metabolic inefficiency, which could negatively affect the sustainability of dairy systems.Hit-and-run crashes not only degrade the morality, but also result in delays of medical services provided to victims. However, class imbalance problem exists as the number of hit-and-run crashes is much smaller than that of non-hit-and-run crashes. The missing label problem also exists in the crash analysis due to reasons like data barrier such that the information hidden in the unlabelled samples has not been effectively utilised. In this paper, a cost-sensitive semi-supervised logistic regression (CS3LR) model is proposed for hit-and-run analysis, in order to tackle class-imbalanced data distribution and missing label problem, based on the crash dataset of Victorian, Australia (2013-2019). By performing label estimation with logistic regression jointly utilising both labelled and unlabelled data with pseudo labels in a well-designed cost-sensitive semi-supervised maximum likelihood framework, the proposed model can obtain an unbiased likelihood parameter for hit-and-run prediction and analysis. Comparing the experimental results of CS3LR model with two logistic regression models and seven machine learning methods, better performance of CS3LR model is demonstrated. The most significant contributing factors to hit-and-run crashes extracted by CS3LR with only 10% labelled data show a high degree of consistency with the true contributing factors obtained by the supervised cost-sensitive logistic regression with complete hit-and-run labels. The effects of class-weighted ratio and hyper-parameter λ on the performance of hit-and-run crash prediction model have also been analysed. The results can further provide recommendations and implications on the policies and counter-measures for preventing hit-and-run collisions and crimes. The methodology proposed in this paper can also be employed to analyse crash data with other types of missing labels, such as crash severity. To evaluate the findings and compare the diagnostic yield of computed tomography (CT) versus magnetic resonance imaging (MRI) in children with bilateral sensorineural hearing loss (BSNHL). Multi-institutional retrospective review. Three tertiary referral centers. A multi-institutional retrospective chart review was performed in patients under the age of 18 years with BSNHL (diagnosis codes 389.00-389.22) who underwent both CT and MRI from 2010 to 2012. An abnormal imaging finding was defined as an abnormality of the cochleovestibular or central nervous system known to directly correlate with sensorineural hearing loss. Diagnostic yield of CT versus MRI was compared by McNemar's test. Of 2081 charts reviewed, 313 patients met inclusion criteria. The diagnostic yield of CT and MRI were 25% and 18%, respectively. Approximately one-quarter of patients had an abnormal finding on CT or MRI. The concordance rate was 92%. CT was more likely to yield an abnormal finding versus MRI (p-value=0.0001). The most common findings on CT were cochlear and semicircular canal abnormalities. On MRI, the most common findings were cochlear nerve aplasia/hypoplasia and semicircular canal abnormalities. This study evaluates and directly compares the diagnostic yield of CT versus MRI for pediatric BSNHL. While both modalities have important and often complementary diagnostic utility, CT had superior diagnostic yield in identifying abnormalities associated with BSNHL.This study evaluates and directly compares the diagnostic yield of CT versus MRI for pediatric BSNHL. While both modalities have important and often complementary diagnostic utility, CT had superior diagnostic yield in identifying abnormalities associated with BSNHL. Cannabis use and cannabis use disorders are increasing in prevalence, including among pregnant women. The objective was to evaluate the association of a cannabis-related diagnosis (CRD) in pregnancy and adverse maternal and infant outcomes. We queried an administrative birth cohort of singleton deliveries in California between 2011-2017 linked to maternal and infant hospital discharge records. We classified pregnancies with CRD from International Classification of Disease codes. We identified nicotine and other substance-related diagnoses (SRD) in the same manner. Outcomes of interest included maternal (hypertensive disorders) and infant (prematurity, small for gestational age, NICU admission, major structural malformations) adverse outcomes. From 3,067,069 pregnancies resulting in live births, 29,112 (1.0 %) had a CRD. CRD was associated with an increased risk of all outcomes studied; the strongest risks observed were for very preterm birth (aRR 1.4, 95 % CI 1.3, 1.6) and small for gestational age (aRR 1.4, 95 % CI 1.3, 1.4). (S)-Glutamic acid When analyzed with or without co-exposure diagnoses, CRD alone conferred increased risk for all outcomes compared to no use. The strongest effects were seen for CRD with other SRD (preterm birth aRR 2.3, 95 % CI 2.2, 2.5; very preterm birth aRR 2.6, 95 % CI 2.3, 3.0; gastrointestinal malformations aRR 2.0, 95 % CI 1.6, 2.6). The findings were generally robust to unmeasured confounding and misclassification analyses. CRD in pregnancy was associated with increased risk of adverse maternal and infant outcomes. Providing education and effective treatment for women with a CRD during prenatal care may improve maternal and infant health.CRD in pregnancy was associated with increased risk of adverse maternal and infant outcomes. Providing education and effective treatment for women with a CRD during prenatal care may improve maternal and infant health. Availability of medications for opioid use disorder (MOUD) has increased during the past two decades but treatment retention and adherence remain low. This study aimed to measure the impact of out-of-pocket buprenorphine cost on treatment retention and adherence among US commercially insured patients. Medical payment records from IBM MarketScan were analyzed for 6,439 adults age 18-64 years with commercial insurance who initiated buprenorphine treatment during January 1, 2016 to June 30, 2017. Regression models analyzed the relationship between patients' average daily out-of-pocket buprenorphine cost and buprenorphine retention (at least 80 % days covered by buprenorphine) at three different thresholds (180, 360, and 540 days) and adherence (the number of days of buprenorphine coverage) within each retention threshold. Models controlled for patient demographic and clinical characteristics including age, sex, presence of other substance use disorders, psychiatric and pain diagnoses, and receipt of prescription medications.