Categories
Nevin Manimala Statistics

A deep attention LSTM embedded aggregation network for multiple histopathological images

PLoS One. 2023 Jun 29;18(6):e0287301. doi: 10.1371/journal.pone.0287301. eCollection 2023.

ABSTRACT

Recent advancements in computer vision and neural networks have facilitated the medical imaging survival analysis for various medical applications. However, challenges arise when patients have multiple images from multiple lesions, as current deep learning methods provide multiple survival predictions for each patient, complicating result interpretation. To address this issue, we developed a deep learning survival model that can provide accurate predictions at the patient level. We propose a deep attention long short-term memory embedded aggregation network (DALAN) for histopathology images, designed to simultaneously perform feature extraction and aggregation of lesion images. This design enables the model to efficiently learn imaging features from lesions and aggregate lesion-level information to the patient level. DALAN comprises a weight-shared CNN, attention layers, and LSTM layers. The attention layer calculates the significance of each lesion image, while the LSTM layer combines the weighted information to produce an all-encompassing representation of the patient’s lesion data. Our proposed method performed better on both simulated and real data than other competing methods in terms of prediction accuracy. We evaluated DALAN against several naive aggregation methods on simulated and real datasets. Our results showed that DALAN outperformed the competing methods in terms of c-index on the MNIST and Cancer dataset simulations. On the real TCGA dataset, DALAN also achieved a higher c-index of 0.803±0.006 compared to the naive methods and the competing models. Our DALAN effectively aggregates multiple histopathology images, demonstrating a comprehensive survival model using attention and LSTM mechanisms.

PMID:37384648 | DOI:10.1371/journal.pone.0287301

Categories
Nevin Manimala Statistics

Geologists are using artificial intelligence to predict landslides

Many factors influence where a landslide will occur, including the shape of the terrain, its slope and drainage areas, the material properties of soil and bedrock, and environmental conditions like climate, rainfall, hydrology and ground motion resulting from earthquakes. Geologists have developed a new technique that uses artificial intelligence to better predict where and why landslides may occur could bolster efforts to protect lives and property in some of the world’s most disaster-prone areas. The new method improves the accuracy and interpretability of AI-based machine-learning techniques, requires far less computing power and is more broadly applicable than traditional predictive models.
Categories
Nevin Manimala Statistics

Economic Evaluation of Digital Therapeutic Care Apps for Unsupervised Treatment of Low Back Pain: Monte Carlo Simulation

JMIR Mhealth Uhealth. 2023 Jun 29;11:e44585. doi: 10.2196/44585.

ABSTRACT

BACKGROUND: Digital therapeutic care (DTC) programs are unsupervised app-based treatments that provide video exercises and educational material to patients with nonspecific low back pain during episodes of pain and functional disability. German statutory health insurance can reimburse DTC programs since 2019, but evidence on efficacy and reasonable pricing remains scarce. This paper presents a probabilistic sensitivity analysis (PSA) to evaluate the efficacy and cost-utility of a DTC app against treatment as usual (TAU) in Germany.

OBJECTIVE: The aim of this study was to perform a PSA in the form of a Monte Carlo simulation based on the deterministic base case analysis to account for model assumptions and parameter uncertainty. We also intend to explore to what extent the results in this probabilistic analysis differ from the results in the base case analysis and to what extent a shortage of outcome data concerning quality-of-life (QoL) metrics impacts the overall results.

METHODS: The PSA builds upon a state-transition Markov chain with a 4-week cycle length over a model time horizon of 3 years from a recently published deterministic cost-utility analysis. A Monte Carlo simulation with 10,000 iterations and a cohort size of 10,000 was employed to evaluate the cost-utility from a societal perspective. Quality-adjusted life years (QALYs) were derived from Veterans RAND 6-Dimension (VR-6D) and Short-Form 6-Dimension (SF-6D) single utility scores. Finally, we also simulated reducing the price for a 3-month app prescription to analyze at which price threshold DTC would result in being the dominant strategy over TAU in Germany.

RESULTS: The Monte Carlo simulation yielded on average a €135.97 (a currency exchange rate of EUR €1=US $1.069 is applicable) incremental cost and 0.004 incremental QALYs per person and year for the unsupervised DTC app strategy compared to in-person physiotherapy in Germany. The corresponding incremental cost-utility ratio (ICUR) amounts to an additional €34,315.19 per additional QALY. DTC yielded more QALYs in 54.96% of the iterations. DTC dominates TAU in 24.04% of the iterations for QALYs. Reducing the app price in the simulation from currently €239.96 to €164.61 for a 3-month prescription could yield a negative ICUR and thus make DTC the dominant strategy, even though the estimated probability of DTC being more effective than TAU is only 54.96%.

CONCLUSIONS: Decision-makers should be cautious when considering the reimbursement of DTC apps since no significant treatment effect was found, and the probability of cost-effectiveness remains below 60% even for an infinite willingness-to-pay threshold. More app-based studies involving the utilization of QoL outcome parameters are urgently needed to account for the low and limited precision of the available QoL input parameters, which are crucial to making profound recommendations concerning the cost-utility of novel apps.

PMID:37384379 | DOI:10.2196/44585

Categories
Nevin Manimala Statistics

External Control Arms in Idiopathic Pulmonary Fibrosis Using Clinical Trial and Real World Data Sources

Am J Respir Crit Care Med. 2023 Jun 29. doi: 10.1164/rccm.202210-1947OC. Online ahead of print.

ABSTRACT

RATIONALE: Idiopathic pulmonary fibrosis (IPF) is a progressive lung disease for which novel therapies are needed. External controls (ECs) could enhance IPF trial efficiency, though the direct comparability of ECs to concurrent controls is unknown.

OBJECTIVES: To develop IPF ECs by fit for purpose data standards to historical randomized clinical trial (RCT), multicenter registry (Pulmonary Fibrosis Foundation Patient Registry [PFF-PR]), and electronic health record (EHR) data and to evaluate endpoint comparability among ECs and the BMS-986020 phase 2 RCT.

METHODS: After data curation, the rate of change in forced vital capacity (FVC) from baseline to 26 weeks among participants receiving BMS-986020 600 mg twice daily was compared with the BMS-placebo arm and ECs using mixed effects models with inverse probability weights.

MEASUREMENTS AND MAIN RESULTS: At 26 weeks, the rate of change in FVC was -32.71 mL (BMS-986020) versus -130.09 mL (BMS-placebo; difference, 97.4 mL, 95% CI, 24.6, 170.2), replicating the original BMS-986020 RCT. RCT-ECs showed treatment effect point estimates within the 95% CI of the original BMS-986020 RCT. Both PFF-PR-ECs and EHR-ECs experienced a slower rate of FVC decline compared with the BMS-placebo arm, resulting in treatment effect point estimates outside of the 95% CI of the original BMS-986020 RCT.

CONCLUSIONS: IPF ECs generated from historical RCT placebo arms result in comparable primary treatment effects to that of the original clinical trial, while ECs from real-world data sources, including registry or EHR data, do not. RCT-ECs may serve as a potentially useful supplement to future IPF RCTs.

PMID:37384378 | DOI:10.1164/rccm.202210-1947OC

Categories
Nevin Manimala Statistics

Evaluating the Effectiveness of Interactive Virtual Patients for Medical Education in Zambia: Randomized Controlled Trial

JMIR Med Educ. 2023 Jun 29;9:e43699. doi: 10.2196/43699.

ABSTRACT

BACKGROUND: Zambia is facing a severe shortage of health care workers, particularly in rural areas. Innovative educational programs and infrastructure have been established to bridge this gap; however, they encounter substantial challenges because of constraints in physical and human resources. In response to these shortcomings, strategies such as web-based and blended learning approaches have been implemented, using virtual patients (VPs) as a means to promote interactive learning at the Levy Mwanawasa Medical University (LMMU) in Zambia.

OBJECTIVE: This study aimed to evaluate the students’ knowledge acquisition and acceptance of 2 VP medical topics as a learning tool on a Zambian higher education e-learning platform.

METHODS: Using a mixed methods design, we assessed knowledge acquisition using pre- and posttests. In a randomized controlled trial setting, students were assigned (1:1) to 2 medical topics (topic 1: appendicitis and topic 2: severe acute malnutrition) and then to 4 different learning tools within their respective exposure groups: VPs, textbook content, preselected e-learning materials, and self-guided internet materials. Acceptance was evaluated using a 15-item questionnaire with a 5-point Likert scale.

RESULTS: A total of 63 third- and fourth-year Bachelor of Science clinical science students participated in the study. In the severe acute malnutrition-focused group, participants demonstrated a significant increase in knowledge within the textbook group (P=.01) and the VP group (P=.01). No substantial knowledge gain was observed in the e-learning group or the self-guided internet group. For the appendicitis-focused group, no statistically significant difference in knowledge acquisition was detected among the 4 intervention groups (P=.62). The acceptance of learning materials exhibited no substantial difference between the VP medical topics and other learning materials.

CONCLUSIONS: In the context of LMMU, our study found that VPs were well accepted and noninferior to traditional teaching methods. VPs have the potential to serve as an engaging learning resource and can be integrated into blended learning approaches at LMMU. However, further research is required to investigate the long-term knowledge gain and the acceptance and effectiveness of VPs in medical education.

TRIAL REGISTRATION: Pan African Clinical Trials Registry (PACTR) PACTR202211594568574; https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=20413.

PMID:37384369 | DOI:10.2196/43699

Categories
Nevin Manimala Statistics

Use of Electronic Ecological Momentary Assessment Methodologies in Physical Activity, Sedentary Behavior, and Sleep Research in Young Adults: Systematic Review

J Med Internet Res. 2023 Jun 29;25:e46783. doi: 10.2196/46783.

ABSTRACT

BACKGROUND: Recent technological advances allow for the repeated sampling of real-time data in natural settings using electronic ecological momentary assessment (eEMA). These advances are particularly meaningful for investigating physical activity, sedentary behavior, and sleep in young adults who are in a critical life stage for the development of healthy lifestyle behaviors.

OBJECTIVE: This study aims to describe the use of eEMA methodologies in physical activity, sedentary behavior, and sleep research in young adults.

METHODS: The PubMed, CINAHL, PsycINFO, Embase, and Web of Science electronic databases were searched through August 2022. Inclusion criteria were use of eEMA; sample of young adults aged 18 to 25 years; at least 1 measurement of physical activity, sedentary behavior, or sleep; English language; and a peer-reviewed report of original research. Study reports were excluded if they were abstracts, protocols, or reviews. The risk of bias assessment was conducted using the National Heart, Lung, and Blood Institute’s Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies. Screening, data extraction, and risk of bias assessments were conducted by independent authors, with discrepancies resolved by consensus. Descriptive statistics and narrative synthesis were used to identify overarching patterns within the following categories guided by the Checklist for Reporting Ecological Momentary Assessments Studies: study characteristics, outcomes and measures, eEMA procedures, and compliance.

RESULTS: The search resulted in 1221 citations with a final sample of 37 reports describing 35 unique studies. Most reports (28/37, 76%) were published in the last 5 years (2017-2022), used observational designs (35/37, 95%), consisted of samples of college students or apprentices (28/35, 80%), and were conducted in the United States (22/37, 60%). The sample sizes ranged from 14 to 1584 young adults. Physical activity was measured more frequently (28/37, 76%) than sleep (16/37, 43%) or sedentary behavior (4/37, 11%). Of the 37 studies, 11 (30%) reports included 2 movement behaviors and no reports included 3 movement behaviors. eEMA was frequently used to measure potential correlates of movement behaviors, such as emotional states or feelings (25/37, 68%), cognitive processes (7/37, 19%), and contextual factors (9/37, 24%). There was wide variability in the implementation and reporting of eEMA procedures, measures, missing data, analysis, and compliance.

CONCLUSIONS: The use of eEMA methodologies in physical activity, sedentary behavior, and sleep research in young adults has greatly increased in recent years; however, reports continue to lack standardized reporting of features unique to the eEMA methodology. Additional areas in need of future research include the use of eEMA with more diverse populations and the incorporation of all 3 movement behaviors within a 24-hour period. The findings are intended to assist investigators in the design, implementation, and reporting of physical activity, sedentary behavior, and sleep research using eEMA in young adults.

TRIAL REGISTRATION: PROSPERO CRD42021279156; https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021279156.

PMID:37384367 | DOI:10.2196/46783

Categories
Nevin Manimala Statistics

Osmoprotectants play a major role in the Portulaca oleracea resistance to high levels of salinity stress-insights from a metabolomics and proteomics integrated approach

Front Plant Sci. 2023 Jun 13;14:1187803. doi: 10.3389/fpls.2023.1187803. eCollection 2023.

ABSTRACT

INTRODUCTION: Purslane (Portulaca oleracea L.) is a non-conventional food plant used extensively in folk medicine and classified as a multipurpose plant species, serving as a source of features of direct importance to the agricultural and agri-industrial sectors. This species is considered a suitable model to study the mechanisms behind resistance to several abiotic stresses including salinity. The recently achieved technological developments in high-throughput biology opened a new window of opportunity to gain additional insights on purslane resistance to salinity stress-a complex, multigenic, and still not well-understood trait. Only a few reports on single-omics analysis (SOA) of purslane are available, and only one multi-omics integration (MOI) analysis exists so far integrating distinct omics platforms (transcriptomics and metabolomics) to characterize the response of purslane plants to salinity stress.

METHODS: The present study is a second step in building a robust database on the morpho-physiological and molecular responses purslane to salinity stress and its subsequent use in attempting to decode the genetics behind its resistance to this abiotic stress. Here, the characterization of the morpho-physiological responses of adult purslane plants to salinity stress and a metabolomics and proteomics integrative approach to study the changes at the molecular level in their leaves and roots is presented.

RESULTS AND DISCUSSION: Adult plants of the B1 purslane accession lost approximately 50% of the fresh and dry weight (from shoots and roots) whensubmitted to very high salinity stress (2.0 g of NaCl/100 g of the substrate). The resistance to very high levels of salinity stress increases as the purslane plant matures, and most of the absorbed sodium remains in the roots, with only a part (~12%) reaching the shoots. Crystal-like structures, constituted mainly by Na+, Cl, and K+, were found in the leaf veins and intercellular space near the stoma, indicating that this species has a mechanism of salt exclusion operating on the leaves, which has its role in salt tolerance. The MOI approach showed that 41 metabolites were statistically significant on the leaves and 65 metabolites on the roots of adult purslane plants. The combination of the mummichog algorithm and metabolomics database comparison revealed that the glycine, serine, and threonine, amino sugar and nucleotide sugar, and glycolysis/gluconeogenesis pathways were the most significantly enriched pathways when considering the total number of occurrences in the leaves (with 14, 13, and 13, respectively) and roots (all with eight) of adult plants; and that purslane plants employ the adaptive mechanism of osmoprotection to mitigate the negative effect of very high levels of salinity stress; and that this mechanism is prevalent in the leaves. The multi-omics database built by our group underwent a screen for salt-responsive genes, which are now under further characterization for their potential to promote resistance to salinity stress when heterologously overexpressed in salt-sensitive plants.

PMID:37384354 | PMC:PMC10296175 | DOI:10.3389/fpls.2023.1187803

Categories
Nevin Manimala Statistics

TRICHOMONAS VAGINALIS INFECTION AMONG ASYMPTOMATIC UNDERGRADUATE STUDENTS IN A PRIVATE UNIVERSITY IN OGUN STATE, NIGERIA

Ann Ib Postgrad Med. 2022 Dec;20(2):135-142.

ABSTRACT

BACKGROUND: Trichomonas vaginalis is one of the common non- viral sexually transmitted infections that infect both men and women worldwide. It is largely asymptomatic and its association with the risk of HIV transmission has made it a compelling public health concern. Therefore, this study aims to determine the prevalence and the risk factors associated with T. vaginalis among asymptomatic undergraduate students at Babcock University, Ilisan-Remo, Ogun state, Nigeria.

MATERIALS AND METHODS: This is a descriptive cross-sectional study involving 246 asymptomatic students of Babcock University between February 2019 to April 2020. Information on socio-demographic and associated risk factors was obtained by structured-questionnaire in an interview-based manner. First-void urine was collected from each participant for the detection of T. vaginalis using the traditional wet prep method and TV in-pouch. The data were analyzed by SPSS Version 23.

RESULTS: The overall prevalence of T. vaginalis among the participants was 12.2% (30/246). The use of wet-preparation showed 8.5% (21/246) while the use of TV inpouch yielded 12. 2% (30/246) prevalence of positive results. The results of the wet prep in comparison to the in-pouch technique was statistically significant among the study population. (P < 0.001). Sexual intercourse, use of hormonal contraceptives and practice of internet-based sex seeking behaviour were factors that had increase likelihood of T. vaginalis infection on multivariate analysis.

CONCLUSION: The occurrence of T. vaginalis and, its associated risk factors among the asymptomatic population in this study is very high. We advocate for the screening of young people.

PMID:37384345 | PMC:PMC10295101

Categories
Nevin Manimala Statistics

BIOCHEMICAL PREDICTORS OF ENTEROCOLITIS IN CHILDREN WITH COLORECTAL ANOMALIES POST COLOSTOMY AT UNIVERSITY COLLEGE HOSPITAL, IBADAN

Ann Ib Postgrad Med. 2022 Dec;20(2):129-134.

ABSTRACT

BACKGROUND: A large proportion of patients with preoperative enterocolitis still have enterocolitis persisting even after surgery while others resolve thereafter. Some researchers have studied Calprotectin, C-reactive protein (CRP), Blood and Plasma viscosity as markers of inflammation, hence, the choice of their use. The aim of the study is to determine the sensitivity and reliability of Calprotectin, C-reactive protein (CRP), Blood and Plasma viscosity as biochemical predictors of enterocolitis in children with colorectal anomaly post-surgery at University College Hospital Ibadan.

PATIENTS AND METHODS: This is an observational analytic study of 32 patients with either Hirschsprung’s disease or Anorectal malformation carried out over a year period. The demographic data of the patients, clinical condition and the preoperative and postoperative readings of the biochemical analytes were recorded in a chart. Statistical analysis were carried out using SPSS version 23 and test for statistical association done.

RESULTS: The incidence of Hirschsprung associated enterocolitis is 12.5% and for Ano rectal malformation 6.3 %. Gender difference was not statistically significant even with the observed clinical difference. Plasma viscosity and blood viscosity correlate positively with each order. C-reactive Protein and Calprotectin did not predict enterocolitis in this study and the Sensitivity of blood viscosity at T1 and T2 is as low as 66% with a Positive Predictive Value of 25 %.

CONCLUSION: The incidence of Enterocolitis associated with Hirschsprung’s disease and Anorectal malformation is 19 %. Calprotectin and C-reactive protein did not predict enterocolitis in this patients. The outcome of care was satisfactory in over 90 % of the patients.

PMID:37384344 | PMC:PMC10295100

Categories
Nevin Manimala Statistics

SKINFOLD MEASUREMENTS AS DETERMINANTS OF BLOOD PRESSURE LEVELS AMONG ADULT HYPERTENSIVES ATTENDING A SECONDARY HEALTHCARE CENTRE IN NIGERIA

Ann Ib Postgrad Med. 2022 Dec;20(2):169-176.

ABSTRACT

BACKGROUND: Background: The prevalence of obesity is increasing globally, making it a growing pandemic affecting adults and children. Obesity is associated with multiple morbidities and mortalities increasing the burden on the health care system.

OBJECTIVE: There is inadequacy of data in Nigeria on the prevalence of obesity among adult patients with hypertension and adequate data on these conditions would help in their comprehensive management.

METHODS: This was a cross-sectional study of 354 patients with hypertension, and the systematic sampling technique was used to recruit patients. The data were analysed using SPSS software version 23. Logistic regressions and linear regressions were done to determine the predictors of obesity and blood pressure levels.

RESULTS: The mean age of the respondents was 52.60(SD±8.26) years and the prevalence of obesity was 53.1%. After adjusting for other variables, the predictors of obesity were female sex. Females were about six times more likely to be obese than males (OR=6.23; 95%CI= 3.16 – 12.32). For every 1 unit increase in triceps skinfold, there was a statistically significant increase in diastolic blood pressure by about 2.77units (95% C.I equals 2.63 to 2.91, p-value= 0.0001). Also, for every 1 unit increase in biceps skinfold, there was a statistically significant increase in systolic blood pressure by about 5.78 units (95% C.I equals 5.46- 6.10, p-value= 0.0001).

CONCLUSION: The prevalence of obesity was high, and the predictors of obesity were female sex. Triceps skinfold measurements were predictors of diastolic blood pressure while biceps skinfold measurements were predictors of systolic blood pressure.

PMID:37384341 | PMC:PMC10295094