Category: Nevin Manimala

BMC Endocr Disord. 2022 Aug 26;22(1):214. doi: 10.1186/s12902-022-01121-4.

ABSTRACT

OBJECTIVE: The internal workings ofmachine learning algorithms are complex and considered as low-interpretation “black box” models, making it difficult for domain experts to understand and trust these complex models. The study uses metabolic syndrome (MetS) as the entry point to analyze and evaluate the application value of model interpretability methods in dealing with difficult interpretation of predictive models.

METHODS: The study collects data from a chain of health examination institution in Urumqi from 2017 ~ 2019, and performs 39,134 remaining data after preprocessing such as deletion and filling. RFE is used for feature selection to reduce redundancy; MetS risk prediction models (logistic, random forest, XGBoost) are built based on a feature subset, and accuracy, sensitivity, specificity, Youden index, and AUROC value are used to evaluate the model classification performance; post-hoc model-agnostic interpretation methods (variable importance, LIME) are used to interpret the results of the predictive model.

RESULTS: Eighteen physical examination indicators are screened out by RFE, which can effectively solve the problem of physical examination data redundancy. Random forest and XGBoost models have higher accuracy, sensitivity, specificity, Youden index, and AUROC values compared with logistic regression. XGBoost models have higher sensitivity, Youden index, and AUROC values compared with random forest. The study uses variable importance, LIME and PDP for global and local interpretation of the optimal MetS risk prediction model (XGBoost), and different interpretation methods have different insights into the interpretation of model results, which are more flexible in model selection and can visualize the process and reasons for the model to make decisions. The interpretable risk prediction model in this study can help to identify risk factors associated with MetS, and the results showed that in addition to the traditional risk factors such as overweight and obesity, hyperglycemia, hypertension, and dyslipidemia, MetS was also associated with other factors, including age, creatinine, uric acid, and alkaline phosphatase.

CONCLUSION: The model interpretability methods are applied to the black box model, which can not only realize the flexibility of model application, but also make up for the uninterpretable defects of the model. Model interpretability methods can be used as a novel means of identifying variables that are more likely to be good predictors.

PMID:36028865 | DOI:10.1186/s12902-022-01121-4

J Hematol Oncol. 2022 Aug 26;15(1):116. doi: 10.1186/s13045-022-01333-0.

ABSTRACT

BACKGROUND: Patients with chronic lymphocytic leukemia (CLL) may be more susceptible to COVID-19 related poor outcomes, including thrombosis and death, due to the advanced age, the presence of comorbidities, and the disease and treatment-related immune deficiency. The aim of this study was to assess the risk of thrombosis and bleeding in patients with CLL affected by severe COVID-19.

METHODS: This is a retrospective multicenter study conducted by ERIC, the European Research Initiative on CLL, including patients from 79 centers across 22 countries. Data collection was conducted between April and May 2021. The COVID-19 diagnosis was confirmed by the real-time polymerase chain reaction (RT-PCR) assay for SARS-CoV-2 on nasal or pharyngeal swabs. Severe cases of COVID-19 were defined by hospitalization and the need of oxygen or admission into ICU. Development and type of thrombotic events, presence and severity of bleeding complications were reported during treatment for COVID-19. Bleeding events were classified using ISTH definition. STROBE recommendations were used in order to enhance reporting.

RESULTS: A total of 793 patients from 79 centers were included in the study with 593 being hospitalized (74.8%). Among these, 511 were defined as having severe COVID: 162 were admitted to the ICU while 349 received oxygen supplementation outside the ICU. Most patients (90.5%) were receiving thromboprophylaxis. During COVID-19 treatment, 11.1% developed a thromboembolic event, while 5.0% experienced bleeding. Thrombosis developed in 21.6% of patients who were not receiving thromboprophylaxis, in contrast to 10.6% of patients who were on thromboprophylaxis. Bleeding episodes were more frequent in patients receiving intermediate/therapeutic versus prophylactic doses of low-molecular-weight heparin (LWMH) (8.1% vs. 3.8%, respectively) and in elderly. In multivariate analysis, peak D-dimer level and C-reactive protein to albumin ratio were poor prognostic factors for thrombosis occurrence (OR = 1.022, 95%CI 1.007‒1.038 and OR = 1.025, 95%CI 1.001‒1.051, respectively), while thromboprophylaxis use was protective (OR = 0.199, 95%CI 0.061‒0.645). Age and LMWH intermediate/therapeutic dose administration were prognostic factors in multivariate model for bleeding (OR = 1.062, 95%CI 1.017-1.109 and OR = 2.438, 95%CI 1.023-5.813, respectively).

CONCLUSIONS: Patients with CLL affected by severe COVID-19 are at a high risk of thrombosis if thromboprophylaxis is not used, but also at increased risk of bleeding under the LMWH intermediate/therapeutic dose administration.

PMID:36028857 | DOI:10.1186/s13045-022-01333-0

BMC Public Health. 2022 Aug 26;22(1):1621. doi: 10.1186/s12889-022-14000-3.

ABSTRACT

BACKGROUND: Malnutrition happens when there are insufficient amounts of nutrients and energy consumed improperly. Included are both undernutrition and overnutrition. This study is aimed to evaluate the relationship among undernutrition indicators of stunting, underweight, and wasting among those under 5 years given other predictors.

METHODS: The data were obtained from the measure of DHS program. A total of 2399 under-five children were involved in this study. A multivariate binary logistic regression model is used to assess the association between stunting, wasting, and being underweight given the effect of other predictors.

RESULTS: Of the 2399 under-five children considered in this study, 13.5, 18.7, and 5.9% of them suffered from stunting, underweight, and wasting, respectively. The majority of children (40.1%) were obtained from the Brikama local government area of Gambia; more than half of the children (52.9%) were male, and 63.3% of children lived in urban areas. The association between stunting and underweight, underweight and wasting, and stunting and wasting was measured by the odds ratio (OR) of 15.87, 46.34, and 1.75, respectively, given the other predictors. The estimated odds ratio for children who had an average birth size to become stunted, underweight, and wasted were 0.965, 0.885, and 0.989 times the estimated odds ratio of children who had a small birth size, respectively.

CONCLUSION: The prevalence of stunting and wasting for under-five children in Gambia was lower than the world prevalence, but the prevalence of being underweight was higher. Children who are underweight have a significant association with both stunting and wasting. The age of the child, the child’s anemia level, and the birth type of the child are the common important determinants of stunting and underweight. The small birth size of a child was highly associated with a higher risk of stunting, underweight, and wasting among under five-year-olds.

PMID:36028850 | DOI:10.1186/s12889-022-14000-3

BMC Complement Med Ther. 2022 Aug 26;22(1):225. doi: 10.1186/s12906-022-03709-8.

ABSTRACT

BACKGROUND: Complementary and alternative medicine (CAM) has played a critical role in ensuring universal access to basic health care services around the world. In Bangladesh, conventional medicine is a common approach for health care practices, yet, due to Bangladesh’s high out-of-pocket payment, millions of people utilise CAM-based healthcare services for illnesses. In Bangladesh, there is a scarcity of data on how CAM is perceived and utilised. The goal of this study was to determine the proportion and correlates of the utilisation of CAM among patients visiting a tertiary level hospital, in Bangladesh.

METHODS: A cross-sectional survey involving 1,183 patients who received health care from a hospital in Bangladesh was interviewed for this study. The associated factors on utilising CAM were identified using multivariable logistic regression analysis.

RESULTS: Thirty-three percent of patients utilised CAM exclusively to treat their illnesses, whereas the rest utilised conventional medicine before CAM. Young adult patients aged 26 to 45 years (AOR = 6.26, 95% CI:3.24-12.07), patients without education (AOR = 2.99, 1.81-4.93), and married patients (AOR = 1.79, 1.08-2.97) were the most likely to be only CAM users. The most common reasons for using CAM were belief in its effectiveness, less adverse effects, affordability and lower costs.

CONCLUSION: In Bangladesh, CAM plays a significant role in health care delivery, with high-levels of patient satisfaction and health benefits. Patients who are older and have a higher level of education are more hesitant to use CAM for their illness, yet CAM has the potential to play a significant role in reducing hospitalisation by providing high reliability and low costs.

PMID:36028844 | DOI:10.1186/s12906-022-03709-8

J Cardiothorac Surg. 2022 Aug 26;17(1):208. doi: 10.1186/s13019-022-01964-x.

ABSTRACT

BACKGROUND: Minimally invasive aortic surgery is growing in popularity among surgeons. Although many clinical reports have proven both the safety and efficacy from a surgical point of view, there are few data regarding its impact on patients’ quality of life and whether there is a difference between ministernotomy and minithoracotomy from the patient perspective.

METHODS: This prospective, questionnaire-based, nonrandomized study included 189 patients who underwent aortic valve replacement via a minimally invasive incision between May 2014 and December 2020 and completed at least 1 year of follow-up. The study uses the RAND SF 36-Item Health Survey 1.0 to assess and compare health-related quality of life between ministernotomy and minithoracotomy.

RESULTS: There was a statistically significant improvement in the minithoracotomy group with regard to physical functioning, role limitation due to a physical problem, and social functioning (79.69 ± 20.72, 75.28 ± 26.52, 87.91 ± 16.98) compared to the ministernotomy group (70.31 ± 22.88, 58.59 ± 31.17, 66.15 ± 27.32) with p values (0.0036, 0.0001, < 0.0001), respectively.

CONCLUSIONS: Both minimally invasive aortic valve incisions positively impacted patient quality of life. The minithoracotomy incision showed significant improvements in physical capacity and successful patient re-engagement in daily physical and social activities. This, in turn, positively improved their general health status compared to the 1-year preoperative status.

TRIAL REGISTRATION: This study was approved by the Research Ethics Committee (REC) at the Faculty of Medicine, Ain Shams University, under the number code (FWA 000017585, FAMSU R 91 /2021).

PMID:36028838 | DOI:10.1186/s13019-022-01964-x

BMC Emerg Med. 2022 Aug 26;22(1):148. doi: 10.1186/s12873-022-00708-3.

ABSTRACT

BACKGROUND: Diagnostic errors constitute an important medical safety problem that needs improvement, and their frequency and severity are high in emergency room settings. Previous studies have suggested that diagnostic errors occur in 0.6-12% of first-time patients in the emergency room and that one or more cognitive factors are involved in 96% of these cases. This study aimed to identify the types of cognitive biases experienced by physicians in emergency rooms in Japan.

METHODS: We conducted a questionnaire survey using Nikkei Medical Online (Internet) from January 21 to January 31, 2019. Of the 159,519 physicians registered with Nikkei Medical Online when the survey was administered, those who volunteered their most memorable diagnostic error cases in the emergency room participated in the study. EZR was used for the statistical analyses.

RESULTS: A total of 387 physicians were included. The most common cognitive biases were overconfidence (22.5%), confirmation (21.2%), availability (12.4%), and anchoring (11.4%). Of the error cases, the top five most common initial diagnoses were upper gastrointestinal disease (22.7%), trauma (14.7%), cardiovascular disease (10.9%), respiratory disease (7.5%), and primary headache (6.5%). The corresponding final diagnoses for these errors were intestinal obstruction or peritonitis (27.3%), overlooked traumas (47.4%), other cardiovascular diseases (66.7%), cardiovascular disease (41.4%), and stroke (80%), respectively.

CONCLUSIONS: A comparison of the initial and final diagnoses of cases with diagnostic errors shows that there were more cases with diagnostic errors caused by overlooking another disease in the same organ or a disease in a closely related organ.

PMID:36028810 | DOI:10.1186/s12873-022-00708-3

BMC Pregnancy Childbirth. 2022 Aug 26;22(1):664. doi: 10.1186/s12884-022-04978-4.

ABSTRACT

BACKGROUND: Postpartum depression is a common condition in the pregnancy and postpartum cycle. The development of this condition is multifactorial and can be influenced by previous traumas. This study sought to verify whether there is an association between having been exposed to mistreatment during childbirth and presenting symptoms suggestive of postpartum depression.

METHODS: This is a cross-sectional study, with the inclusion of 287 women without complications in childbirth, randomly selected from two maternity hospitals of Porto Alegre, southern Brazil, in 2016. Four weeks after delivery, the postpartum women answered a face-to-face interview about socioeconomic aspects, obstetric history, health history, and childbirth experience (practices and interventions applied) and completed the Edinburgh Postnatal Depression Scale (EPDS). From the perception of women regarding the practices performed in the context of childbirth care, a composite variable was created, using item response theory, to measure the level of mistreatment during childbirth. The items that made up this variable were: absence of a companion during delivery, feeling insecure and not welcome, lack of privacy, lack of skin-to-skin contact after delivery, not having understood the information shared with them, and not having felt comfortable to ask questions and make decisions about their care. To define symptoms suggestive of postpartum depression, reflecting on increased probability of this condition, the EPDS score was set at ≥ 8. Poisson Regression with robust variance estimation was used for modeling.

RESULTS: Women who experienced mistreatment during childbirth had a higher prevalence of symptoms suggestive of postpartum depression (PR 1.55 95% CI 1.07-2.25), as well as those with a history of mental health problems (PR 1.69 95% CI 1.16-2.47), while higher socioeconomic status (A and B) had an inverse association (PR 0.53 95% CI 0.33-0.83).

CONCLUSIONS: Symptoms suggestive of postpartum depression seem to be more prevalent in women who have suffered mistreatment during childbirth, of low socioeconomic status, and with a history of mental health problems. Thus, qualifying care for women during pregnancy, childbirth and postpartum and reducing social inequalities are challenges to be faced in order to eliminate mistreatment during childbirth and reduce the occurrence of postpartum depression.

PMID:36028806 | DOI:10.1186/s12884-022-04978-4

Brief Bioinform. 2022 Aug 26:bbac346. doi: 10.1093/bib/bbac346. Online ahead of print.

ABSTRACT

Anatomical Therapeutic Chemical (ATC) classification for compounds/drugs plays an important role in drug development and basic research. However, previous methods depend on interactions extracted from STITCH dataset which may make it depend on lab experiments. We present a pilot study to explore the possibility of conducting the ATC prediction solely based on the molecular structures. The motivation is to eliminate the reliance on the costly lab experiments so that the characteristics of a drug can be pre-assessed for better decision-making and effort-saving before the actual development. To this end, we construct a new benchmark consisting of 4545 compounds which is with larger scale than the one used in previous study. A light-weight prediction model is proposed. The model is with better explainability in the sense that it is consists of a straightforward tokenization that extracts and embeds statistically and physicochemically meaningful tokens, and a deep network backed by a set of pyramid kernels to capture multi-resolution chemical structural characteristics. Its efficacy has been validated in the experiments where it outperforms the state-of-the-art methods by 15.53% in accuracy and by 69.66% in terms of efficiency. We make the benchmark dataset, source code and web server open to ease the reproduction of this study.

PMID:36027578 | DOI:10.1093/bib/bbac346

J Cardiovasc Pharmacol. 2022 Aug 23. doi: 10.1097/FJC.0000000000001364. Online ahead of print.

ABSTRACT

Only a few meta-analyses evaluated the effect of finerenone on cardiovascular events in type 2 diabetes mellitus (T2DM) with chronic kidney disease (CKD). The main aim of this meta-analysis was to gain more reliable assessments of the efficacy and safety of finerenone for prevention of cardiovascular events in diabetic kidney disease. We searched for finerenone in the treatment of diabetic kidney disease from database (PubMed, Embase and ClinicalTrials.gov) until December 30, 2021. Relative risks (RRs) with 95% confidence intervals (CIs) calculated by the Mantel-Haenszel random-effects model were used as summary statistics for the categorical data. We included four studies that met the inclusion criteria with 13,943 participants. The finerenone group demonstrated a great benefit in reducing the incidence of major adverse cardiac events (MACEs) (RR: 0.88; 95% CI 0.80 – 0.96; P = 0.003), all-cause mortality (RR: 0.89; 95% CI 0.80 – 0.99; P = 0.04), myocardial infarction (RR: 0.79; 95% CI 0.67 – 0.92; P = 0.003) and new-onset hypertension (RR: 0.71; 95% CI 0.62 – 0.81; P < 0.00001). No difference was found in adverse events between the finerenone and placebo groups (RR: 1.00; 95% CI [0.98, 1.01], P = 0.59), whereas, a higher risk of hyperkalemia was observed in the finerenone group than in the placebo group (RR = 2.04, 95% CI 1.80 – 2.32; P < 0.00001). Besides, cerebrovascular events and new-onset atrial fibrillation did not increase in patients taking finerenone. Overall, finerenone treatment showed a great benefit of reducing the risk of MACEs, all-cause mortality, myocardial infarction, and new-onset hypertension events in patients with T2DM and CKD.

PMID:36027585 | DOI:10.1097/FJC.0000000000001364

ACS Appl Bio Mater. 2022 Aug 26. doi: 10.1021/acsabm.2c00434. Online ahead of print.

ABSTRACT

Nanoparticle (NP) internalization by cells is complex, highly heterogeneous, and fundamentally important for nanomedicine. We report powerful probabilistic statistics from single-cell data on quantitative NP uptake of PEG-coated transferrin receptor-targeted gold NPs for cancer-derived and fibroblast cells according to their cell size, receptor expression, and receptor density. The smaller cancer cells had a greater receptor density and more efficient uptake of targeted NPs. However, simply due to fibroblasts being larger with more receptors, they exhibited greater NP uptake. While highly heterogeneous, targeted NP uptake strongly correlated with receptor expression. When uptake was normalized to cell size, no correlation existed. Consequently, skewed population distributions in cell sizes explain the distribution in NP uptake. Furthermore, exposure to the transferrin receptor-targeted NPs alters the fibroblast size and receptor expression, suggesting that the receptor-targeted NPs may interfere with the metabolic flux and nutrient exchange, which could assist in explaining the altered regulation of cells exposed to nanoparticles.

PMID:36027561 | DOI:10.1021/acsabm.2c00434