (mg/day), a score of bioavailability-enhancing behaviours and recall of one nutrition knowledge indicator. Costs and cost-effectiveness of the intervention will be estimated from a provider perspective. Using constrained randomisation, we allocated clusters to study arms, ensuring similarity with respect to cluster size, ethnicity, religion and distance to a health facility. Analysis is by intention-to-treat at the individual level, using mixed-effects regression. DISCUSSION: Findings will inform Nepal government policy on approaches to increase adherence to IFA, improve diets and reduce anaemia in pregnancy. TRIAL REGISTRATION: ISRCTN 12272130 .

Strategies for responding to the COVID-19 pandemic in a rural health system in New York state.

Research and media reports about coronavirus disease 2019 (COVID-19) have largely focused on urban areas due to their high caseloads. However, the COVID-19 pandemic presents distinct and under-recognized challenges to rural areas. This report describes the challenges faced by Bassett Healthcare Network (BHN), a health network in rural upstate New York, and the strategies BHN devised in response. The response to COVID-19 at BHN focused on 4 strategies: (1) Expansion of intensive-care capacity. (2) Redeployment and retraining of workforce. (3) Provision of COVID-19 information, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral testing, and appropriate follow-up for a geographically dispersed population. (4) Coordination of the response to the pandemic across a large, diverse organization. Rural health systems and hospitals can take steps to address the specific challenges posed by the COVID-19 pandemic in their communities. We believe that the strategies BHN employed to adapt to COVID-19 may be useful to other rural health systems. More research is needed to determine which strategies have been most effective in responding to the pandemic in other rural settings.

Does isolated traumatic subarachnoid hemorrhage merit a lower intensity level of observation than other traumatic brain injury?

Evidence is emerging that isolated traumatic subarachnoid hemorrhage (ITSAH) may be a milder form of traumatic brain injury (TBI). If true, ITSAH may not benefit from intensive care unit (ICU) admission, which would, in turn, decrease resource utilization. We conducted a retrospective review of all TBI admissions to our institution between February 2010 and November 2012 to compare the presentation and clinical course of subjects with ITSAH to all other TBI. We then performed descriptive statistics on the subset of ITSAH subjects presenting with a Glasgow Coma Score (GCS) of 13-15. Of 698 subjects, 102 had ITSAH and 596 had any other intracranial hemorrhage pattern. Compared to all other TBI, ITSAH had significantly lower injury severity scores (p<0.0001), lower head abbreviated injury scores (p<0.0001), higher emergency department GCS (p<0.0001), shorter ICU stays (p=0.007), higher discharge GCS (p=0.005), lower mortality (p=0.003), and significantly fewer head computed tomography scans (p<0.0001). Of those ITSAH subjects presenting with a GCS of 13-15 (n=77), none underwent placement of an intracranial monitor or craniotomy. One subject (1.3%) demonstrated a change in exam (worsened headache and dizziness) concomitant with a progression of his intracranial injury. His symptoms resolved with readmission to the ICU and continued observation. Our results suggest that ITSAH are less-severe brain injuries than other TBI. ITSAH patients with GCS scores of 13-15 demonstrate low rates of clinical progression, and when progression occurs, it resolves without further intervention. This subset of TBI patients does not appear to benefit from ICU admission.