Nevin Manimala

Phys Med Biol. 2022 Jan 20. doi: 10.1088/1361-6560/ac4d85. Online ahead of print.

ABSTRACT

OBJECTIVE: This work investigates the use of deep convolutional neural networks (CNN) to automatically perform measurements of fetal body parts, including head circumference, biparietal diameter, abdominal circumference and femur length, and to estimate gestational age and fetal weight using fetal ultrasound videos.

APPROACH: We developed a novel multi-task CNN-based spatio-temporal fetal US feature extraction and standard plane detection algorithm (called FUVAI) and evaluated the method on 50 freehand fetal US video scans. We compared FUVAI fetal biometric measurements with measurements made by five experienced sonographers at two time points separated by at least two weeks. Intra- and inter-observer variabilities were estimated.

MAIN RESULTS: We found that automated fetal biometric measurements obtained by FUVAI were comparable to the measurements performed by experienced sonographers The observed differences in measurement values were within the range of inter- and intra-observer variability. Moreover, analysis has shown that these differences were not statistically significant when comparing any individual medical expert to our model.

SIGNIFICANCE: We argue that FUVAI has the potential to assist sonographers who perform fetal biometric measurements in clinical settings by providing them with suggestions regarding the best measuring frames, along with automated measurements. Moreover, FUVAI is able perform these tasks in just a few seconds, which is a huge difference compared to the average of six minutes taken by sonographers. This is significant, given the shortage of medical experts capable of interpreting fetal ultrasound images in numerous countries.

PMID:35051921 | DOI:10.1088/1361-6560/ac4d85

Biomed Phys Eng Express. 2022 Jan 20. doi: 10.1088/2057-1976/ac4d43. Online ahead of print.

ABSTRACT

In this study, we investigated the possibility of predicting expression levels of programmed death-ligand 1 (PD-L1) using radiomic features of intratumoral and peritumoral tumors on computed tomography (CT) images. We retrospectively analyzed 161 patients with non-small cell lung cancer. We extracted radiomics features for intratumoral and peritumoral regions on CT images. The null importance, least absolute shrinkage, and selection operator model were used to select the optimized feature subset to build the prediction models for the PD-L1 expression level. LightGBM with five-fold cross-validation was used to construct the prediction model and evaluate the receiver operating characteristics. The corresponding area under the curve (AUC) was calculated for the training and testing cohorts. The proportion of ambiguously clustered pairs was calculated based on consensus clustering to evaluate the validity of the selected features. In addition, Radscore was calculated for the training and test cohorts. For expression level of PD-L1 above 1%, prediction models that included radiomic features from the intratumoral region and a combination of radiomic features from intratumoral and peritumoral regions yielded an AUC of 0.83 and 0.87 and 0.64 and 0.74 in the training and test cohorts, respectively. In contrast, the models above 50% prediction yielded an AUC of 0.80, 0.97, and 0.74, 0.83, respectively. The selected features were divided into two subgroups based on PD-L1 expression levels ≥ 50% or ≥ 1%. Radscore was statistically higher for subgroup one than subgroup two when radiomic features for intratumoral and peritumoral regions were combined. We constructed a predictive model for PD-L1 expression level using CT images. The model using a combination of intratumoral and peritumoral radiomic features had a higher accuracy than the model with only intratumoral radiomic features.

PMID:35051908 | DOI:10.1088/2057-1976/ac4d43

Nurse Educ Today. 2022 Jan 14;110:105266. doi: 10.1016/j.nedt.2022.105266. Online ahead of print.

ABSTRACT

BACKGROUND: Obstetric violence is a problem that has grown worldwide, and a particularly worrying one in Spain. Such violence has repercussions for women, and for the professionals who cause them. Preventing this problem seems fundamental.

OBJECTIVE: This study evaluated how health sciences students perceived obstetric violence.

DESIGN: A cross-sectional study conducted between October 2019 and November 2020.

PARTICIPANTS: A sample of Spanish health sciences students studying degrees of nursing, medicine, midwifery, and psychology.

METHODS: A validated questionnaire was used: Perception of Obstetric Violence in Students (PercOV-S). Socio-demographic and control variables were included. A descriptive and comparative multivariate analysis was performed with the obtained data.

RESULTS: 540 questionnaires were completed with an overall mean score of 3.83 points (SD ± 0.63), with 2.83 points (SD ± 0.91) on the protocolised-visible dimension and 4.15 points (SD ± 0.67) on the non-protocolised-invisible obstetric violence dimension. Statistically significant differences were obtained for degree studied (p < 0.001), gender (p < 0.001), experience (p < 0.001), ethnic group (p < 0.001), the obstetric violence concept (p < 0.001) and academic year (p < 0.005). There were three significant multivariate models for the questionnaire’s overall score and dimensions.

CONCLUSIONS: Health sciences students perceived obstetric violence mainly as non-protocolised aspects while attending women. Degree studied and academic year might be related to perceived obstetric violence.

PMID:35051872 | DOI:10.1016/j.nedt.2022.105266

Epilepsy Behav. 2022 Jan 17;127:108538. doi: 10.1016/j.yebeh.2021.108538. Online ahead of print.

ABSTRACT

AIM: The aim of this study was to evaluate the potential impact of cannabidiol (CBD) on healthcare resource utilization and determine the effect of CBD on seizure-related emergency departments (ED) and hospital admissions in patients with epilepsy.

METHODS: This single-center, retrospective longitudinal cohort study included patients ≥1 year on CBD, excluding participants in CBD clinical trials or on <6 months of CBD therapy. Demographics, antiseizure medications (ASM), ED and hospital admissions were collected from the electronic medical record. Co-primary outcomes included change in seizure-related ED and hospital admissions. Secondary outcomes included change in ASMs and total ED or hospital admissions. Co-primary outcomes were assessed using generalized linear modeling. Descriptive statistics were used to analyze all other variables.

RESULTS: In the one-hundred total patients, there was no difference in either seizure-related ED visits (0.012 vs 0.011, p = 0.85) or hospital admissions per month (0.019 vs 0.021, p = 0.7). However, given the low percentage of the total cohort (n = 100) with either a seizure-related ED visits and hospital admissions (9% and 18%, respectively), a subgroup analysis was conducted. Those with seizure-related hospital admissions prior to CBD (n = 18) had significantly less seizure-related hospital admissions after initiation of CBD (23 admissions [0.104 per month] vs 15 admissions [0.055 per month], p = 0.007).

CONCLUSION: Despite the lack of statistically significant difference in primary outcomes for the total cohort, CBD may have a potential for a clinically beneficial impact in real-world settings on those patients with prior seizure-related admissions, who are the highest utilizers of healthcare resources.

PMID:35051868 | DOI:10.1016/j.yebeh.2021.108538

Comput Biol Med. 2022 Jan 11;142:105230. doi: 10.1016/j.compbiomed.2022.105230. Online ahead of print.

ABSTRACT

OBJECTIVE: To investigate the impact of harmonization on the performance of CT, PET, and fused PET/CT radiomic features toward the prediction of mutations status, for epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma viral oncogene (KRAS) genes in non-small cell lung cancer (NSCLC) patients.

METHODS: Radiomic features were extracted from tumors delineated on CT, PET, and wavelet fused PET/CT images obtained from 136 histologically proven NSCLC patients. Univariate and multivariate predictive models were developed using radiomic features before and after ComBat harmonization to predict EGFR and KRAS mutation statuses. Multivariate models were built using minimum redundancy maximum relevance feature selection and random forest classifier. We utilized 70/30% splitting patient datasets for training/testing, respectively, and repeated the procedure 10 times. The area under the receiver operator characteristic curve (AUC), accuracy, sensitivity, and specificity were used to assess model performance. The performance of the models (univariate and multivariate), before and after ComBat harmonization was compared using statistical analyses.

RESULTS: While the performance of most features in univariate modeling was significantly improved for EGFR prediction, most features did not show any significant difference in performance after harmonization in KRAS prediction. Average AUCs of all multivariate predictive models for both EGFR and KRAS were significantly improved (q-value < 0.05) following ComBat harmonization. The mean ranges of AUCs increased following harmonization from 0.87-0.90 to 0.92-0.94 for EGFR, and from 0.85-0.90 to 0.91-0.94 for KRAS. The highest performance was achieved by harmonized F_R0.66_W0.75 model with AUC of 0.94, and 0.93 for EGFR and KRAS, respectively.

CONCLUSION: Our results demonstrated that regarding univariate modelling, while ComBat harmonization had generally a better impact on features for EGFR compared to KRAS status prediction, its effect is feature-dependent. Hence, no systematic effect was observed. Regarding the multivariate models, ComBat harmonization significantly improved the performance of all radiomics models toward more successful prediction of EGFR and KRAS mutation statuses in lung cancer patients. Thus, by eliminating the batch effect in multi-centric radiomic feature sets, harmonization is a promising tool for developing robust and reproducible radiomics using vast and variant datasets.

PMID:35051856 | DOI:10.1016/j.compbiomed.2022.105230

Biosens Bioelectron. 2022 Jan 2;202:113955. doi: 10.1016/j.bios.2021.113955. Online ahead of print.

ABSTRACT

Proteins secreted by cells play significant roles in mediating many physiological, developmental, and pathological processes due to their functions in intra/intercellular communication and signaling. Conventional end-point methods are insufficient for understanding the temporal response in cell secretion process, which is often highly dynamic. Furthermore, cellular heterogeneity makes it essential to analyze secretory proteins from single cells. To uncover individual cellular activities and the underlying kinetics, new technologies are needed for real-time analysis of the secretomes of many cells at single-cell resolution. This study reports a high-throughput biosensing microarray platform, which is capable of label-free and real-time secretome monitoring from a large number of living single cells using a biochip integrating ultrasensitive nanoplasmonic substrate and microwell compartments having volumes of ∼0.4 nL. Precise synchronization of image acquisition and microscope stage movement of the developed optical platform enables spectroscopic analysis with high temporal and spectral resolution. In addition, our system allows simultaneous optical imaging of cells to track morphology changes for a comprehensive understanding of cellular behavior. We demonstrated the platform performance by detecting interleukin-2 secretion from hundreds of single lymphoma cells in real-time over many hours. Significantly, the analysis of the secretion kinetics allows us to study cellular response to the stimulations in a statistical way. The new platform is a promising tool for the characterization of single-cell functionalities given its versatility, throughput and label-free configuration.

PMID:35051850 | DOI:10.1016/j.bios.2021.113955

Comput Methods Programs Biomed. 2022 Jan 10;215:106624. doi: 10.1016/j.cmpb.2022.106624. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Lyme disease which is one of the most common infectious vector-borne diseases manifests itself in most cases with erythema migrans (EM) skin lesions. Recent studies show that convolutional neural networks (CNNs) perform well to identify skin lesions from images. Lightweight CNN based pre-scanner applications for resource-constrained mobile devices can help users with early diagnosis of Lyme disease and prevent the transition to a severe late form thanks to appropriate antibiotic therapy. Also, resource-intensive CNN based robust computer applications can assist non-expert practitioners with an accurate diagnosis. The main objective of this study is to extensively analyze the effectiveness of CNNs for diagnosing Lyme disease from images and to find out the best CNN architectures considering resource constraints.

METHODS: First, we created an EM dataset with the help of expert dermatologists from Clermont-Ferrand University Hospital Center of France. Second, we benchmarked this dataset for twenty-three CNN architectures customized from VGG, ResNet, DenseNet, MobileNet, Xception, NASNet, and EfficientNet architectures in terms of predictive performance, computational complexity, and statistical significance. Third, to improve the performance of the CNNs, we used custom transfer learning from ImageNet pre-trained models as well as pre-trained the CNNs with the skin lesion dataset HAM10000. Fourth, for model explainability, we utilized Gradient-weighted Class Activation Mapping to visualize the regions of input that are significant to the CNNs for making predictions. Fifth, we provided guidelines for model selection based on predictive performance and computational complexity.

RESULTS: Customized ResNet50 architecture gave the best classification accuracy of 84.42% ±1.36, AUC of 0.9189±0.0115, precision of 83.1%±2.49, sensitivity of 87.93%±1.47, and specificity of 80.65%±3.59. A lightweight model customized from EfficientNetB0 also performed well with an accuracy of 83.13%±1.2, AUC of 0.9094±0.0129, precision of 82.83%±1.75, sensitivity of 85.21% ±3.91, and specificity of 80.89%±2.95. All the trained models are publicly available at https://dappem.limos.fr/download.html, which can be used by others for transfer learning and building pre-scanners for Lyme disease.

CONCLUSION: Our study confirmed the effectiveness of even some lightweight CNNs for building Lyme disease pre-scanner mobile applications to assist people with an initial self-assessment and referring them to expert dermatologist for further diagnosis.

PMID:35051835 | DOI:10.1016/j.cmpb.2022.106624

J Commun Disord. 2022 Jan 17;95:106181. doi: 10.1016/j.jcomdis.2021.106181. Online ahead of print.

ABSTRACT

INTRODUCTION: A feasibility study of The Fluency Trust Residential Course (FTRC) for adolescents who stutter was conducted. The study aimed to measure key areas of a feasibility trial, for example, recruitment and retention, outcome measure completion, outcome measure reliability, and acceptability of the intervention to inform future research into the FTRC.

METHODS: Quantitative and qualitative methods were used. Participants were 23 adolescents (12-17 years), 23 parents and 2 Speech-Language Pathologists (SLPs) from the FTRC. Data collection included: outcome measure collection via a pre-test post-test quasi-experimental design (including two baseline measures), intervention fidelity checklists, semi-structured interviews with adolescents to explore acceptability of the intervention and semi-structured interviews with SLPs to explore their experiences of research participation and views on a future trial.

RESULTS: Recruitment, retention and outcome measure completion levels were all 100%. Intervention fidelity was 95% and there were no adverse events. Outcome measures showed good test- re-test reliability: Progress Questionnaire Child Intraclass Correlation Coefficient (ICC) = 0.87 (95% CI = 0.69-0.94 sig< 0.001) and Progress Questionnaire Parent ICC = 0.88 (95% CI = 0.70-0.95 sig< 0.001). Descriptive statistics showed that group medians and means of all outcome measures shifted in a positive direction between pre and post-tests (9 weeks follow-up). Twenty-five percent of young people showed changes on the Progress Questionnaire Child that were above the minimal important difference. Seventy-five percent of parents showed changes on the Progress Questionnaire Parent that were above the minimal important difference. Acceptability of the intervention by adolescents was high. SLPs reported participation was manageable and they were pleased to be part of the research.

CONCLUSION: Quantitative and qualitative data suggest that a future definitive trial of the FTRC is indicated after additional development work and feasibility testing. Recommendations for further research are included.

PMID:35051833 | DOI:10.1016/j.jcomdis.2021.106181

Am J Dent. 2021 Dec;34(6):338-344.

ABSTRACT

PURPOSE: To compare the plaque removal efficacy of a new children’s powered toothbrush to a children’s manual toothbrush.

METHODS: 55 subjects aged 5-8 years old, who met entry criteria, participated in this single-center, two-cell, examiner-blind, randomized, crossover, single use clinical study. Subjects brushed at home with their first assigned toothbrush and fluoride toothpaste, under supervision of a parent or legal guardian, at least once daily for 2 minutes during a 1-week acclimation period. After refraining from oral hygiene for 12-16 hours, and from eating and drinking for 4 hours, subjects returned to the clinical site where they were assessed for plaque using the Rustogi Modified Navy Plaque Index (RMNPI). Subjects then brushed their teeth with their assigned toothbrush and toothpaste for 2 minutes and plaque levels were reassessed. Subjects were then given their second assigned toothbrush and the acclimation period and clinical site visit were repeated. Safety-in-use was also assessed during each clinic visit. Differences between pre-and post-brushing scores were analyzed for each toothbrush and between toothbrush groups for whole mouth plaque and 12 subset sites using baseline adjusted ANCOVA.

RESULTS: Both toothbrushes significantly (P< 0.0001) reduced whole mouth and 12 subset site plaque scores from the pre-brushing baseline. Between treatment comparisons showed that use of the powered toothbrush resulted in statistically significant reductions in whole mouth plaque (55%, P< 0.0001) and in 12 subset site scores (40-208%) compared to the manual brush. This clinical study showed that brushing with a new children’s powered toothbrush was safe and significantly more effective than brushing with a manual toothbrush in reducing whole mouth plaque scores, as well as plaque scores at a range of subset sites in the mouth.

CLINICAL SIGNIFICANCE: This new powered toothbrush may enable children to safely achieve significant and meaningful improvements in oral hygiene compared to brushing with a manual toothbrush.

PMID:35051323

Am J Dent. 2021 Dec;34(6):327-332.

ABSTRACT

PURPOSE: To evaluate the marginal sealing ability of different restorative materials used in deep margin elevation (DME) on zirconia-reinforced lithium silicate CAD-CAM ceramic restorations.

METHODS: A total of 30 Class II cavities were prepared in freshly extracted human molars with the proximal margin located 1 mm below the cemento-enamel junction (CEJ). All specimens were randomly assigned to one of three groups (n=10): control group, resin composite group (Filtek Z350 XT), and resin-modified glass-ionomer group (RMGI) (Vitremer Tricure). In Group 1, control group, no DME was performed. The inlay margin of the control group was placed directly on the dentin. In Groups 2 and 3, DME was used to elevate the margin to 1 mm above the CEJ with resin composite and RMGI, respectively. Zirconia-reinforced lithium silicate CAD-CAM ceramic restorations were manufactured and bonded on all specimens with universal bonding and resin luting cement. All specimens were aged by water storage for 6 months. Marginal sealing ability at different interfaces was evaluated with a stereomicroscope at 40x magnification by scoring the depth of silver nitrate penetrating along the adhesive surfaces. Statistical differences between groups were analyzed using the Kruskal-Wallis and Mann-Whitney U tests.

RESULTS: At the dentin interface, there was no significant difference in microleakage scores in the control group and resin composite group (P= 0.577); however, the RMGI group had significantly higher microleakage compared to the control group (P= 0.004) and resin composite group (P= 0.007).

CLINICAL SIGNIFICANCE: Deep margin elevation can be achieved with resin composite. Resin-modified glass-ionomer must be used with caution due to the high microleakage scores.

PMID:35051321