Tag: statistics

PLoS One. 2022 Dec 27;17(12):e0279472. doi: 10.1371/journal.pone.0279472. eCollection 2022.

ABSTRACT

The automatic sorting of construction waste (CW) is an essential procedure in the field of CW recycling due to its remarkable efficiency and safety. The classification of CW is the primary task that guides automatic and precise sorting. In our work, a new method of CW classification based on two-level fusion is proposed to promote classification performance. First, statistical histograms are used to obtain global hue information and local oriented gradients, which are called the hue histogram (HH) and histogram of oriented gradients (HOG), respectively. To fuse these visual features, a bag-of-visual-words (BoVW) method is applied to code HOG descriptors in a CW image as a vector, and this process is named B-HOG. Then, based on feature-level fusion, we define a new feature to combine HH and B-HOG, which represent the global and local visual characteristics of an object in a CW image. Furthermore, two base classifiers are used to learn the information from the color feature space and the new feature space. Based on decision-level fusion, we propose a joint decision-making model to combine the decisions from the two base classifiers for the final classification result. Finally, to verify the performance of the proposed method, we collect five types of CW images as the experimental data set and use these images to conduct experiments on three different base classifiers. Moreover, we compare this method with other extant methods. The results demonstrate that our method is effective and feasible.

PMID:36574416 | DOI:10.1371/journal.pone.0279472

PLoS One. 2022 Dec 27;17(12):e0279518. doi: 10.1371/journal.pone.0279518. eCollection 2022.

ABSTRACT

PURPOSE: To assess the rate of pseudophakic cystoid macular edema (pCME) in uneventful cataract surgery in surgeons in training vs experienced surgeons and to analyze the rate of pCME according to surgeon’s sex.

METHODS: Medical reports post phacoemulsification between 2010 and 2018 at the Department of Ophthalmology, Medical University of Graz, Austria, were reviewed for pCME. A running lifetime number of preceding cataract surgeries was used to express hands-on experience. A cut-off number of 300 surgeries was defined to distinguish between surgeons in training and experienced surgeons. Outcome parameters were incidence of pCME, patient’s sex and age, laterality of eye, coexistence of pseudoexfoliation syndrome (PEX), duration of surgery and surgeon’s sex.

RESULTS: 25.422 surgeries on 18.266 patients were included. The majority was performed by experienced surgeons (23.139, 91.0%) vs 2.283 (9.0%) by surgeons in training (25 surgeons, 9 (36%) female and 16 (64%) male). pCME occurred in 32 eyes (1.4%) following surgery by surgeons in training and in 152 eyes (0.7%) following surgery by experienced surgeons. Chance for pCME was 1.57 higher in training surgeries (95% CI 1.03-2.41, p = 0.034) and longer duration (OR = 1.04; 95% CI 1.02-1.07, p = 0.001). After excluding the first 100 surgeries for every surgeon in training similar results were observed. No difference in risk for pCME was found between female and male surgeons in both groups (training and experienced surgeons).

CONCLUSION: In conclusion, the rate for pCME after uneventful cataract surgery is significantly higher for surgeons in training but steadily decreasing and associated to surgical time. No difference in the risk for pCME was found between female and male surgeons.

PMID:36574394 | DOI:10.1371/journal.pone.0279518

PLoS One. 2022 Dec 27;17(12):e0274724. doi: 10.1371/journal.pone.0274724. eCollection 2022.

ABSTRACT

Specific microRNAs expressions may accurately characterize different stages of canine myxomatous mitral valve disease. This preliminary pilot study aimed to (1) describe the clinical and echocardiographic parameters of Cavalier King Charles Spaniels affected by myxomatous mitral valve disease at different American College of Veterinary Internal Medicine (ACVIM) stages (B1, B2 and C) and healthy control group (ACVIM A), comparing the parameters collected during the first examination (T0) and the end of the follow-up (T1); (2) assess the association between the values of echocardiographic parameters at T1 and the expression profile of miR-30b-5p at T0. Thirty-five Cavalier King Charles Spaniels (median age 4.29 years and median weight 9 Kg) in different ACVIM stages were included (7 A, 19 B1, 6 B2 and 3 C). Inverse probability weighting analysis was performed to estimate the association of the exposure variable (miR-30b-5p) with the outcome variables (clinical and echocardiographic variables). Time was included as variable. The results pointed out that high levels of plasma miR-30b-5p corresponded to lower values of left ventricular end-diastolic diameter normalized for body weight, end-diastolic and end-systolic volumes indexed for body weight, and left atrium-to aortic root ratio. Hence, higher miR-30b-5p expressions were associated with milder forms of mitral valve disease in our study population. In contrast, the results obtained for the intensity of heart murmur, the mitral regurgitation severity, and the Mitral INsufficiency Echocardiographic score) were not statistically significant. A relationship between high abundance of miR-30b-5p and myxomatous mitral valve disease that appear echocardiographically more stable over time has been demonstrated. In conclusion, Cavalier King Charles Spaniels affected by myxomatous mitral valve disease that at the first cardiologic evaluation showed an upregulation of miR-30b-5p are expected to experience lesser variations on their echocardiographic examination between T0 and T1.

PMID:36574372 | DOI:10.1371/journal.pone.0274724

PLoS One. 2022 Dec 27;17(12):e0279441. doi: 10.1371/journal.pone.0279441. eCollection 2022.

ABSTRACT

BACKGROUND: Highly connected individuals disseminate information effectively within their social network. To apply this concept to inflammatory bowel disease (IBD) care and lay the foundation for network interventions to disseminate high-quality treatment, we assessed the need for improving the IBD practices of highly connected clinicians. We aimed to examine whether highly connected clinicians who treat IBD patients were more likely to provide high-quality treatment than less connected clinicians.

METHODS: We used network analysis to examine connections among clinicians who shared patients with IBD in the Veterans Health Administration between 2015-2018. We created a network comprised of clinicians connected by shared patients. We quantified clinician connections using degree centrality (number of clinicians with whom a clinician shares patients), closeness centrality (reach via shared contacts to other clinicians), and betweenness centrality (degree to which a clinician connects clinicians not otherwise connected). Using weighted linear regression, we examined associations between each measure of connection and two IBD quality indicators: low prolonged steroids use, and high steroid-sparing therapy use.

RESULTS: We identified 62,971 patients with IBD and linked them to 1,655 gastroenterologists and 7,852 primary care providers. Clinicians with more connections (degree) were more likely to exhibit high-quality treatment (less prolonged steroids beta -0.0268, 95%CI -0.0427, -0.0110, more steroid-sparing therapy beta 0.0967, 95%CI 0.0128, 0.1805). Clinicians who connect otherwise unconnected clinicians (betweenness) displayed more prolonged steroids use (beta 0.0003, 95%CI 0.0001, 0.0006). The presence of variation is more relevant than its magnitude.

CONCLUSIONS: Clinicians with a high number of connections provided more high-quality IBD treatments than less connected clinicians, and may be well-positioned for interventions to disseminate high-quality IBD care. However, clinicians who connect clinicians who are otherwise unconnected are more likely to display low-quality IBD treatment. Efforts to improve their quality are needed prior to leveraging their position to disseminate high-quality care.

PMID:36574370 | DOI:10.1371/journal.pone.0279441

Eur J Gastroenterol Hepatol. 2023 Feb 1;35(2):212-218. doi: 10.1097/MEG.0000000000002476. Epub 2022 Nov 8.

ABSTRACT

OBJECTIVES: This study aimed to evaluate the efficacy and safety of long-term postpartum tenofovir disoproxil fumarate (TDF) therapy in hepatitis B virus (HBV)-infected mothers with high viral load.

METHODS: In this retrospective cohort study, HBV-infected mothers with HBV DNA>2 × 105 IU/mL who initiated TDF prophylaxis treatment during pregnancy were divided into TDF continuation and discontinuation groups according to whether they stopped TDF treatment within 3 months after birth or not. Virological and biochemical markers were collected before TDF treatment, antepartum and postpartum.

RESULTS: In 131 women followed for a median of 18 months postpartum, alanine aminotransferase (ALT) abnormality rate was significantly lower in TDF continuation group vs. discontinuation group (39.4% vs. 56.9%, P = 0.045), and continuous TDF therapy in postpartum was independently associated with lower risk of ALT flares [OR = 0.308, 95% confidence interval (CI), 0.128-0.742; P = 0.009]. Long-term postpartum TDF treatment can promote the decline of hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg) levels, but the HBeAg seroconversion rate in two groups was not significant (15.5% vs. 11.7%, P = 0.541). There were no statistical differences in bone metabolism markers between two groups (P > 0.05). Compared with the TDF discontinuation group, TDF continuation group had a significantly lower estimated glomerular filtration rate level and higher creatinine level in postpartum but within normal ranges (P < 0.05).

CONCLUSIONS: For pregnant women who received prophylactic TDF treatment, long-term TDF therapy continued in postpartum can reduce the risk of ALT flares and promote the rapid decline of HBeAg and HBsAg levels.

PMID:36574312 | DOI:10.1097/MEG.0000000000002476

Eur J Gastroenterol Hepatol. 2023 Feb 1;35(2):174-180. doi: 10.1097/MEG.0000000000002486. Epub 2022 Dec 8.

ABSTRACT

OBJECTIVE: The first choice of treatment for rectal neuroendocrine tumors (R-NETs) 10 mm in size is endoscopic resection, there is still controversy concerning the optimal endoscopic treatment for resecting R-NETs. This study evaluated the efficacy and safety of multiple ligation-assisted endoscopic submucosal resection combined with endoscopic ultrasonography (EMR-MLUS) for R-NETs.

METHODS: We retrospectively analyzed the data of 62 patients with R-NETs ≤10 mm in size who underwent EMR-MLUS or ligation-assisted endoscopic submucosal resection combined with endoscopic ultrasonography (EMR-LUS) between May 2019 and April 2022, including tumor characteristics, endoscopic complete resection, pathological complete resection, the procedure time, adverse events, and follow-up were compared between the two groups of patients.

RESULTS: Of the 62 patients, 19 underwent EMR-MLUS and 43 underwent EMR-LUS. The endoscopic morphology of lesions was statistically different between group EMR-MLUS and group EMR-LUS (P = 0.015), and most of them were flat and slightly raised lesions in group EMR-MLUS. Although the pathological complete resection rate was slightly higher in the EMR-MLUS group than in the EMR-LUS group (94.74% vs. 90.70%; P = 0.290), the endoscopic complete resection rate was high in both groups. Involvement of the lateral resection margin was found four cases in the ESMR-LUS group; one case of deep resection margin involvement in the EMR-MLUS group. The mean procedure time was longer in the EMR-MLUS group than in the EMR-LUS group (12.79 ± 1.01 min vs. 11.08 ± 1.89 min; P = 0.041). In group EMR-LUS, there were two cases of immediate bleeding; in group EMR-MLUS, one case of perforation, all of them were successfully treated by endoscopy. No recurrence, progression, or metastasis was found in all patients.

CONCLUSION: EMR-MLUS is a safe and effective technique that could be considered when removing small rectal NETs, especially flat and slightly raised lesions.

PMID:36574308 | DOI:10.1097/MEG.0000000000002486

J Cardiovasc Med (Hagerstown). 2022 Dec 23. doi: 10.2459/JCM.0000000000001410. Online ahead of print.

ABSTRACT

BACKGROUND: Infective endocarditis (IE) is a significant disease characterized by high mortality and complications. The aim of this study was to evaluate the incidence/100 000 inhabitants and the in-hospital mortality/100 000 inhabitants of IE during the last 10 years in the province of Ravenna.

METHODS AND RESULTS: We reviewed the public hospitals discharge database from January 2010 to December 2020 using the international classification of disease codification (ICD-9) for IE. We used the Italian national statistical institute (ISTAT) archive to estimate the number of Ravenna inhabitants/year. In 10 years, we identified a total of 407 patients with diagnosis of IE.The incidence of IE increased significantly from 6.29 cases/100 000 inhabitants in 2010 to 19.58 cases/100 000 inhabitants in 2020 (P < 0.001). Also, the in-hospital mortality from IE increased over the same number of years, from 1.8 deaths/100 000 inhabitants in 2010 to 4.4 deaths/100 000 inhabitants in 2020 (P < 0.001). The mortality rate (%) of IE over the years did not increase (P = 0.565). Also, over the years there was no difference in the site of infection (P = 0.372), irrespective of the valve localization or type, native valve (P = 0.347) or prosthetic valve (P = 0.145). On logistic regression analysis, age was the only predictor of in-hospital mortality (odds ratio 1.045, 95% confidence interval: 1.015; 1.075, P = 0.003).

CONCLUSIONS: Ravenna-based data on IE showed increased disease incidence but unchanged mortality rate over 10 years of follow-up. Age remains the sole predictor of population-based mortality, irrespective of the nature of the valve, native or substitute, and the organism detected on microbiology.

PMID:36574285 | DOI:10.2459/JCM.0000000000001410

J Med Internet Res. 2022 Dec 27;24(12):e41972. doi: 10.2196/41972.

ABSTRACT

BACKGROUND: Health care delivery and patient satisfaction are improved when patients engage with their medical information through patient portals. Despite their wide availability and multiple functionalities, patient portals and their functionalities are still underused.

OBJECTIVE: We seek to understand factors that lead to patient engagement through multiple portal functionalities. We provide recommendations that could lead to higher patients’ usage of their portals.

METHODS: Using data from the Health Information National Trends Survey 5, Cycle 3 (N=2093), we performed descriptive statistics and used a chi-square test to analyze the association between the demographic variables and the use of mobile health apps for accessing medical records. We further fitted a generalized linear model to examine the association between access type and the use of portal functionalities. We further examined the moderation effects of age groups on the impact of access type on portal usage.

RESULTS: Our results show that accessing personal health records using a mobile health app is positively associated with greater patient usage of access capabilities (β=.52; P<.001), patient-provider interaction capabilities (β=.24, P=.006), and patient-personal health information interaction capabilities (β=.23, P=.009). Patients are more likely to interact with their records and their providers when accessing their electronic medical records using a mobile health app. The impacts of mobile health app usage fade with age for tasks consisting of viewing, downloading, and transmitting medical results to a third party (β=-.43, P=.005), but not for those involving patient-provider interaction (β=.05, P=.76) or patient-personal health information interaction (β=-.15, P=.19).

CONCLUSIONS: These findings provide insights on how to increase engagement with diverse portal functionalities for different age groups and thus improve health care delivery and patient satisfaction.

PMID:36574284 | DOI:10.2196/41972

Med Glas (Zenica). 2023 Feb 1;20(1). doi: 10.17392/1517-22. Online ahead of print.

ABSTRACT

Aim Anterior cruciate ligament (ACL) injury is among the most common orthopaedic injuries. The elevated body mass index (BMI) can contribute to non-contact ACL injury. This study aims to assess the risk of ACL injury among elevated BMI population people (BMI ≥25 Kgm2 ). Methods This is a cross sectional study that was conducted in a tertiary care centre in the Kingdom of Saudi Arabia. A total of 302 patients, who had an ACL reconstruction surgery in a ten-year-period (January 2008 to December 2018) were included. Results Sport related injury is significantly higher among the overweight and obese groups (p=0.002). Moreover, the combined ACL tear was higher among the overweight and obese groups (p=0.001). In univariate regression analysis for the selected baseline characteristics, it was found that individuals with higher BMI have chance to develop combined (ACL) injury 2 times higher when compared to those with isolated ACL injury (p=0.003). Also, the ACL type, mode of injury, types of injury and type of sports were statistically significant in univariate regression analysis. However, only the mode of injury was statistically significant after controlling the confounding factors. Other selected variables like type of sport, type of injury and ACL type were not significant. Conclusion Elevated BMI was associated with a higher risk of developing combined ACL tear as well as reinjured individuals.

PMID:36574270 | DOI:10.17392/1517-22

JAMA Netw Open. 2022 Dec 1;5(12):e2248559. doi: 10.1001/jamanetworkopen.2022.48559.

ABSTRACT

IMPORTANCE: Machine learning approaches can assist opioid stewardship by identifying high-risk opioid prescribing for potential interventions.

OBJECTIVE: To develop a machine learning model for deployment that can estimate the risk of adverse outcomes within 30 days of an opioid dispensation as a potential component of prescription drug monitoring programs using access to real-world data.

DESIGN, SETTING, AND PARTICIPANTS: This prognostic study used population-level administrative health data to construct a machine learning model. This study took place in Alberta, Canada (from January 1, 2018, to December 31, 2019), and included all patients 18 years and older who received at least 1 opioid dispensation from a community pharmacy within the province.

EXPOSURES: Each opioid dispensation served as the unit of analysis.

MAIN OUTCOMES AND MEASURES: Opioid-related adverse outcomes were identified from administrative data sets. An XGBoost model was developed on 2018 data to estimate the risk of hospitalization, an emergency department visit, or mortality within 30 days of an opioid dispensation; validation on 2019 data was done to evaluate model performance. Model discrimination, calibration, and other relevant metrics are reported using daily and weekly predictions on both ranked predictions and predicted probability thresholds using all data from 2019.

RESULTS: A total of 853 324 participants represented 6 181 025 opioid dispensations, with 145 016 outcome events reported (2.3%); 46.4% of the participants were men and 53.6% were women, with a mean (SD) age of 49.1 (15.6) years for men and 51.0 (18.0) years for women. Of the outcome events, 77 326 (2.6% pretest probability) occurred within 30 days of a dispensation in the validation set (XGBoost C statistic, 0.82 [95% CI, 0.81-0.82]). The top 0.1 percentile of estimated risk had a positive likelihood ratio (LR) of 28.7, which translated to a posttest probability of 43.1%. In our simulations, the weekly measured predictions had higher positive LRs in both the highest-risk dispensations and percentiles of estimated risk compared with predictions measured daily. Net benefit analysis showed that using machine learning prediction may not add additional benefit over the entire range of probability thresholds.

CONCLUSIONS AND RELEVANCE: These findings suggest that prescription drug monitoring programs can use machine learning classifiers to identify patients at risk of opioid-related adverse outcomes and intervene on high-risk ranked predictions. Better access to available administrative and clinical data could improve the prediction performance of machine learning classifiers and thus expand opioid stewardship efforts.

PMID:36574245 | DOI:10.1001/jamanetworkopen.2022.48559