Receiver operating characteristic (ROC) curve for medical researchers

Predicting the Risk of Driving Under the Influence of Alcohol Using EEG-Based Machine Learning

Driving under the influence of alcohol (DUIA) is closely associated with alcohol use disorder (AUD). Our previous study on machine learning (ML) algorithms revealed a very high accuracy of decision trees with neuropsychological features in predicting the risk of DUIA despite limited data availability. Thus, this study aimed at comparing six well-known ML algorithms based on electroencephalographic (EEG) signals to differentiate adults with AUD and DUIA (AUD-DD) from those with AUD without DUIA (AUD-NDD) and controls. Fifteen AUD-DD and 10 AUD-NDD participants were recruited from a single tertiary referral center. Fourteen social drinkers without DUIA served as controls. Their EEG signals related to driving conditions were gathered using a VR headset with eight electrodes (F3, F4, Fz, C3, C4, Cz, P3, and P4). Based on the labeled features of EEG asymmetry and theta/beta ratio (TBR), comparisons between different algorithms were conducted. Fz and Cz electrodes exhibited differences in TBR across the three groups (all p < 0.02), while there were no significant differences between AUD-DD individuals and social drinkers. In contrast, asymmetries of between-group differences were not observed (all p > 0.09). K-nearest neighbors (KNN) with TBR showed the highest accuracy (83 %) in distinguishing AUD-DD individuals from controls, while logistic regression (LR), support vector machines (SVM), and naive Bayes (NB) with EEG asymmetric features demonstrated high accuracy in identifying DUIA (all 80 %) in AUD adults. LR, SVM, and NB with asymmetry may be employed in predicting DUIA among AUD adults, while KNN with TBR may be used for identifying DUIA in the general population.

Decoding skin cancer classification: perspectives, insights, and advances through researchers' lens

Skin cancer is a significant global health concern, with timely and accurate diagnosis playing a critical role in improving patient outcomes. In recent years, computer-aided diagnosis systems have emerged as powerful tools for automated skin cancer classification, revolutionizing the field of dermatology. This survey analyzes 107 research papers published over the last 18 years, providing a thorough evaluation of advancements in classification techniques, with a focus on the growing integration of computer vision and artificial intelligence (AI) in enhancing diagnostic accuracy and reliability. The paper begins by presenting an overview of the fundamental concepts of skin cancer, addressing underlying challenges in accurate classification, and highlighting the limitations of traditional diagnostic methods. Extensive examination is devoted to a range of datasets, including the HAM10000 and the ISIC archive, among others, commonly employed by researchers. The exploration then delves into machine learning techniques coupled with handcrafted features, emphasizing their inherent limitations. Subsequent sections provide a comprehensive investigation into deep learning-based approaches, encompassing convolutional neural networks, transfer learning, attention mechanisms, ensemble techniques, generative adversarial networks, vision transformers, and segmentation-guided classification strategies, detailing various architectures, tailored for skin lesion analysis. The survey also sheds light on the various hybrid and multimodal techniques employed for classification. By critically analyzing each approach and highlighting its limitations, this survey provides researchers with valuable insights into the latest advancements, trends, and gaps in skin cancer classification. Moreover, it offers clinicians practical knowledge on the integration of AI tools to enhance diagnostic decision-making processes. This comprehensive analysis aims to bridge the gap between research and clinical practice, serving as a guide for the AI community to further advance the state-of-the-art in skin cancer classification systems.

Integrated analysis ceRNA network of autophagy-related gene RNF144B in steroid-induced necrosis of the femoral head

This study aimed to investigate the regulatory mechanisms that influenced autophagy in Steroid-induced necrosis of the femoral head (SONFH) by constructing a competing endogenous RNA (ceRNA) network. Blood sample data from the SONFH patients were obtained from the Gene Expression Omnibus (GEO) database under the accession number GSE123568. Autophagy-related genes were identified from the Human Autophagy Database (HADb). Differential analysis and weighted gene co-expression network analysis (WGCNA) were performed on the GSE123568 dataset to screen for core genes and validation was performed with the validation set. Based on the GEO dataset (GSE74089), we performed differential lncRNA analysis. Meanwhile, we utilized three databases, namely miRDB, TargetScan, and StarBase, to predict the miRNAs of target genes and corresponding lncRNAs. Cytoscape software was used to construct and visualize the ceRNA networks. We also employed reverse transcription-quantitative polymerase chain reaction (RT-qPCR) to quantify their expression levels. A total of 1692 differentially expressed genes (DEGs) were identified in the GSE123568 dataset. By intersecting with the HADb database, 47 autophagy-related genes were identified from these DEGs. Furthermore, we found the significant correlation between RNF144B and 37 autophagy genes. Importantly, we established a regulatory axis involving TUG1, hsa-miR-31-5p, and RNF144B., and both TUG1 and RNF144B were upregulated, while hsa-miR-31-5p was downregulated in the SONFH cell model. A TUG1-hsa-miR-31-5p-RNF144B axis was related to autophagy genes, which potentially provided insights into the RNA interactions triggering autophagy in SONFH.

Diagnostic Performance of Artificial Intelligence in Chest Radiographs Referred from the Emergency Department

BACKGROUND

The increasing integration of AI in chest X-ray evaluation holds promise for enhancing diagnostic accuracy and optimizing clinical workflows. However, understanding its performance in real-world clinical settings is essential.

OBJECTIVES

In this study, we evaluated the sensitivity (Se) and specificity (Sp) of an AI-based software (Arterys MICA v29.4.0) alongside a radiology resident in interpreting chest X-rays referred from the emergency department (ED), using a senior radiologist's assessment as the gold standard (GS). We assessed the concordance between the AI system and the resident, noted the frequency of doubtful cases for each category, identified how many were considered positive by the GS, and assessed variables that AI was not trained to detect.

METHODS

We conducted a retrospective observational study analyzing chest X-rays from a sample of 784 patients referred from the ED at our hospital. The AI system was trained to detect five categorical variables-pulmonary nodule, pulmonary opacity, pleural effusion, pneumothorax, and fracture-and assign each a confidence label ("positive", "doubtful", or "negative").

RESULTS

Sensitivity in detecting fractures and pneumothorax was high (100%) for both AI and the resident, moderate for pulmonary opacity (AI = 76%, resident = 71%), and acceptable for pleural effusion (AI = 60%, resident = 67%), with negative predictive values (NPV) above 95% and areas under the curve (AUC) exceeding 0.8. The resident showed moderate sensitivity (75%) for pulmonary nodules, while AI's sensitivity was low (33%). AI assigned a "doubtful" label to some diagnoses, most of which were deemed negative by the GS; the resident expressed doubt less frequently. The Kappa coefficient between the resident and AI was fair (0.3) across most categories, except for pleural effusion, where concordance was moderate (0.5). Our study highlighted additional findings not detected by AI, including 16% prevalence of mediastinal abnormalities, 20% surgical materials, and 20% other pulmonary findings.

CONCLUSIONS

Although AI demonstrated utility in identifying most primary findings-except for pulmonary nodules-its high NPV suggests it may be valuable for screening. Further training of the AI software and broadening its scope to identify additional findings could enhance its detection capabilities and increase its applicability in clinical practice.

The mediation effect of liver and anthropometric indices on the relationship between incidence of diabetes and physical activity: results of 5-year follow up azar cohort study

BACKGROUND

It has been documented that regular physical activity is considered one of the most effective strategies for preventing diabetes; however, it is not the sole contributing factor. Therefore, we decided to evaluate the meditation effect of liver function and anthropometric indices on the relationship between incidence of diabetes and physical activity (PA) in the Azar cohort population.

MATERIALS AND METHODS

Subjects who were diabetic in the baseline phase from 15,006 participants in study of azar cohort population were excluded and to follow up, a total of 13,253 people was included in the analysis. Demographic characteristics, physical activity, 10 anthropometric indices (AI) and seven liver indices (LI) were measured. Evaluated and displayed using Pearson correlation heatmap and canonical correlation of liver and anthropometric indices. The Generalized Structural Equation Modeling (GSEM) with the Maximum Likelihood method employed to estimate the model.

RESULTS

During the follow-up years, a total of 685 participants developed diabetes. The measurements of the AI were significantly higher in subjects with diabetes (P < .001). Patients with diabetes were older, had a higher proportion of women, and had lower values of PA (P < .05). Body Roundness Index (BRI) and Waist height ratio (WHtR) exhibited the largest AUCs for predicting diabetes onset risk (both AUC = 0.6989) among these anthropometric measures. The increase in AI (RR [95%CI] = 1.25 [1.22,1.29], P < .001) and liver enzyme (LE) (RR [95%CI] = 1.14 [1.08.1.19], P < .001) increase the risk of diabetes by 25% and 14%, respectively. Despite the mediation effects of AI and Liver Enzymes for an increase of one MET of PA, the risk of developing diabetes decreases by 5% (RR [95% CI] = .95 [.92,.99], P = .013). Around VAF = 53% of the association between PA and diabetes onset (Total effect: RR [95% CI] = .90 [.87,.94], P < .001) was mediated by AI and LE.

CONCLUSIONS

A low level of PA was found to be significantly correlated with high levels of AI and LI, all of which are associated with an increased risk of developing diabetes. These analyses provide evidence that when the relationship between PA and diabetes is mediated by AI and LI this association becomes stronger, with AI playing a more significant role than LI.

Emotional Dysregulation in Emerging Adult ADHD: A Key Consideration in Explaining and Classifying Impairment and Co-Occurring Internalizing Problems

OBJECTIVE

The current study sought to clarify and harness the incremental validity of emotional dysregulation and unawareness (EDU) in emerging adulthood, beyond ADHD symptoms and with respect to concurrent classification of impairment and co-occurring problems, using machine learning techniques.

METHOD

Participants were 1,539 college students (Mage = 19.5, 69% female) with self-reported ADHD diagnoses from a multisite study who completed questionnaires assessing ADHD symptoms, EDU, and co-occurring problems.

RESULTS

Random forest analyses suggested EDU dimensions significantly improved model performance (ps < .001) in classifying participants with impairment and internalizing problems versus those without, with the resulting ADHD + EDU classification model demonstrating acceptable to excellent performance (except in classification of Work Impairment) in a distinct sample. Variable importance analyses suggested inattention sum scores and the Limited Access to Emotional Regulation Strategies EDU dimension as the most important features for facilitating model classification.

CONCLUSION

Results provided support for EDU as a key deficit in those with ADHD that, when present, helps explain ADHD's co-occurrence with impairment and internalizing problems. Continued application of machine learning techniques may facilitate actuarial classification of ADHD-related outcomes while also incorporating multiple measures.

Nocturnal pulse oximetry for the detection and prediction of acute mountain sickness: An observational study

Acute mountain sickness (AMS) is a well-studied illness defined by clinical features (e.g., headache and nausea), as assessed by the Lake Louise score (LLS). Although obvious in its severe form, early stages of AMS are poorly defined and easily confused with common travel-related conditions. Measurement of hypoxaemia, the cause of AMS, should be helpful, yet to date its utility for identifying AMS susceptibility remains unclear. This study quantified altitude-induced hypoxaemia in individuals during an ascent to 4800 m to determine the utility of nocturnal pulse oximetry measurements for prediction of AMS. Eighteen individuals (36 ± 16 years of age) ascended to 4800 m over 12 days. Symptomology of AMS was assessed each morning via LLS criteria, with participants categorized as either AMS-positive (LLS ≥ 3 with headache) or AMS-negative. Overnight peripheral oxygen saturations (ov-S p O 2 ${{S}_{{\mathrm{p}}{{{\mathrm{O}}}_2}}}$ ) were recorded continuously (1 Hz) using portable oximeters. Derivatives of these recordings were compared between AMS-positive and -negative subjects (Mann-Whitney U-test). Exploratory analyses (Pearson's) were conducted to investigate relationships between overnight parameters and AMS severity. Overnight derivatives, including ov-S p O 2 ${{S}_{{\mathrm{p}}{{{\mathrm{O}}}_2}}}$ , heart rate/ov-S p O 2 ${{S}_{{\mathrm{p}}{{{\mathrm{O}}}_2}}}$ , variance, oxygen desaturation index, hypoxic burden and total sleep time at <80%S p O 2 ${{S}_{{\mathrm{p}}{{{\mathrm{O}}}_2}}}$ , all differed significantly between AMS-positive and -negative subjects (all P < 0.01), with cumulative/relative frequency plots highlighting these differences visually. Exploratory analysis revealed that ov-S p O 2 ${{S}_{{\mathrm{p}}{{{\mathrm{O}}}_2}}}$ from 3850 m was correlated with peak LLS at 4800 m (r = 0.58-0.61). The findings highlight the potential for overnight oximetry to predict AMS susceptibility during ascent to high altitude. Further investigation is required to develop, evaluate and optimize predictive models to improve AMS management and prevention.

Multi-modality multi-label ocular abnormalities detection with transformer-based semantic dictionary learning

Blindness is preventable by early detection of ocular abnormalities. Computer-aided diagnosis for ocular abnormalities is built by analyzing retinal imaging modalities, for instance, Color Fundus Photography (CFP). This research aims to propose a multi-label detection of 28 ocular abnormalities consisting of frequent and rare abnormalities from a single CFP by using transformer-based semantic dictionary learning. Rare labels are usually ignored because of a lack of features. We tackle this condition by adding the co-occurrence dependency factor to the model from the linguistic features of the labels. The model learns the relation between spatial features and linguistic features represented as a semantic dictionary. The proposed method treats the semantic dictionary as one of the main important parts of the model. It acts as the query while the spatial features are the key and value. The experiments are conducted on the RFMiD dataset. The results show that the proposed method achieves the top 30% in Evaluation Set on the RFMiD dataset challenge. It also shows that treating the semantic dictionary as one of the strong factors in model detection increases the performance when compared with the method that treats the semantic dictionary as a weak factor.

Is blue light imaging without magnification satisfactory as screening for esophageal squamous cell carcinoma? Post-hoc analysis of multicenter randomized controlled trial

OBJECTIVES

Narrow light observation is currently recommended as an alternative to Lugol chromoendoscopy (LCE) to detect esophageal squamous cell carcinoma (ESCC). Studies revealed little difference in sensitivity between the two modalities in expert settings; however, these included small numbers of cases. We aimed to determine whether blue light imaging (BLI) without magnification is satisfactory for preventing misses of ESCC.

METHODS

This was a post-hoc analysis of a multicenter randomized controlled trial targeting patients at high risk of ESCC in expert settings. In this study, BLI without magnification followed by LCE was performed. The evaluation parameters included: (i) the diagnostic abilities of ESCC; (ii) the endoscopic characteristics of lesions with diagnostic differences between the two modalities; and (iii) the color difference between cancerous and noncancerous areas in BLI and LCE.

RESULTS

This study identified ESCC in 49 of 699 cases. Of these cases, nine (18.4%) were missed by BLI but detected by LCE. In per-patient analysis, the sensitivity of BLI was lower than that of LCE following BLI (83.7% vs. 100.0%; P = 0.013), whereas the specificity and accuracy of BLI were higher (88.2% vs. 81.2%; P < 0.001 and 87.8% vs. 82.5%; P < 0.001, respectively). No significant endoscopic characteristics were identified, but the color difference was lower in BLI than in LCE (21.4 vs. 25.1; P = 0.003).

CONCLUSION

LCE following BLI outperformed BLI in terms of sensitivity in patients with high-risk ESCC. Therefore, LCE, in addition to BLI, would still be required in screening esophagogastroduodenoscopy even by expert endoscopists.

An initial exploration of factors that may impact radiographer performance in reporting mammograms

OBJECTIVES

In the United Kingdom, radiographers with a qualification in image interpretation have interpreted mammograms since 1995. These radiographers work under the title of radiography advanced practitioners (RAP) or Consultant Radiographer. This study extends upon what has been very recently published by exploring further clinical, non-clinical and experiential factors that may impact the reporting performance of RAPs.

METHODS

Fifteen RAPs interpreted an image test set of 60 2D mammograms of known truth using the Detected-X software platform. Unknown to the reader, twenty cases contained a malignancy. Sensitivity, specificity, lesion sensitivity, receiver operating characteristic (ROC) and jack-knife free response operating characteristic (AFROC) values were established for each RAP. Specific features that had significant impact on accuracy were identified using Student's-T and Mann Whitney tests.

RESULTS

RAPs with more than 10 years' experience in image interpretation, compared to those with less than 10 years' experience, demonstrated lower specificity (51.3% vs 84.8%, p = 0.0264), ROC (0.83 vs 0.91, p = 0.0264) and AFROC (0.75 vs 0.87, p = 0.0037) values. Further, higher sensitivity values of 90.7% were seen in those RAPs who had an eye test in the last year compared to those who had not, 82% (p = 0.021). Other changes are presented in the paper.

CONCLUSION

These data reveal previously unidentified factors that impact the diagnostic efficacy of RAPs when interpreting mammographic images. Highlighting such findings will empower screening authorities to better examine ways of standardising performance and offer a baseline for performance benchmarks.

IMPLICATIONS FOR PRACTICE

This study for the first time performs an initial exploration of the factors that may be associated with RAP performance when interpreting screening mammograms.

Non-invasive detection of bilirubin concentrations during the first week of life in a low-resource setting along the Thailand-Myanmar border

BACKGROUND

Neonatal hyperbilirubinaemia (NH) is a common problem worldwide and is a cause of morbidity and mortality especially in low-resource settings.

METHODS

A study was carried out at Shoklo Malaria Research Unit (SMRU) clinics along the Thailand-Myanmar border to evaluate a non-invasive test for diagnosis of NH in a low-resource setting. Performance of a transcutaneous bilirubinometer Dräger Jaundice Meter JM-105 was assessed against routine capillary serum bilirubin testing (with BR-501 microbilirubinometer) before phototherapy during neonatal care in the first week of life. Results were analysed by direct agreement and by various bilirubin thresholds used in clinical practice. Total serum bilirubin was also measured in cord blood at birth and tested for prediction of hyperbilirubinaemia requiring phototherapy in the first week of life.

RESULTS

Between April 2020 and May 2023, 742 neonates born at SMRU facilities were included in the study. A total of 695 neonates provided one to nine capillary blood samples for analysis of serum bilirubin (total 1244 tests) during the first week of life. Performance of transcutaneous bilirubinometer was assessed in 307 neonates who provided 687 paired transcutaneous capillary blood tests. Bilirubin levels were also measured in 738 cord blood samples. Adjusted values of transcutaneous bilirubinometer showed excellent agreement with capillary serum bilirubin concentration (intraclass correlation coefficient=0.923) and high sensitivity (>98%) at all clinical thresholds analysed across 3 years of sampling and multiple users. Concentrations of bilirubin detected in cord blood were not useful in identifying neonates at risk of hyperbilirubinaemia requiring treatment.

CONCLUSIONS

The transcutaneous bilirubinometer is a reliable tool to screen neonates and identify those needing confirmatory blood testing. Bilirubin concentrations in cord blood are not predictive of hyperbilirubinaemia in neonates.

Novel prediction model of early screening lung adenocarcinoma with pulmonary fibrosis based on haematological index

BACKGROUND

Lung cancer (LC), a paramount global life-threatening condition causing significant mortality, is most commonly characterized by its subtype, lung adenocarcinoma (LUAD). Concomitant with LC, pulmonary fibrosis (PF) and interstitial lung disease (ILD) contribute to an intricate landscape of respiratory diseases. Idiopathic pulmonary fibrosis (IPF) in association with LC has been explored. However, other fibrotic interrelations remain underrepresented, especially for LUAD-PF and LUAD-ILD.

METHODS

We analysed data with statistical analysis from 7,137 healthy individuals, 7,762 LUAD patients, 7,955 ILD patients, and 2,124 complex PF patients collected over ten years. Furthermore, to identify blood indicators related to lung disease and its complications and compare the relationships between different indicators and lung diseases, we successfully applied the naive Bayes model for a biomarker-based prediction of diagnosis and development into complex PF.

RESULTS

Males predominantly marked their presence in all categories, save for complex PF where females took precedence. Biomarkers, specifically AGR, MLR, NLR, and PLR emerged as pivotal in discerning lung diseases. A machine-learning-driven predictive model underscored the efficacy of these markers in early detection and diagnosis, with NLR exhibiting unparalleled accuracy.

CONCLUSIONS

Our study elucidates the gender disparities in lung diseases and illuminates the profound potential of serum biomarkers, including AGR, MLR, NLR, and PLR in early lung cancer detection. With NLR as a standout, therefore, this study advances the exploration of indicator changes and predictions in patients with pulmonary disease and fibrosis, thereby improving early diagnosis, treatment, survival rate, and patient prognosis.

NAT10-mediated ac4C acetylation of TFRC promotes sepsis-induced pulmonary injury through regulating ferroptosis

BACKGROUND

Sepsis-induced pulmonary injury (SPI) is a common complication of sepsis with a high rate of mortality. N4-acetylcytidine (ac4C) is mediated by the ac4C "writer", N-acetyltransferase (NAT)10, to regulate the stabilization of mRNA. This study aimed to investigate the role of NAT10 in SPI and the underlying mechanism.

METHODS

Twenty-three acute respiratory distress syndrome (ARDS) patients and 27 non-ARDS volunteers were recruited. A sepsis rat model was established. Reverse transcription-quantitative polymerase chain reaction was used to detect the expression of NAT10 and transferrin receptor (TFRC). Cell viability was detected by cell counting kit-8. The levels of Fe2+, glutathione, and malondialdehyde were assessed by commercial kits. Lipid reactive oxygen species production was measured by flow cytometric analysis. Western blot was used to detect ferroptosis-related protein levels. Haematoxylin & eosin staining was performed to observe the pulmonary pathological symptoms.

RESULTS

The results showed that NAT10 was increased in ARDS patients and lipopolysaccharide-treated human lung microvascular endothelial cell line-5a (HULEC-5a) cells. NAT10 inhibition increased cell viability and decreased ferroptosis in HULEC-5a cells. TFRC was a downstream regulatory target of NAT10-mediated ac4C acetylation. Overexpression of TFRC decreased cell viability and promoted ferroptosis. In in vivo study, NAT10 inhibition alleviated SPI.

CONCLUSION

NAT10-mediated ac4C acetylation of TFRC aggravated SPI through promoting ferroptosis.

Construction of circRNA-mediated ceRNA network and immunoassay for investigating pathogenesis of COPD

Background

The chronic respiratory condition known as chronic obstructive pulmonary disease (COPD) was one of the main causes of death and disability worldwide. This study aimed to explore and elucidate new targets and molecular mechanisms of COPD by constructing competitive endogenous RNA (ceRNA) networks.

Methods

GSE38974 and GSE106986 were used to select DEGs in COPD samples and normal samples. Cytoscape software was used to construct and present protein-protein interaction (PPI) network, mRNA-miRNA co-expression network and ceRNA network. The CIBERSORT algorithm and the Lasso model were used to screen the immune infiltrating cells and hub genes associated with COPD, and the correlation between them was analyzed. COPD cell models were constructed in vitro and the expression level of ceRNA network factors mediated by hub gene was detected by reverse transcription-quantitative polymerase chain reaction (RT-qPCR).

Results

In this study, 852 differentially expressed genes were screened in the GSE38974 dataset, including 439 upregulated genes and 413 downregulated genes. Gene clustering analysis of PPI network results was performed using the Minimum Common Tumor Data Element (MCODE) in Cytoscape, and seven hub genes were screened using five algorithms in cytoHubba. CCL20 was verified as an important hub gene based on mRNA-miRNA co-expression network, GSE106986 database validation and the analysis of ROC curve results. Finally, we successfully constructed the circDTL-hsa-miR-330-3p-CCL20 network by Cytoscape. Immune infiltration analysis suggested that CCL20 can co-regulate immune cell migration and infiltration through chemokines CCL7 and CXCL3. In vitro experiments, the expression of circDTL and CCL20 was increased, while the expression of hsa-miR-330-3p was decreased in the COPD cell model.

Conclusion

By constructing the circDTL-hsa-miR-330-3p-CCL20 network, this study contributes to a better understanding of the molecular mechanism of COPD development, which also provides important clues for the development of new therapeutic strategies and drug targets.

Clinical utility of the Saint Louis University Mental Status Examination (SLUMS) in a mixed neurological sample: Proposed revised cutoff scores for normal cognition, mild cognitive impairment, and dementia

Early detection of cognitive impairment is of paramount importance in clinical settings, with several brief screening tools having been developed for that purpose. The present study sought to evaluate the clinical utility of the Saint Louis University Mental Status examination (SLUMS) at identifying examinees with normal cognition, mild cognitive impairment, or dementia syndrome using the criterion of a comprehensive neuropsychological assessment. Two hundred sixty-three examinees (M age = 67.84 ± 12.72; 59.3% female; 81.4% white) were referred for comprehensive neuropsychological evaluation at a private, Mid-Atlantic medical center. Using original cutoff scores, the SLUMS correctly classified just over half (55.1%) of examinees. Classification statistics suggested modified cutoff scores for mild cognitive impairment (≤24) and dementia (≤17) with strong discriminability between cognitive status groups (AUCs ranged from .834 to .986). These proposed revised cutoff scores improved overall concordance between SLUMS and diagnostic conclusions from comprehensive clinical neuropsychological testing, correctly classifying nearly two-thirds of examinees (65.4%). The SLUMS and its revised cutoff scores appear to have clinical utility for cognitive screening in primary care and neurological settings to inform treatment plans and appropriate referrals for comprehensive neuropsychological assessment.

Mesoscale air motion and thermodynamics predict heavy hourly U.S. precipitation

Predicting heavy precipitation remains scientifically challenging. Here we combine Atmospheric Infrared Sounder (AIRS) temperature and moisture soundings and weather forecast winds to predict the formation of thermodynamic conditions favourable for convection in the hours following satellite overpasses. Here we treat AIRS retrievals as air parcels that are moved adiabatically to generate time-varying fields. Over much of the Central-Eastern Continental U.S. during the non-winter months of 2019-2020, our derived convective available potential energy alone predicts intense precipitation. For hourly precipitation above the all-hours 99.9th percentile, performance is marginally lower than forecasts from a convection permitting model, but similar to the ERA5 reanalysis and substantially better than using the original AIRS soundings. Our results illustrate how mesoscale advection is a major contributor to developing heavy precipitation in the region. Enhancing the full AIRS record as described here would provide an alternative approach to quantify multi-decade trends in heavy precipitation risk.

Precision mapping of snail habitat in lake and marshland areas: Integrating environmental and textural indicators using Random Forest modeling

Schistosomiasis japonica continues to pose a significant public health challenge in China, primarily due to the widespread distribution of Oncomelania hupensis, the sole intermediate host of Schistosoma. This study aims to address the constraints of existing remote sensing analyses for identifying snail habitats, which frequently neglect spatial scale and seasonal variations. To this end, we adopt a multi-source data-driven Random Forest approach that integrates bottomland and ground-surface texture data with traditional environmental variables, enhancing the accuracy of snail habitat assessments. We developed four distinct models for the lake and marshland areas of Guichi, China: a baseline model incorporating ground-surface texture, bottomland variables, and environmental variables; Model 1 with only environmental variables; Model 2 adding ground-surface texture and environmental variables; and Model 3 integrating bottomland with environmental variables. The baseline model outperformed the others, achieving a true skill statistic of 0.93, an accuracy of 0.97, a kappa statistic of 0.94, and an area under the curve of 0.99. Our analysis pinpointed critical high-risk snail habitats distributed in a belt-like pattern along major water bodies, near the Yangtze River, QiuPu River, and around Shengjin Lake, Jiuhua River, and Qingtong River. These insights can aid local health authorities in more efficiently allocating limited resources, developing effective snail surveillance and control strategies to combat schistosomiasis. Additionally, this approach can be adapted to localize other endemic hosts with similar ecological characteristics.

Prediction of Clinical Outcomes in Psychotic Disorders Using Artificial Intelligence Methods: A Scoping Review

BACKGROUND

Psychotic disorders are major psychiatric disorders that can impact multiple domains including physical, social, and psychological functioning within individuals with these conditions. Being able to better predict the outcomes of psychotic disorders will allow clinicians to identify illness subgroups and optimize treatment strategies in a timely manner.

OBJECTIVE

In this scoping review, we aimed to examine the accuracy of the use of artificial intelligence (AI) methods in predicting the clinical outcomes of patients with psychotic disorders as well as determine the relevant predictors of these outcomes.

METHODS

This review was guided by the PRISMA Guidelines for Scoping Reviews. Seven electronic databases were searched for relevant published articles in English until 1 February 2024.

RESULTS

Thirty articles were included in this review. These studies were mainly conducted in the West (63%) and Asia (37%) and published within the last 5 years (83.3%). The clinical outcomes included symptomatic improvements, illness course, and social functioning. The machine learning models utilized data from various sources including clinical, cognitive, and biological variables such as genetic, neuroimaging measures. In terms of main machine learning models used, the most common approaches were support vector machine, random forest, logistic regression, and linear regression models. No specific machine learning approach outperformed the other approaches consistently across the studies, and an overall range of predictive accuracy was observed with an AUC from 0.58 to 0.95. Specific predictors of clinical outcomes included demographic characteristics (gender, socioeconomic status, accommodation, education, and employment); social factors (activity level and interpersonal relationships); illness features (number of relapses, duration of relapses, hospitalization rates, cognitive impairments, and negative and disorganization symptoms); treatment (prescription of first-generation antipsychotics, high antipsychotic doses, clozapine, use of electroconvulsive therapy, and presence of metabolic syndrome); and structural and functional neuroimaging abnormalities, especially involving the temporal and frontal brain regions.

CONCLUSIONS

The current review highlights the potential and need to further refine AI and machine learning models in parsing out the complex interplay of specific variables that contribute to the clinical outcome prediction of psychotic disorders.

A semi-structured interview for the dimensional assessment of internalizing and externalizing symptoms in children and adolescents: Interview Version of the Symptoms and Functioning Severity Scale (SFSS-I)

BACKGROUND

This study evaluates the psychometric properties of the newly developed semi-structured interview, Interview Version of the Symptoms and Functioning Severity Scale (SFSS-I), which is designed to provide a dimensional assessment of internalizing and externalizing symptoms.

METHODS

Multi-informant baseline data from the OPTIE study was used, involving 358 children and adolescents aged 6 to 17 years (M = 11.54, SD = 3.4, n = 140 [39.1%] were female). Participants were screened for internalizing and externalizing symptoms. For validity analyses, caregiver (Child Behavior Checklist), youth (Youth Self Report), and teacher ratings (Teacher Report Form) were used. We performed Receiver Operating Characteristic (ROC) analyses to evaluate the effectiveness of the SFSS-I subscales in distinguishing between children and adolescents diagnosed with internalizing and externalizing disorders, as determined by clinical judgement in routine care.

RESULTS

Confirmatory factor analyses supported a correlated two-factor model for internalizing and externalizing symptoms. Acceptable to good internal consistencies (α = 0.76 to 0.89; ω = 0.76 to 0.90) and excellent interrater reliability on the scale level (ICC ≥ 0.91) was found. The ROC analyses showed an acceptable accuracy in identifying internalizing diagnoses (AUC = 0.76) and excellent accuracy for externalizing diagnoses (AUC = 0.84).

CONCLUSION

The SFSS-I demonstrates potential as a clinically-rated instrument for screening and routine outcome monitoring, offering utility in both clinical practice and research settings for the dimensional assessment of broad psychopathological dimensions.

TRIAL REGISTRATION

German Clinical Trials Register (DRKS) DRKS00016737 ( https://www.drks.de/DRKS00016737 ). Registered 17 September, 2019.

Application of multivariate binary logistic regression grouped outlier statistics and geospatial logistic model to identify villages having unusual health-seeking habits for childhood malaria in Malawi

BACKGROUND

Early diagnosis and prompt treatment of malaria in young children are crucial for preventing the serious stages of the disease. If delayed treatment-seeking habits are observed in certain areas, targeted campaigns and interventions can be implemented to improve the situation.

METHODS

This study applied multivariate binary logistic regression model diagnostics and geospatial logistic model to identify traditional authorities in Malawi where caregivers have unusual health-seeking behaviour for childhood malaria. The data from the 2021 Malawi Malaria Indicator Survey were analysed using R software version 4.3.0 for regressions and STATA version 17 for data cleaning.

RESULTS

Both models showed significant variability in treatment-seeking habits of caregivers between villages. The mixed-effects logit model residual identified Vuso Jere, Kampingo Sibande, Ngabu, and Dzoole as outliers in the model. Despite characteristics that promote late reporting of malaria at clinics, most mothers in these traditional authorities sought treatment within twenty-four hours of the onset of malaria symptoms in their children. On the other hand, the geospatial logit model showed that late seeking of malaria treatment was prevalent in most areas of the country, except a few traditional authorities such as Mwakaboko, Mwenemisuku, Mwabulambya, Mmbelwa, Mwadzama, Zulu, Amidu, Kasisi, and Mabuka.

CONCLUSIONS

These findings suggest that using a combination of multivariate regression model residuals and geospatial statistics can help in identifying communities with distinct treatment-seeking patterns for childhood malaria within a population. Health policymakers could benefit from consulting traditional authorities who demonstrated early reporting for care in this study. This could help in understanding the best practices followed by mothers in those areas which can be replicated in regions where seeking care is delayed.

Screening accuracy and cut-offs of the Polish version of Communication and Symbolic Behavior Scales-Developmental Profile Infant-Toddler Checklist

BACKGROUND

The first stage of diagnosing autism spectrum disorders usually involves population screening to detect children at risk. This study aims to assess the predictive convergent validity of the Polish version of the Communication and Symbolic Behavior Scales-Developmental Profile Infant-Toddler Checklist (CSBS-DP ITC) with the Autism Spectrum Rating Scales (ASRS), evaluate its sensitivity and specificity and assess the cut-off points for the possibility of using this questionnaire in population screening among children aged 6 to 24 months.

METHOD

The study was conducted among 602 children from the general population who had previously participated in the earlier phase of validation of the questionnaire for Polish conditions. The collected data were statistically processed to calculate the accuracy (i.e. sensitivity, specificity) of the questionnaire.

RESULTS

In individual age groups, the sensitivity of the questionnaire varies from 0.667 to 0.750, specificity from 0.854 to 0.939, positive predictive value from 0.261 to 0.4 and negative predictive value-from 0.979 to 0.981. Screening accuracy ranges from 0.847 to 0.923 depending on the age group. The adopted cut-off points are 21 points for children aged 9-12 months, 36 for children aged 13-18 months, 39 for children aged 19-24 months. Cut-off points could not be established for children aged 6-8 months. The convergent validity values with the ASRS ranged from -0.28 to -0.431 and were highest in the group of the oldest children.

CONCLUSIONS

These results indicate that the Polish version of the CSBS-DP ITC can be used as an effective tool for ASD universal screening.

High expression of 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2) associated with Diquat-induced damage

Diquat (DQ) is a commonly used bipyridine herbicide known for its toxic properties and adverse effects on individuals. However, the mechanism underlying DQ-induced damage remain elusive. Our research aimed to uncover the regulatory network involved in DQ-induced damage. We analyzed publicly accessible gene expression patterns and performed research using a DQ-induced damage animal model. The GSE153959 dataset from the Gene Expression Omnibus collection and the animal model of DQ-induced kidney injury were used to identify differentially expressed genes (DEGs). Pathways including the regulation of DNA-templated transcription in response to stress, RNA polymerase II transcription regulator complex and transcription coregulatory activity were shown to be enriched in 21 DEGs. We used least absolute shrinkage and selection operator (LASSO) regression analysis to find possible diagnostic biomarkers for DQ-induced damage. Then, we used an HK-2 cell model to confirm these results. Additionally, we confirmed that 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2) was the major gene associated with DQ-induced damage using multi-omics screening. The sample validation strongly suggested that HMGCS2 has promise as a diagnostic marker and may provide new targets for therapy in the context of DQ-induced damage.

Exploration of TCM syndrome types of the material basis and risk prediction of Wilson disease liver fibrosis based on 1H NMR metabolomics

Wilson disease (WD) is an autosomal recessive disorder characterized by abnormal copper metabolism. The accumulation of copper in the liver can progress to liver fibrosis and, ultimately, cirrhosis, which is a primary cause of death in WD patients. Metabonomic technology offers an effective approach to investigate the traditional Chinese medicine (TCM) syndrome types of WD-related liver fibrosis by monitoring the alterations in small molecule metabolites within the body. In this study, we employed 1H-Nuclear Magnetic Resonance (1H NMR) metabonomics to assess the metabolic profiles associated with five TCM syndrome types of WD-related liver fibrosis and analyzed the diagnostic and predictive capabilities of various metabolites. The study found a variety of metabolites, each with varying levels of diagnostic and predictive capabilities. Furthermore, the discerned differential metabolic pathways were primarily associated with various pathways involving carbohydrate metabolism, amino acid metabolism, and lipid metabolism. This study has identified various characteristic metabolic markers and pathways associated with different TCM syndromes of liver fibrosis in WD, providing a substantial foundation for investigating the mechanisms underlying these TCM syndromes.

Hybrid deep learning models for the screening of Diabetic Macular Edema in optical coherence tomography volumes

Several studies published so far used highly selective image datasets from unclear sources to train computer vision models and that may lead to overestimated results, while those studies conducted in real-life remain scarce. To avoid image selection bias, we stacked convolutional and recurrent neural networks (CNN-RNN) to analyze complete optical coherence tomography (OCT) cubes in a row and predict diabetic macular edema (DME), in a real-world diabetic retinopathy screening program. A retrospective cohort study was carried out. Throughout 4-years, 5314 OCT cubes from 4408 subjects who attended to the diabetic retinopathy (DR) screening program were included. We arranged twenty-two (22) pre-trained CNNs in parallel with a bidirectional RNN layer stacked at the bottom, allowing the model to make a prediction for the whole OCT cube. The staff of retina experts built a ground truth of DME later used to train a set of these CNN-RNN models with different configurations. For each trained CNN-RNN model, we performed threshold tuning to find the optimal cut-off point for binary classification of DME. Finally, the best models were selected according to sensitivity, specificity, and area under the receiver operating characteristics curve (AUROC) with their 95% confidence intervals (95%CI). An ensemble of the best models was also explored. 5188 cubes were non-DME and 126 were DME. Three models achieved an AUROC of 0.94. Among these, sensitivity, and specificity (95%CI) ranged from 84.1-90.5 and 89.7-93.3, respectively, at threshold 1, from 89.7-92.1 and 80-83.1 at threshold 2, and from 80.2-81 and 93.8-97, at threshold 3. The ensemble model improved these results, and lower specificity was observed among subjects with sight-threatening DR. Analysis by age, gender, or grade of DME did not vary the performance of the models. CNN-RNN models showed high diagnostic accuracy for detecting DME in a real-world setting. This engine allowed us to detect extra-foveal DMEs commonly overlooked in other studies, and showed potential for application as the first filter of non-referable patients in an outpatient center within a population-based DR screening program, otherwise ended up in specialized care.

Biochemical Signals of Survival: A Study on Mortality Markers in Coronary Bypass Surgery Patients

BACKGROUND

Coronary bypass surgery remains a cornerstone treatment for advanced coronary artery disease. Identifying reliable predictors of postoperative mortality can significantly enhance patient care and outcomes. This study investigates the prognostic value of preoperative and postoperative amylase levels, creatinine, alanine aminotransferase, and aspartate aminotransferase as mortality markers in coronary bypass surgery patients.

METHODS

We conducted a retrospective analysis of 343 patients who underwent coronary bypass surgery. We compared the preoperative and postoperative biochemical markers (amylase, creatinine, alanine aminotransferase, and aspartate aminotransferase) of patients who died within the first week post-surgery (n = 52) and those who survived (n = 291). Statistical analyses included chi-square tests for categorical variables, t-tests for continuous variables, and receiver operating characteristic analysis for predicting mortality.

RESULTS

No significant difference was observed in the distribution of blood groups between deceased and surviving patients. However, significant differences were noted in gender distribution and mean ages, with higher mortality observed in older and male patients. Preoperative creatinine levels were significantly higher in patients who died compared to survivors. Postoperatively, deceased patients exhibited significantly higher levels of amylase, creatinine, alanine aminotransferase, and aspartate aminotransferase. Receiver operating characteristic analysis revealed that postoperative amylase, creatinine, alanine aminotransferase, and aspartate aminotransferase values were good predictors of mortality, with amylase being the most significant predictor.

CONCLUSION

This study highlights the importance of biochemical markers, particularly amylase, as predictors of mortality in patients undergoing coronary bypass surgery. The findings suggest that monitoring and managing amylase, creatinine, alanine aminotransferase, and aspartate aminotransferase levels pre- and post-surgery could improve patient outcomes. This study lays the groundwork for further research into the mechanistic links between these biochemical markers and patient survival, potentially leading to improved prognostic tools and therapeutic strategies.

ESM1 May Be Used as a New Indicator for the Diagnosis and Prognosis of Early and Advanced Stage Digestive Tract Cancers

Background

The biological function and prognostic significance of endothelial cell specific molecule 1 (ESM1) in various cancers have been validated. This study aimed to explore the expression and clinical diagnosis values in patients with stomach adenocarcinoma (STAD) and esophageal carcinoma (ESCA).

Methods

Online database Gene Expression Omnibus was used to screen for abnormally expressed genes in STAD and ESCA. Besides, 36 STAD and 36 ESCA patients were enrolled, and their corresponding control groups were also 36 people each. Reverse transcription-quantitative polymerase chain reaction and Western blot were performed to analyze the expression of ESM1. Overall survival (OS) curve and receiver operating characteristics curve (ROC) analysis were used to assess the prognosis, and the sensitivity and specificity of ESM1 for the diagnosis of STAD and ESCA, respectively. Additionally, the effects of ESM1 on cell viability, migration, and invasion were analyzed by cell counting kit-8, transwell migration and invasion assays.

Results

The results showed that the poor OS of STAD and ESCA patients was correlated with high ESM1. Besides, ESM1 was increased in ESCA and STAD in in vivo and in vitro studies. ESM1 has a high accuracy [area under the curve (AUC) > 0.79] at stage I and IV of STAD and ESCA. Knockdown of ESM1 suppressed the cell viability, migration, and invasion and increased the apoptosis rate of AGS and TE1 cells.

Conclusion

Our study suggested that ESM1 might be used as a new indicator for the diagnosis and prognosis of early and advanced stage digestive tract cancers.

Detection of common pathogenesis of rheumatoid arthritis and atherosclerosis via microarray data analysis

Despite extensive research reveal rheumatoid arthritis (RA) is related to atherosclerosis (AS), common pathogenesis between these two diseases still needs to be explored. In current study, we explored the common pathogenesis between rheumatoid arthritis (RA) and atherosclerosis (AS) by identifying 297 Differentially Expressed Genes (DEGs) associated with both diseases. Through KEGG and GO functional analysis, we highlighted the correlation of these DEGs with crucial biological processes such as the vesicle transport, immune system process, signaling receptor binding, chemokine signaling and many others. Employing Protein-Protein Interaction (PPI) network analysis, we elucidated the associations between DEGs, revealing three gene modules enriched in immune system process, vesicle, signaling receptor binding, Pertussis, and among others. Additionally, through CytoHubba analysis, we pinpointed 11 hub genes integral to intergrin-mediated signaling pathway, plasma membrane, phosphotyrosine binding, chemokine signaling pathway and so on. Further investigation via the TRRUST database identified two key Transcription Factors (TFs), SPI1 and RELA, closely linked with these hub genes, shedding light on their regulatory roles. Finally, leveraging the collective insights from hub genes and TFs, we proposed 10 potential drug candidates targeting the molecular mechanisms underlying RA and AS pathogenesis. Further investigation on xCell revealed that 14 types of cells were all different in both AS and RA. This study underscores the shared pathogenic mechanisms, pivotal genes, and potential therapeutic interventions bridging RA and AS, offering valuable insights for future research and clinical management strategies.

Aging's Effect on Working Memory-Modality Comparison

Research exploring the impact of development and aging on working memory (WM) has primarily concentrated on visual and verbal domains, with limited attention paid to the tactile modality. The current study sought to evaluate WM encompassing storage and manipulation across these three modalities, spanning from childhood to old age. The study included 134 participants, divided into four age groups: 7-8, 11-12, 25-35, and 60-69. Each participant completed the Visuospatial Span, Digit Span, and Tactual Span, with forward and backward recall. The findings demonstrated a consistent trend in both forward and backward stages. Performance improved until young adulthood, progressively diminishing with advancing age. In the forward stage, the Tactual Span performance was worse than that of the Digit and Visuospatial Span for all participants. In the backward stage, the Visuospatial Span outperformed the Digit and Tactual Span across all age groups. Furthermore, the Tactual Span backward recall exhibited significantly poorer performance than the other modalities, primarily in the youngest and oldest age groups. In conclusion, age impacts WM differently across modalities, with tactile storage capacity being the most vulnerable. Additionally, tactile manipulation skills develop later in childhood but deteriorate sooner in adulthood, indicating a distinct component within tactile WM.

Psychometric properties of the cannabis abuse screening test (CAST) in a sample of Moroccans with cannabis use

BACKGROUND

The Cannabis Abuse Screening Test (CAST) is a widely used screening tool for identifying patterns of cannabis use that have negative health or social consequences for both the user and others involved. This brief screening instrument has been translated into multiple languages, and several studies examining its psychometric properties have been published. However, studies on the factorial validity and psychometric properties of a Moroccan version of the CAST are not yet available. The objective of this study is to validate the CAST, translated, and adapted to the Moroccan Arabic dialect among persons with cannabis use.

METHODS

A total of 370 participants from an addictology center in Fez City, were selected over two phases to form the study sample. First, in phase I, exploratory factor analysis was employed to evaluate the factor structure in the pilot sample (n1 = 150). Subsequently, in the second phase (Phase II), confirmatory factor analysis was utilized to confirm this structure in the validation sample (n2 = 220). All statistical analyses were carried out using the R program.

RESULTS

The CFA unveiled a three-factor structure that showed a good overall fit (χ2/df = 2.23, RMSEA = 0.07, SRMR = 0.02, CFI = 0.99, NFI = 0.98) and satisfactory local parameters (standardized factor loadings between 0.72 and 0.88). The model demonstrates satisfactory reliability and convergent validity, as evidenced by the acceptable values of composite reliability (CR) (0.76-0.88) and average variance extracted (AVE) (0.62-0.78), respectively. The square roots of the AVE exceeded the correlations of the factor pairs, and the heterotrait-monotrait (HTMT) ratio of the correlation values was below 0.85, indicating acceptable discriminant validity.

CONCLUSION

The reliability, convergent validity, and discriminant validity tests all demonstrated that the Moroccan version of the CAST performed well and can be considered a valid tool for screening of problematic cannabis use.

Minimal clinically important differenceof fatigue severity scale in patients with chronic stroke

BACKGROUND

One of the most prevalent symptoms of stroke is fatigue. Fatigue severity scale is the most often used tool for evaluating fatigue in stroke patients, its minimal clinically important difference threshold has not been determined. This study aimed to identify the minimal clinically important difference of fatigue severity scale in stroke patients.

METHODS

All study participants were examined using fatigue severity scale and multidimensional fatigue symptom inventory-short form before and after the intervention. The 6-week intervention combined graded activity training and pacing therapy employed to reduce fatigue severity. Participants reported changes in their fatigue severity after the intervention with the global rating of change and visual analog scale. The minimal clinically important difference of the fatigue severity scale calculated using both anchor- and distribution-based methods.

RESULTS

A total of 117 stroke patients were included in the study. Using multidimensional fatigue symptom inventory-short form, global rating of change, and visual analog scale as an anchor, the minimal clinically important difference of fatigue severity scale was obtained at 3.5, 4.5, and 4.5, respectively. The minimal clinically important difference for fatigue severity scale varied from 4.28 to 12.90 using the distribution-based method, with SEM = 4.28 displaying the best sensitivity and specificity for use as minimal clinically important difference.

CONCLUSIONS

The minimal clinically important difference value for the fatigue severity scale was estimated at 3.5_12.90 using anchor-based and distribution-based methods. The study's results can be utilized to understand the effectiveness of fatigue interventions in stroke patients in clinical and research settings.

Comparison of Sepsis-1, 2 and 3 for Predicting Mortality in Septic Patients of a Middle-Income Country: A Retrospective Observational Cohort Study

INTRODUCTION

The diagnosis of sepsis is based on expert consensus and does not yet have a "gold standard." With the aim of improving and standardizing diagnostic methods, there have already been three major consensuses on the subject. However, there are still few studies in middle-income countries comparing the methods. This study describes the characteristics of patients diagnosed with sepsis and evaluates and compares the performance of Sepsis-1, 2, and 3 criteria in predicting 28 days, and in-hospital mortality.

PATIENTS AND METHODS

A retrospective observational cohort study was conducted in the intensive care unit of a tertiary hospital. All admissions between January 1, 2018, and December 31, 2019, were reviewed. Patients diagnosed with sepsis were included.

RESULTS

During the study period, 653 patients diagnosed with sepsis (by any of the studied criteria) were included in the study. The 28 days mortality rate was 45.8%, and the in-hospital mortality rate was 59.7%. We observed that 72.1% of patients met the minimum criteria for diagnosing sepsis according to the three protocols, and this group also had the highest mortality rate. Age and comorbidities such as cancer and liver cirrhosis were significantly associated with in-hospital mortality. The most common microorganisms were Escherichia coli, Klebsiella spp., and Staphylococcus spp.

CONCLUSIONS

The study found that most patients met the diagnostic criteria for sepsis using the three methods. Sepsis-2 showed greater sensitivity to predict mortality, while Sequential Organ Failure Assessment showed low accuracy, but was the only significant one. Furthermore, quick Sequential Organ Failure Assessment (qSOFA) had the highest specificity for mortality. Overall, these findings suggest that, although all three methods contribute to the diagnosis and prognosis of sepsis, Sepsis-2 is particularly sensitive in predicting mortality. Sepsis-3 shows some accuracy but requires improvement, and qSOFA exhibits the highest specificity. More research is needed to improve predictive capabilities and patient outcomes.

A practical approach to interpretation of 18F-fluorodeoxyglucose positron emission tomography/computed tomography for postoperative spine infection

OBJECTIVE

18F-fluorodeoxyglucose-PET/CT is the imaging modality of choice for the diagnosis of postoperative spine infection. Published interpretation criteria are variable and often incompletely described. The objective was to develop a practical and standardized approach.

MATERIALS AND METHODS

Two-hundred-twenty-seven FDG-PET/CTs performed on 140 postoperative patients over a 7-year period were reviewed retrospectively. The presence or absence of infection was determined from clinical history, microbiology, other investigations, and clinical outcome during a minimum 6-month follow-up.

RESULTS

No activity attributable to normal healing was seen in the post-discectomy space or at the bone-hardware interface in the absence of a complication at any stage. Within the incision, activity from normal healing persisted for months. Wound infections were diagnosed clinically, and most had already been treated before FDG-PET/CT was done to assess deep structures. With proven infection, 95% of cases had activity in bone or soft tissue outside the surgical field. The remaining 5% had activity confined to the post-discectomy space. Sterile hardware loosening may cause elevated activity which remains confined to the bone/hardware interface. Pathogens are introduced directly at the time of surgery and may be avirulent resulting in indolent infection with low-grade activity. At the same time, activity from non-infectious causes can be intense. A semi-quantitative method using SUVmax performed poorly compared with assessment of the distribution of activity.

CONCLUSION

These observations have been incorporated into a checklist which is now being used at the time of interpretation. The potential sensitivity and specificity in the diagnosis of infection are close to 100%.

Role of Rain in the Spore Dispersal of Fungal Pathogens Associated with Grapevine Trunk Diseases

Grapevine trunk diseases are caused by a complex of fungi that belong to different taxa, which produce different spore types and have different spore dispersal mechanisms. It is commonly accepted that rainfall plays a key role in spore dispersal, but there is conflicting information in the literature on the relationship between rain and spore trapping in aerobiology studies. We conducted a systematic literature review, extracted quantitative data from published papers, and used the pooled data for Bayesian analysis of the effect of rain on spore trapping. We selected 17 papers covering 95 studies and 8,778 trapping periods, concerning a total of 26 fungal taxa causing Botryosphaeria dieback (BD), Esca complex (EC), and Eutypa dieback (ED). Results confirmed the role of rain in the spore dispersal of these fungi but revealed differences among the different fungi. Rain was a good predictor of spore trapping for ED (AUROC = 0.820) and BD (0.766) but not for the ascomycetes involved in EC (0.569) and not for the only basidiomycetes, Fomitiporella viticola, studied as for spore discharge (AUROC not significant). Prediction of spore trapping was more accurate for negative prognosis than for positive prognosis; a rain cutoff of ≥0.2 mm provided an overall accuracy of ≥0.61 for correct prognoses. Spores trapped in rainless periods accounted for only <10% of the total spores. Our analysis had some drawbacks, which were mainly caused by knowledge gaps and limited data availability; these drawbacks are discussed to facilitate further research.

Investigation of the validity and reliability of the 3-meter backward walk test in high functional level adults with lower limb amputation

BACKGROUND

Backward walk training has an important place in the rehabilitation programs of lower extremity amputees.

OBJECTIVE

This study aimed to investigate the test-retest validity and reliability of the 3-meter backward walk test (3MBWT), minimal detectable change, and the cutoff time in high functional level adults with lower limb amputations (LLAs). Adults with LLA (n = 30) and healthy adults (n = 29) were included in the study.

STUDY DESIGN

This is a randomized cross-sectional study.

METHODS

The Modified Fall Efficacy Score, Rivermead Mobility Index, and Timed Up and Go test with the 3MBWT were used to evaluate the concurrent validity of the test. The second evaluation (retest) was performed by the same physiotherapist 1 week following the first evaluation (test). The validity was assessed by correlating the 3MBWT times with the scores of other measures and by comparing the 3MBWT times between adults with LLA and healthy adults.

RESULTS

Test-retest reliability of the 3MBWT was excellent. The intraclass correlation coefficient for the 3MBWT was 0.950. The standard error of measurement and minimal detectable change values were 0.38 and 0.53, respectively. A moderate correlation was found between the 3MBWT, Modified Fall Efficacy Score, Timed Up and Go test, and Rivermead Mobility Index ( p < 0.001). Significant differences in the 3MBWT times were found between adults with LLA and healthy controls ( p < 0.001). The cutoff time of 3.11 s discriminates healthy adults from high functional level adults with LLA.

CONCLUSIONS

The 3MBWT was determined to be valid, reliable, and easy-to-apply tool in high functional level adults with LLA. This assessment is a useful and practical measurement for dynamic balance in high functional level adults with LLA.

Identification of potential biomarkers for aging diagnosis of mesenchymal stem cells derived from the aged donors

BACKGROUND

The clinical application of human bone-marrow derived mesenchymal stem cells (MSCs) for the treatment of refractory diseases has achieved remarkable results. However, there is a need for a systematic evaluation of the quality and safety of MSCs sourced from donors. In this study, we sought to assess one potential factor that might impact quality, namely the age of the donor.

METHODS

We downloaded two data sets from each of two Gene Expression Omnibus (GEO), GSE39035 and GSE97311 databases, namely samples form young (< 65 years of age) and old (> 65) donor groups. Through, bioinformatics analysis and experimental validation to these retrieved data, we found that MSCs derived from aged donors can lead to differential expression of gene profiles compared with those from young donors, and potentially affect the function of MSCs, and may even induce malignant tumors.

RESULTS

We identified a total of 337 differentially expressed genes (DEGs), including two upregulated and eight downregulated genes from the databases of both GSE39035 and GSE97311. We further identified 13 hub genes. Six of them, TBX15, IGF1, GATA2, PITX2, SNAI1 and VCAN, were highly expressed in many human malignancies in Human Protein Atlas database. In the MSCs in vitro senescent cell model, qPCR analysis validated that all six hub genes were highly expressed in senescent MSCs. Our findings confirm that aged donors of MSCs have a significant effect on gene expression profiles. The MSCs from old donors have the potential to cause a variety of malignancies. These TBX15, IGF1, GATA2, PITX2, SNAI1, VCAN genes could be used as potential biomarkers to diagnosis aging state of donor MSCs, and evaluate whether MSCs derived from an aged donor could be used for therapy in the clinic. Our findings provide a diagnostic basis for the clinical use of MSCs to treat a variety of diseases.

CONCLUSIONS

Therefore, our findings not only provide guidance for the safe and standardized use of MSCs in the clinic for the treatment of various diseases, but also provide insights into the use of cell regeneration approaches to reverse aging and support rejuvenation.

A 2D video-based assessment is associated with 3D biomechanical contributors to dynamic knee valgus in the coronal plane

Introduction

Adolescent athletes involved in sports that involve cutting and landing maneuvers have an increased risk of anterior cruciate ligament (ACL) tears, highlighting the importance of identifying risky movement patterns such as dynamic knee valgus (DKV). Qualitative movement screenings have explored two-dimensional (2D) scoring criteria for DKV, however, there remains limited data on the validity of these screening tools. Determining a 2D scoring criterion for DKV that closely aligns with three-dimensional (3D) biomechanical measures will allow for the identification of poor knee position in adolescent athletes on a broad scale. The purpose of this study was to establish a 2D scoring criterion that corresponds to 3D biomechanical measures of DKV.

Methods

A total of 41 adolescent female club volleyball athletes performed a three-task movement screen consisting of a single-leg squat (SLS), single-leg drop landing (SLDL), and double-leg vertical jump (DLVJ). A single rater scored 2D videos of each task using four criteria for poor knee position. A motion capture system was used to calculate 3D joint angles, including pelvic obliquity, hip adduction, knee abduction, ankle eversion, and foot progression angle. Receiver operating characteristic curves were created for each 2D scoring criterion to determine cut points for the presence of movement faults, and areas under the curve (AUC) were computed to describe the accuracy of each 2D criterion compared to 3D biomechanical data.

Results

3D measures indicated knee abduction angles between 2.4°-4.6° (SD 4.1°-4.3°) at the time point when the center of the knee joint was most medial during the three tasks. AUCs were between 0.62 and 0.93 across scoring items. The MEDIAL scoring item, defined as the knee joint positioned inside the medial border of the shoe, demonstrated the greatest association to components of DKV, with AUCs ranging from 0.67 to 0.93.

Conclusion

The MEDIAL scoring criterion demonstrated the best performance in distinguishing components of DKV, specifically pelvic obliquity, hip adduction, ankle eversion, and foot progression. Along with the previously published scoring definitions for trunk-specific risk factors, the authors suggest that the MEDIAL criterion may be the most indicative of DKV, given an association with 3D biomechanical risk factors.

What do Hyperhidrosis Quality of Life Index (HidroQoL©) scores mean? Transferring science into practice by establishing a score banding system

BACKGROUND

The Hyperhidrosis Quality of Life Index (HidroQoL©) is a measure of quality of life (QoL) impacts in hyperhidrosis (HH).

OBJECTIVES

We aimed to establish score banding systems for the HidroQoL total score for specific contexts representing different severity/impact categories by using the Dermatology Life Quality Index (DLQI) and the Hyperhidrosis Disease Severity Scale (HDSS) as anchors, including data from 357 patients from a phase III clinical trial.

METHODS

We used the HDSS, the established DLQI score bands and two single items (items 5 and 7) of the DLQI as anchors for the creation of banding systems for the HidroQoL. These anchors were chosen via consensus among an expert group according to relevance to patient experience. Due to the distribution of the HDSS and the single DLQI item 7, receiver operating characteristic curves were computed in order to create an optimal cut-off value of the HidroQoL total score. For the DLQI banding system and the single DLQI item 5, we created a banding system for the HidroQoL based on the distribution of their different categories.

RESULTS

A score of 30 and greater is proposed as the cut-off value for sweating that 'always interferes in daily activities', based on the HDSS as anchor. In terms of overall skin QoL effects, score bands of 0-6, 7-18, 19-25, 26-32 and 33-36 represent 'no effect', 'small effect', 'moderate effect', 'very large effect' and 'extremely large effect' on the patient's life, respectively.

CONCLUSIONS

In this study, we propose different banding systems for four different contexts: skin-specific QoL (DLQI banding), HH severity (HDSS), working and studying (single DLQI item 7) and social and leisure activities (single DLQI item 5). These banding systems and cut-off values can be used in clinical research and practice to place the patients in different severity categories.

Elucidating the role of TWIST1 in ulcerative colitis: a comprehensive bioinformatics and machine learning approach

Background: Ulcerative colitis (UC) is a common and progressive inflammatory bowel disease primarily affecting the colon and rectum. Prolonged inflammation can lead to colitis-associated colorectal cancer (CAC). While the exact cause of UC remains unknown, this study aims to investigate the role of the TWIST1 gene in UC. Methods: Second-generation sequencing data from adult UC patients were obtained from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified, and characteristic genes were selected using machine learning and Lasso regression. The Receiver Operating Characteristic (ROC) curve assessed TWIST1's potential as a diagnostic factor (AUC score). Enriched pathways were analyzed, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Variation Analysis (GSVA). Functional mechanisms of marker genes were predicted, considering immune cell infiltration and the competing endogenous RNA (ceRNA) network. Results: We found 530 DEGs, with 341 upregulated and 189 downregulated genes. TWIST1 emerged as one of four potential UC biomarkers via machine learning. TWIST1 expression significantly differed in two datasets, GSE193677 and GSE83687, suggesting its diagnostic potential (AUC = 0.717 in GSE193677, AUC = 0.897 in GSE83687). Enrichment analysis indicated DEGs associated with TWIST1 were involved in processes like leukocyte migration, humoral immune response, and cell chemotaxis. Immune cell infiltration analysis revealed higher rates of M0 macrophages and resting NK cells in the high TWIST1 expression group, while TWIST1 expression correlated positively with M2 macrophages and resting NK cell infiltration. We constructed a ceRNA regulatory network involving 1 mRNA, 7 miRNAs, and 32 long non-coding RNAs (lncRNAs) to explore TWIST1's regulatory mechanism. Conclusion: TWIST1 plays a significant role in UC and has potential as a diagnostic marker. This study sheds light on UC's molecular mechanisms and underscores TWIST1's importance in its progression. Further research is needed to validate these findings in diverse populations and investigate TWIST1 as a therapeutic target in UC.

Temporal Dynamics and Dispersal Patterns of the Primary Inoculum of Coniella diplodiella, the Causal Agent of Grape White Rot

Grape white rot can cause considerable yield losses in viticulture areas worldwide and is principally caused by Coniella diplodiella. The fungus overwinters in berry mummies on the soil surface or on the trellis and produces pycnidia and conidia that serve as primary inoculum. However, little is known about the temporal dynamics and dispersal pattern of C. diplodiella conidia. In this study, we investigated the production and dispersal of C. diplodiella conidia from a primary inoculum source, namely, affected mummified berries that overwintered in two vineyards in northern Italy in 2021 and 2022. Conidia of C. diplodiella were repeatedly produced in berry mummies from the budburst of vines to harvesting, with approximately 50 and 75% of the total conidia in a season being produced before fruit set and véraison, respectively. The production dynamics of C. diplodiella conidia over time were described by a Weibull equation in which the thermal time is the independent variable, with a concordance correlation coefficient of ≥0.964. A rainfall cutoff of ≥0.2 mm provided an overall accuracy of ≥0.86 in predicting conidial dispersal through rain splashes from berry mummies on the soil surface, with the number of dispersed conidia increasing with the amount of rainfall. The dispersal of conidia from mummies on the trellis by washing with rain required at least 6.1 mm of rain. The proposed mathematical equations and rain cutoffs can be used to predict periods with a high dispersal risk of C. diplodiella.

Cloud-Based Machine Learning Platform to Predict Clinical Outcomes at Home for Patients With Cardiovascular Conditions Discharged From Hospital: Clinical Trial

BACKGROUND

Hospitalizations account for almost one-third of the US $4.1 trillion health care cost in the United States. A substantial portion of these hospitalizations are attributed to readmissions, which led to the establishment of the Hospital Readmissions Reduction Program (HRRP) in 2012. The HRRP reduces payments to hospitals with excess readmissions. In 2018, >US $700 million was withheld; this is expected to exceed US $1 billion by 2022. More importantly, there is nothing more physically and emotionally taxing for readmitted patients and demoralizing for hospital physicians, nurses, and administrators. Given this high uncertainty of proper home recovery, intelligent monitoring is needed to predict the outcome of discharged patients to reduce readmissions. Physical activity (PA) is one of the major determinants for overall clinical outcomes in diabetes, hypertension, hyperlipidemia, heart failure, cancer, and mental health issues. These are the exact comorbidities that increase readmission rates, underlining the importance of PA in assessing the recovery of patients by quantitative measurement beyond the questionnaire and survey methods.

OBJECTIVE

This study aims to develop a remote, low-cost, and cloud-based machine learning (ML) platform to enable the precision health monitoring of PA, which may fundamentally alter the delivery of home health care. To validate this technology, we conducted a clinical trial to test the ability of our platform to predict clinical outcomes in discharged patients.

METHODS

Our platform consists of a wearable device, which includes an accelerometer and a Bluetooth sensor, and an iPhone connected to our cloud-based ML interface to analyze PA remotely and predict clinical outcomes. This system was deployed at a skilled nursing facility where we collected >17,000 person-day data points over 2 years, generating a solid training database. We used these data to train our extreme gradient boosting (XGBoost)-based ML environment to conduct a clinical trial, Activity Assessment of Patients Discharged from Hospital-I, to test the hypothesis that a comprehensive profile of PA would predict clinical outcome. We developed an advanced data-driven analytic platform that predicts the clinical outcome based on accurate measurements of PA. Artificial intelligence or an ML algorithm was used to analyze the data to predict short-term health outcome.

RESULTS

We enrolled 52 patients discharged from Stanford Hospital. Our data demonstrated a robust predictive system to forecast health outcome in the enrolled patients based on their PA data. We achieved precise prediction of the patients' clinical outcomes with a sensitivity of 87%, a specificity of 79%, and an accuracy of 85%.

CONCLUSIONS

To date, there are no reliable clinical data, using a wearable device, regarding monitoring discharged patients to predict their recovery. We conducted a clinical trial to assess outcome data rigorously to be used reliably for remote home care by patients, health care professionals, and caretakers.

Decoding acute myocarditis in patients with COVID-19: Early detection through machine learning and hematological indices

During the persistent COVID-19 pandemic, the swift progression of acute myocarditis has emerged as a profound concern due to its augmented mortality, underscoring the urgency of prompt diagnosis. This study analyzed blood samples from 5,230 COVID-19 individuals, identifying key blood and myocardial markers that illuminate the relationship between COVID-19 severity and myocarditis. A predictive model, applying Bayesian and random forest methodologies, was constructed for myocarditis' early identification, unveiling a balanced gender distribution in myocarditis cases contrary to a male predominance in COVID-19 occurrences. Particularly, older men exhibited heightened vulnerability to severe COVID-19 strains. The analysis revealed myocarditis was notably prevalent in younger demographics, and two subvariants COVID-19 progression paths were identified, characterized by symptom intensity and specific blood indicators. The enhanced myocardial marker model displayed remarkable diagnostic accuracy, advocating its valuable application in future myocarditis detection and treatment strategies amidst the COVID-19 crisis.

Screening for affective dysregulation in school-aged children: relationship with comprehensive measures of affective dysregulation and related mental disorders

Affective dysregulation (AD) is characterized by irritability, severe temper outbursts, anger, and unpredictable mood swings, and is typically classified as a transdiagnostic entity. A reliable and valid measure is needed to adequately identify children at risk of AD. This study sought to validate a parent-rated screening questionnaire, which is part of the comprehensive Diagnostic Tool for Affective Dysregulation in Children (DADYS-Screen), by analyzing relationships with comprehensive measures of AD and related mental disorders in a community sample of children with and without AD. The sample comprised 1114 children aged 8-12 years and their parents. We used clinical, parent, and child ratings for our analyses. Across all raters, the DADYS-Screen showed large correlations with comprehensive measures of AD. As expected, correlations were stronger for measures of externalizing symptoms than for measures of internalizing symptoms. Moreover, we found negative associations with emotion regulation strategies and health-related quality of life. In receiver operating characteristic (ROC) analyses, the DADYS-Screen adequately identified children with AD and provided an optimal cut-off. We conclude that the DADYS-Screen appears to be a reliable and valid measure to identify school-aged children at risk of AD.

Deep learning and machine learning predictive models for neurological function after interventional embolization of intracranial aneurysms

Objective

The objective of this study is to develop a model to predicts the postoperative Hunt-Hess grade in patients with intracranial aneurysms by integrating radiomics and deep learning technologies, using preoperative CTA imaging data. Thereby assisting clinical decision-making and improving the assessment and prognosis of postoperative neurological function.

Methods

This retrospective study encompassed 101 patients who underwent aneurysm embolization surgery. 851 radiomic features were extracted from CTA images. 512 deep learning features are extracted from last layer of ResNet50 deep convolutional neural network model. The feature screening process pipeline encompassed intraclass correlation coefficient analysis, principal component analysis, U test, spearman correlation analysis, minimum redundancy maximum relevance algorithm and Lasso regression, to identify features most correlated with postoperative Hunt-Hess grading. In the model construction phase, three distinct models were constructed: radiomics feature-based model (RSM), deep learning feature-based model (DLM), and deep learning-radiomics feature fusion model (DLRSCM). The study also calculated the radiomics score and combined it with clinical data to construct a Nomogram for predictive modeling. DLM, RSM and DLRSCM model was constructed by 9 base algorithms and 1 ensemble learning algorithm - Stacking ensemble model. Model performance was evaluated based on the area under the Receiver Operating Characteristic (ROC) curve (AUC), Matthews Correlation Coefficient (MCC), calibration curves, and decision curves analysis.

Results

5 significant radiomic feature and 4 significant deep learning features were obtained through the feature selection process. These features were utilized for model construction. Bootstrap resampling method was used for internal validation of the models. In terms of model evaluation, the DLM model, the stacking ensemble algorithm results achieved an AUC of 0.959 and MCC of 0.815. In the RSM model, the stacking ensemble model AUC was 0.935 and MCC was 0.793. The stacking ensemble model in DLRSCM outperformed others, with an AUC of 0.968 and MCC of 0.820. Results indicated that the ANN performed optimally among all base models, while the stacked ensemble learning model exhibited the highest predictive performance.

Conclusion

This study demonstrates that the combination of radiomics and deep learning is an effective approach to predict the postoperative Hunt-Hess grade in patients with intracranial aneurysms. This holds significant value in the early identification of postoperative neurological complications and in enhancing clinical decision-making.

Machine learning to identify clinically relevant Candida yeast species

Fungal infections, especially due to Candida species, are on the rise. Multi-drug resistant organisms such as Candida auris are difficult and time consuming to identify accurately. Machine learning is increasingly being used in health care, especially in medical imaging. In this study, we evaluated the effectiveness of six convolutional neural networks (CNNs) to identify four clinically important Candida species. Wet-mounted images were captured using bright field live-cell microscopy followed by separating single-cells, budding-cells, and cell-group images which were then subjected to different machine learning algorithms (custom CNN, VGG16, ResNet50, InceptionV3, EfficientNetB0, and EfficientNetB7) to learn and predict Candida species. Among the six algorithms tested, the InceptionV3 model performed best in predicting Candida species from microscopy images. All models performed poorly on raw images obtained directly from the microscope. The performance of all models increased when trained on single and budding cell images. The InceptionV3 model identified budding cells of C. albicans, C. auris, C. glabrata (Nakaseomyces glabrata), and C. haemulonii in 97.0%, 74.0%, 68.0%, and 66.0% cases, respectively. For single cells of C. albicans, C. auris, C. glabrata, and C. haemulonii InceptionV3 identified 97.0%, 73.0%, 69.0%, and 73.0% cases, respectively. The sensitivity and specificity of InceptionV3 were 77.1% and 92.4%, respectively. Overall, this study provides proof of the concept that microscopy images from wet-mounted slides can be used to identify Candida yeast species using machine learning quickly and accurately.

Defining the glenohumeral range of motion required for overhead shoulder mobility: an observational study

Background

Recovery of overhead mobility after shoulder surgery is time-consuming and important for patient satisfaction. Overhead stretching and mobilization of the scapulothoracic and glenohumeral (GH) joints are common treatment interventions. The isolated GH range of motion (ROM) of flexion, abduction, and external rotation required to move above 120° of global shoulder flexion in the clinical setting remains unclear. This study clarified the GH ROM needed for overhead mobility.

Methods

The timely development of shoulder ROM in patients after shoulder surgery was analyzed. Passive global shoulder flexion, GH flexion, abduction, and external rotation ROM were measured using goniometry and visually at 2-week intervals starting 6-week postsurgery until the end of treatment. Receiver operating characteristic curves were used to identify the GH ROM cutoff values allowing overhead mobility.

Results

A total of 21 patients (mean age 49 years; 76% men) after rotator cuff repair (71%), Latarjet shoulder stabilization (19%), and arthroscopic biceps tenotomy (10%) were included. The ROM cutoff value that accurately allowed overhead mobility was 83° for GH flexion and abduction with the area under the curve (AUC) ranging from 0.90 to 0.93 (p < 0.001). The cutoff value for GH external rotation was 53% of the amount of movement on the opposite side (AUC 0.87, p < 0.001).

Conclusions

Global shoulder flexion above 120° needs almost full GH flexion and abduction to be executable. External rotation ROM seems less important as long as it reaches over 53% of the opposite side.

Neural Computation-Based Methods for the Early Diagnosis and Prognosis of Alzheimer's Disease Not Using Neuroimaging Biomarkers: A Systematic Review

Background

The growing number of older adults in recent decades has led to more prevalent geriatric diseases, such as strokes and dementia. Therefore, Alzheimer's disease (AD), as the most common type of dementia, has become more frequent too.

Background

Objective: The goals of this work are to present state-of-the-art studies focused on the automatic diagnosis and prognosis of AD and its early stages, mainly mild cognitive impairment, and predicting how the research on this topic may change in the future.

Methods

Articles found in the existing literature needed to fulfill several selection criteria. Among others, their classification methods were based on artificial neural networks (ANNs), including deep learning, and data not from brain signals or neuroimaging techniques were used. Considering our selection criteria, 42 articles published in the last decade were finally selected.

Results

The most medically significant results are shown. Similar quantities of articles based on shallow and deep ANNs were found. Recurrent neural networks and transformers were common with speech or in longitudinal studies. Convolutional neural networks (CNNs) were popular with gait or combined with others in modular approaches. Above one third of the cross-sectional studies utilized multimodal data. Non-public datasets were frequently used in cross-sectional studies, whereas the opposite in longitudinal ones. The most popular databases were indicated, which will be helpful for future researchers in this field.

Conclusions

The introduction of CNNs in the last decade and their superb results with neuroimaging data did not negatively affect the usage of other modalities. In fact, new ones emerged.

Development and psychometric evaluation of an Azerbaijani-Turkish grammar comprehension test

Due to the lack of appropriate developmental language tests in the Azerbaijani-Turkish language, the present study aimed to develop the Azerbaijani-Turkish grammar comprehension test (ATGCT) for children aged 4-6 years and determine its validity and reliability. This study was conducted in three phases. First, the target grammatical structures were extracted during the item generation phase. Second, item reduction and content validity ratio (CVR) were calculated. The content validity index (CVI) was determined after designing the items. This test was administered to typically developing children (TD) (N = 30). Face validity was confirmed and modifications were applied. The second version of the test was performed on 170 TD and 60 children with developmental language disorder (DLD) aged 4-6 years were selected using the random cluster method. Third, item analysis was performed, and eight items were removed. The construct validity, reliability, and ROC analysis of the final form of the test were evaluated. The psychometric properties considered in the study included construct validity (group, gender, and age discriminative validity) and reliability (test-retest, inter-rater, and internal consistency). The final test version contained 56 items and confirmed face validity. The Scale Content Validity was .91, and the Item Content Validity was between .8 and 1. The test showed a content validity ratio of .96, indicating that it assesses appropriate content. The construct validity analysis revealed significant differences between the TD and DLD groups and among the four age groups. Test-retest and inter-rater reliability were significantly correlated. Furthermore, the high correlation between test items (ICC= .90) demonstrated that the ATGCT had excellent internal consistency. The receiver operating characteristic (ROC) analysis results indicated that the test had high sensitivity and specificity in all four age groups and effectively distinguished children with TD and those with DLD. In conclusion, based on the psychometric assessment of the test, it appears that the ATGCT has appropriate values for reliability and validity measures, and it can be used as the first suitable and quick test by researchers and clinicians.

StableDNAm: towards a stable and efficient model for predicting DNA methylation based on adaptive feature correction learning

BACKGROUND

DNA methylation, instrumental in numerous life processes, underscores the paramount importance of its accurate prediction. Recent studies suggest that deep learning, due to its capacity to extract profound insights, provides a more precise DNA methylation prediction. However, issues related to the stability and generalization performance of these models persist.

RESULTS

In this study, we introduce an efficient and stable DNA methylation prediction model. This model incorporates a feature fusion approach, adaptive feature correction technology, and a contrastive learning strategy. The proposed model presents several advantages. First, DNA sequences are encoded at four levels to comprehensively capture intricate information across multi-scale and low-span features. Second, we design a sequence-specific feature correction module that adaptively adjusts the weights of sequence features. This improvement enhances the model's stability and scalability, or its generality. Third, our contrastive learning strategy mitigates the instability issues resulting from sparse data. To validate our model, we conducted multiple sets of experiments on commonly used datasets, demonstrating the model's robustness and stability. Simultaneously, we amalgamate various datasets into a single, unified dataset. The experimental outcomes from this combined dataset substantiate the model's robust adaptability.

CONCLUSIONS

Our research findings affirm that the StableDNAm model is a general, stable, and effective instrument for DNA methylation prediction. It holds substantial promise for providing invaluable assistance in future methylation-related research and analyses.

Reliability, Validity and Responsiveness of the Slovenian Version of the Patient Evaluation Measure (PEM-Slo) in Patients with Wrist and Hand Disorders

Introduction

The Patient Evaluation Measure (PEM) is a region-specific patient reported outcome measure (PROM) for hand and wrist disorders, first introduced in English for patients with hand surgery in 1995. The purpose of the study was to assess the psychometric properties of the translated and cross-culturally adapted Slovenian version of PEM (PEM-Slo).

Methods

The study was designed as a single-centre observational prospective study conducted from July 2020 to March 2021. The psychometric evaluation was performed on fifty-one patients with miscellaneous hand and wrist disorders. Reliability was tested for internal consistency and test-retest reliability. Convergent and divergent validity, responsiveness, floor and ceiling effect, and interpretability with the determination of minimal detectable change (MDC) and minimal clinically important difference (MCID) were assessed.

Results

The PEM-Slo has excellent internal consistency (Cronbach's α 0.932) and good to excellent test-retest reliability (intraclass correlation coefficient=0.874). Convergent validity was proved with high to moderate correlations of PEM-Slo with DASH, grip strength and self-care, usual activities, and pain EQ-5D-5L subscales, whereas no correlation of PEM-Slo with EQ-5D-5L mobility and anxiety/depression subscale confirmed divergent validity. The PEM-Slo responsiveness was high (standardised response mean=1.42, effect size=1.25). MDC was 18.01 and MCID was 17.31. No floor or ceiling effect was found.

Conclusion

The PEM-Slo is a reliable, valid and responsive PROM for Slovenian-speaking patients with hand and wrist disorders.