Synthetic OCT-A blood vessel maps using fundus images and generative adversarial networks.

Vessel segmentation in fundus images permits understanding retinal diseases and computing image-based biomarkers. However, manual vessel segmentation is a time-consuming process. Optical coherence tomography angiography (OCT-A) allows direct, non-invasive estimation of retinal vessels. Unfortunately, compared to fundus images, OCT-A cameras are more expensive, less portable, and have a reduced field of view. We present an automated strategy relying on generative adversarial networks to create vascular maps from fundus images without training using manual vessel segmentation maps. Further post-processing used for standard en face OCT-A allows obtaining a vessel segmentation map. We compare our approach to state-of-the-art vessel segmentation algorithms trained on manual vessel segmentation maps and vessel segmentations derived from OCT-A. We evaluate them from an automatic vascular segmentation perspective and as vessel density estimators, i.e., the most common imaging biomarker for OCT-A used in studies. Using OCT-A as a training target over manual vessel delineations yields improved vascular maps for the optic disc area and compares to the best-performing vessel segmentation algorithm in the macular region. This technique could reduce the cost and effort incurred when training vessel segmentation algorithms. To incentivize research in this field, we will make the dataset publicly available to the scientific community.

Antibiotic postprescribing modification opportunities among nursing home residents treated for urinary tract infection.

OBJECTIVE: To characterize opportunities to postprescriptively modify antibiotic prescriptions initiated for treatment of suspected urinary tract infection (UTI) in nursing homes. DESIGN: Cross-sectional cohort study. METHODS: Data from the health records of residents treated for UTI between 2013 and 2014 in 5 Wisconsin nursing homes were abstracted using a structured approach. Explicit definitions were used to identify whether the prescribed antibiotic could have been stopped, shortened, or changed to a nonfluoroquinolone alternative. Antibiotic treatments appropriately modified by study nursing home providers in real time were not considered modifiable. Identification of >1 potential modification opportunity (eg, stop and shorten) per antibiotic treatment event was permitted. RESULTS: In total, 356 eligible antibiotic treatment courses among 249 unique residents were identified. Only 59 antibiotic courses prescribed for treatment of suspected UTI (16.6%) were not amenable to any modification. Discontinuation of treatment due to lack of signs or symptoms of infection was the most frequently identified potential modification opportunity (66.2%). Although less common, substantial numbers of antibiotic treatment courses were potentially amenable to shortening (34%) or agent change (19%) modifications. If applied in concert at 72 hours after antibiotic initiation, stop and shorten modifications could eradicate up to 1,326 avoidable antibiotic days, and change modifications could remove a 32 remaining avoidable fluoroquinolone days. CONCLUSIONS: Substantial opportunity exists to enhance the quality of antibiotic prescribing for treatment of suspected UTI in nursing homes through postprescriptive review interventions. Additional studies examining how to best design and implement postprescriptive review interventions in nursing homes are needed.

Detection of Stroke with Retinal Microvascular Density and Self-Supervised Learning Using OCT-A and Fundus Imaging.

Acute cerebral stroke is a leading cause of disability and death, which could be reduced with a prompt diagnosis during patient transportation to the hospital. A portable retina imaging system could enable this by measuring vascular information and blood perfusion in the retina and, due to the homology between retinal and cerebral vessels, infer if a cerebral stroke is underway. However, the feasibility of this strategy, the imaging features, and retina imaging modalities to do this are not clear. In this work, we show initial evidence of the feasibility of this approach by training machine learning models using feature engineering and self-supervised learning retina features extracted from OCT-A and fundus images to classify controls and acute stroke patients. Models based on macular microvasculature density features achieved an area under the receiver operating characteristic curve (AUC) of 0.87-0.88. Self-supervised deep learning models were able to generate features resulting in AUCs ranging from 0.66 to 0.81. While further work is needed for the final proof for a diagnostic system, these results indicate that microvasculature density features from OCT-A images have the potential to be used to diagnose acute cerebral stroke from the retina.

Total retinal thickness: a neglected factor in the evaluation of inner retinal thickness.

AIM: To determine whether macular retinal nerve fibre layer (mRNFL) and ganglion cell-inner plexiform layer (GC-IPL) thicknesses vary by ethnicity after accounting for total retinal thickness. METHODS: We included healthy participants from the UK Biobank cohort who underwent macula-centred spectral domain-optical coherence tomography scans. mRNFL and GC-IPL thicknesses were determined for groups from different self-reported ethnic backgrounds. Multivariable regression models adjusting for covariables including age, gender, ethnicity and refractive error were built, with and without adjusting for total retinal thickness. RESULTS: 20237 participants were analysed. Prior to accounting for total retinal thickness, mRNFL thickness was on average 0.9 µm (-1.2, -0.6; p<0.001) lower among Asians and 1.5 µm (-2.3, -0.6; p<0.001) lower among black participants compared with white participants. Prior to accounting for total retinal thickness, the average GC-IPL thickness was 1.9 µm (-2.5, -1.4; p<0.001) lower among Asians compared with white participants, and 2.4 µm (-3.9, -1.0; p=0.001) lower among black participants compared with white participants. After accounting for total retinal thickness, the layer thicknesses were not significantly different among ethnic groups. When considered as a proportion of total retinal thickness, mRNFL thickness was ~0.1 and GC-IPL thickness was ~0.2 across age, gender and ethnic groups. CONCLUSIONS: The previously reported ethnic differences in layer thickness among groups are likely driven by differences in total retinal thickness. Our results suggest using layer thickness ratio (retinal layer thicknesses/total retinal thickness) rather than absolute thickness values when comparing retinal layer thicknesses across groups.

Effect of antibiotic time-outs on modification of antibiotic prescriptions in nursing homes.

BACKGROUND: Antibiotic overuse and misuse is a common problem in nursing homes. Antibiotic time-out (ATO) interventions have led to improvements in antibiotic uses in hospitals, but their impact in nursing homes remain understudied. OBJECTIVE: To evaluate the impact of a stewardship intervention, promoting use of ATOs on the frequency and types of antibiotic change events (ACEs) in nursing homes. DESIGN: Controlled before-and-after intervention study. SETTING: Nursing homes in Wisconsin and Pennsylvania. METHOD: Data on antibiotic prescriptions in 11 nursing homes were collected for 25 months. We categorized ACEs as (1) early discontinuation, (2) class modification, or (3) administration modification. Class modification ACEs were further classified based on whether the change narrowed, expanded, or had no effect on bacterial spectrum coverage. Analyses were performed using a difference-in-difference (DiD) approach. RESULT: Of 2,647 antibiotic events initiated in study nursing homes, 376 (14.2%) were associated with an ACE. The overall proportion of ACEs did not significantly differ between intervention and control nursing homes. Early discontinuation ACEs increased in intervention nursing homes (DiD, 2.5%; P = .01), primarily affecting residents initiated on broad-spectrum antibiotics (DiD, 2.9%; P < .01). Class modification ACEs decreased in intervention nursing homes but remained unchanged in control nursing homes. CONCLUSION: The impact of an ATO intervention in study nursing homes was mixed with increases in early discontinuation ACEs offset by reductions in class modification ACEs. More research on the potential value of ATO interventions in nursing homes is warranted.

Validation of a short form Wisconsin Upper Respiratory Symptom Survey (WURSS-21).

BACKGROUND: The Wisconsin Upper Respiratory Symptom Survey (WURSS) is an illness-specific health-related quality-of-life questionnaire outcomes instrument. OBJECTIVES: Research questions were: 1) How well does the WURSS-21 assess the symptoms and functional impairments associated with common cold? 2) How well can this instrument measure change over time (responsiveness)? 3) What is the minimal important difference (MID) that can be detected by the WURSS-21? 4) What are the descriptive statistics for area under the time severity curve (AUC)? 5) What sample sizes would trials require to detect MID or AUC criteria? 6) What does factor analysis tell us about the underlying dimensional structure of the common cold? 7) How reliable are items, domains, and summary scores represented in WURSS? 8) For each of these considerations, how well does the WURSS-21 compare to the WURSS-44, Jackson, and SF-8? STUDY DESIGN AND SETTING: People with Jackson-defined colds were recruited from the community in and around Madison, Wisconsin. Participants were enrolled within 48 hours of first cold symptom and monitored for up to 14 days of illness. Half the sample filled out the WURSS-21 in the morning and the WURSS-44 in the evening, with the other half reversing the daily order. External comparators were the SF-8, a 24-hour recall general health measure yielding separate physical and mental health scores, and the eight-item Jackson cold index, which assesses symptoms, but not functional impairment or quality of life. RESULTS: In all, 230 participants were monitored for 2,457 person-days. Participants were aged 14 to 83 years (mean 34.1, SD 13.6), majority female (66.5%), mostly white (86.0%), and represented substantive education and income diversity. WURSS-21 items demonstrated similar performance when embedded within the WURSS-44 or in the stand-alone WURSS-21. Minimal important difference (MID) and Guyatt's responsiveness index were 10.3, 0.71 for the WURSS-21 and 18.5, 0.75 for the WURSS-44. Factorial analysis suggested an eight dimension structure for the WURSS-44 and a three dimension structure for the WURSS-21, with composite reliability coefficients ranging from 0.87 to 0.97, and Cronbach's alpha ranging from 0.76 to 0.96. Both WURSS versions correlated significantly with the Jackson scale (W-21 R=0.85; W-44 R=0.88), with the SF-8 physical health (W-21 R=-0.79; W-44 R=-0.80) and SF-8 mental health (W-21 R=-0.55; W-44 R=-0.60). CONCLUSION: The WURSS-44 and WURSS-21 perform well as illness-specific quality-of-life evaluative outcome instruments. Construct validity is supported by the data presented here. While the WURSS-44 covers more symptoms, the WURSS-21 exhibits similar performance in terms of reliability, responsiveness, importance-to-patients, and convergence with other measures.