Farrerol alleviates insulin resistance and hepatic steatosis of metabolic associated fatty liver disease by targeting PTPN1.

Metabolic associated fatty liver disease (MAFLD) is the most common chronic liver disease worldwide, characterized by excess lipid deposition. Insulin resistance (IR) serves as a fundamental pathogenic factor in MAFLD. However, currently, there are no approved specific agents for its treatment. Farrerol, a novel compound with antioxidant and anti-inflammatory effects, has garnered significant attention in recent years due to its hepatoprotective properties. Despite this, the precise underlying mechanisms of action remain unclear. In this study, a network pharmacology approach predicted protein tyrosine phosphatase non-receptor type 1 (PTPN1) as a potential target for farrerol's action in the liver. Subsequently, the administration of farrerol improved insulin sensitivity and glucose tolerance in MAFLD mice. Furthermore, farrerol alleviated lipid accumulation by binding to PTPN1 and reducing the dephosphorylation of the insulin receptor (INSR) in HepG2 cells and MAFLD mice. Thus, the phosphoinositide 3-kinase/serine/threonine-protein kinases (PI3K/AKT) signalling pathway was active, leading to downstream protein reduction. Overall, the study demonstrates that farrerol alleviates insulin resistance and hepatic steatosis of MAFLD by targeting PTPN1.

Automated machine learning for the identification of asymptomatic COVID-19 carriers based on chest CT images.

BACKGROUND: Asymptomatic COVID-19 carriers with normal chest computed tomography (CT) scans have perpetuated the ongoing pandemic of this disease. This retrospective study aimed to use automated machine learning (AutoML) to develop a prediction model based on CT characteristics for the identification of asymptomatic carriers. METHODS: Asymptomatic carriers were from Yangzhou Third People's Hospital from August 1st, 2020, to March 31st, 2021, and the control group included a healthy population from a nonepizootic area with two negative RT‒PCR results within 48 h. All CT images were preprocessed using MATLAB. Model development and validation were conducted in R with the H2O package. The models were built based on six algorithms, e.g., random forest and deep neural network (DNN), and a training set (n = 691). The models were improved by automatically adjusting hyperparameters for an internal validation set (n = 306). The performance of the obtained models was evaluated based on a dataset from Suzhou (n = 178) using the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score. RESULTS: A total of 1,175 images were preprocessed with high stability. Six models were developed, and the performance of the DNN model ranked first, with an AUC value of 0.898 for the test set. The sensitivity, specificity, PPV, NPV, F1 score and accuracy of the DNN model were 0.820, 0.854, 0.849, 0.826, 0.834 and 0.837, respectively. A plot of a local interpretable model-agnostic explanation demonstrated how different variables worked in identifying asymptomatic carriers. CONCLUSIONS: Our study demonstrates that AutoML models based on CT images can be used to identify asymptomatic carriers. The most promising model for clinical implementation is the DNN-algorithm-based model.

Predicting the Recurrence of Common Bile Duct Stones After ERCP Treatment with Automated Machine Learning Algorithms.

BACKGROUND: Recurrence of common bile duct stones (CBDs) commonly happens after endoscopic retrograde cholangiopancreatography (ERCP). The clinical prediction models for the recurrence of CBDs after ERCP are lacking. AIMS: We aim to develop high-performance prediction models for the recurrence of CBDS after ERCP treatment using automated machine learning (AutoML) and to assess the AutoML models versus the traditional regression models. METHODS: 473 patients with CBDs undergoing ERCP were recruited in the single-center retrospective cohort study. Samples were divided into Training Set (65%) and Validation Set (35%) randomly. Three modeling approaches, including fully automated machine learning (Fully automated), semi-automated machine learning (Semi-automated), and traditional regression were applied to fit prediction models. Models' discrimination, calibration, and clinical benefits were examined. The Shapley additive explanations (SHAP), partial dependence plot (PDP), and SHAP local explanation (SHAPLE) were proposed for the interpretation of the best model. RESULTS: The area under roc curve (AUROC) of semi-automated gradient boost machine (GBM) model was 0.749 in Validation Set, better than the other fully/semi-automated models and the traditional regression models (highest AUROC = 0.736). The calibration and clinical application of AutoML models were adequate. Through the SHAP-PDP-SHAPLE pipeline, the roles of key variables of the semi-automated GBM model were visualized. Lastly, the best model was deployed online for clinical practitioners. CONCLUSION: The GBM model based on semi-AutoML is an optimal model to predict the recurrence of CBDs after ERCP treatment. In comparison with traditional regressions, AutoML algorithms present significant strengths in modeling, which show promise in future clinical practices.

Inhibitory Effects of Bacillus Coagulans TL3 on the Ileal Oxidative Stress and Inflammation Induced by Lipopolysaccharide in Rats.

This study aimed to explore the effect and mechanism of Bacillus coagulans TL3 (B. coagulans TL3) on ileal inflammatory injury induced by lipopolysaccharide (LPS). Animal models were established wherein male Wistar rats were randomly divided into four groups: a control group, an LPS group, a high-concentration B. coagulans (HBC) group, and a low-concentration B. coagulans (LBC) group. The results showed that the biochemical indices changed, significant pathological changes were found, the number of apoptotic cells increased in the ileal tissue of the LPS group rats; the protein expressions of NFκB, MYD88, TLR4, TNF-α, Il-6, IL-1ß, Claudin-1, Occludin, and ZO-1 in the LPS group were significantly decreased. The biochemical indices, pathological changes, and protein expressions in rats subjected to intragastric administration with high or low concentrations of B. coagulans TL3, were significantly reversed compared with the LPS group. These results indicated that TL3 strain could protect rats against ileal oxidative stress and inflammation induced by LPS and the protective mechanism was related to inhibition of the toll-like receptor 4 (TLR4) / myeloid differentiation factor-88 (MyD88) signaling pathway.

A deafness-associated tRNA mutation caused pleiotropic effects on the m1G37 modification, processing, stability and aminoacylation of tRNAIle and mitochondrial translation.

Defects in the posttranscriptional modifications of mitochondrial tRNAs have been linked to human diseases, but their pathophysiology remains elusive. In this report, we investigated the molecular mechanism underlying a deafness-associated tRNAIle 4295A>G mutation affecting a highly conserved adenosine at position 37, 3' adjacent to the tRNA's anticodon. Primer extension and methylation activity assays revealed that the m.4295A>G mutation introduced a tRNA methyltransferase 5 (TRMT5)-catalyzed m1G37 modification of tRNAIle. Molecular dynamics simulations suggested that the m.4295A>G mutation affected tRNAIle structure and function, supported by increased melting temperature, conformational changes and instability of mutated tRNA. An in vitro processing experiment revealed that the m.4295A>G mutation reduced the 5' end processing efficiency of tRNAIle precursors, catalyzed by RNase P. We demonstrated that cybrid cell lines carrying the m.4295A>G mutation exhibited significant alterations in aminoacylation and steady-state levels of tRNAIle. The aberrant tRNA metabolism resulted in the impairment of mitochondrial translation, respiratory deficiency, decreasing membrane potentials and ATP production, increasing production of reactive oxygen species and promoting autophagy. These demonstrated the pleiotropic effects of m.4295A>G mutation on tRNAIle and mitochondrial functions. Our findings highlighted the essential role of deficient posttranscriptional modifications in the structure and function of tRNA and their pathogenic consequence of deafness.

Public Imaging Datasets of Gastrointestinal Endoscopy for Artificial Intelligence: a Review.

With the advances in endoscopic technologies and artificial intelligence, a large number of endoscopic imaging datasets have been made public to researchers around the world. This study aims to review and introduce these datasets. An extensive literature search was conducted to identify appropriate datasets in PubMed, and other targeted searches were conducted in GitHub, Kaggle, and Simula to identify datasets directly. We provided a brief introduction to each dataset and evaluated the characteristics of the datasets included. Moreover, two national datasets in progress were discussed. A total of 40 datasets of endoscopic images were included, of which 34 were accessible for use. Basic and detailed information on each dataset was reported. Of all the datasets, 16 focus on polyps, and 6 focus on small bowel lesions. Most datasets (n = 16) were constructed by colonoscopy only, followed by normal gastrointestinal endoscopy and capsule endoscopy (n = 9). This review may facilitate the usage of public dataset resources in endoscopic research.

Identification of Asymptomatic COVID-19 Patients on Chest CT Images Using Transformer-Based or Convolutional Neural Network-Based Deep Learning Models.

Novel coronavirus disease 2019 (COVID-19) has rapidly spread throughout the world; however, it is difficult for clinicians to make early diagnoses. This study is to evaluate the feasibility of using deep learning (DL) models to identify asymptomatic COVID-19 patients based on chest CT images. In this retrospective study, six DL models (Xception, NASNet, ResNet, EfficientNet, ViT, and Swin), based on convolutional neural networks (CNNs) or transformer architectures, were trained to identify asymptomatic patients with COVID-19 on chest CT images. Data from Yangzhou were randomly split into a training set (n = 2140) and an internal-validation set (n = 360). Data from Suzhou was the external-test set (n = 200). Model performance was assessed by the metrics accuracy, recall, and specificity and was compared with the assessments of two radiologists. A total of 2700 chest CT images were collected in this study. In the validation dataset, the Swin model achieved the highest accuracy of 0.994, followed by the EfficientNet model (0.954). The recall and the precision of the Swin model were 0.989 and 1.000, respectively. In the test dataset, the Swin model was still the best and achieved the highest accuracy (0.980). All the DL models performed remarkably better than the two experts. Last, the time on the test set diagnosis spent by two experts-42 min, 17 s (junior); and 29 min, 43 s (senior)-was significantly higher than those of the DL models (all below 2 min). This study evaluated the feasibility of multiple DL models in distinguishing asymptomatic patients with COVID-19 from healthy subjects on chest CT images. It found that a transformer-based model, the Swin model, performed best.

Automated Multimodal Machine Learning for Esophageal Variceal Bleeding Prediction Based on Endoscopy and Structured Data.

Esophageal variceal (EV) bleeding is a severe medical emergency related to cirrhosis. Early identification of cirrhotic patients with at a high risk of EV bleeding is key to improving outcomes and optimizing medical resources. This study aimed to evaluate the feasibility of automated multimodal machine learning (MMML) for predicting EV bleeding by integrating endoscopic images and clinical structured data. This study mainly includes three steps: step 1, developing deep learning (DL) models using EV images by 12-month bleeding on TensorFlow (backbones include ResNet, Xception, EfficientNet, ViT and ConvMixer); step 2, training and internally validating MMML models integrating clinical structured data and DL model outputs to predict 12-month EV bleeding on an H2O-automated machine learning platform (algorithms include DL, XGBoost, GLM, GBM, RF, and stacking); and step 3, externally testing MMML models. Furthermore, existing clinical indices, e.g., the MELD score, Child‒Pugh score, APRI, and FIB-4, were also examined. Five DL models were transfer learning to the binary classification of EV endoscopic images at admission based on the occurrence or absence of bleeding events during the 12-month follow-up. An EfficientNet model achieved the highest accuracy of 0.868 in the validation set. Then, a series of MMML models, integrating clinical structured data and the output of the EfficientNet model, were automatedly trained to predict 12-month EV bleeding. A stacking model showed the highest accuracy (0.932), sensitivity (0.952), and F1-score (0.879) in the test dataset, which was also better than the existing indices. This study is the first to evaluate the feasibility of automated MMML in predicting 12-month EV bleeding based on endoscopic images and clinical variables.

C9orf72 knockdown alleviates hepatic insulin resistance by promoting lipophagy.

Insulin resistance (IR) attributed by the deficiency of lipophagy, is an abnormal state of downregulation of insulin-mediated glucose uptake and use into the liver. Chromosome 9 open reading frame 72 (C9orf72) variously modulates autophagy. We investigated the role and the downstream pathway of C9orf72 in hepatic IR. We found that C9orf72 knockdown alleviated hepatic IR by lipophagy promotion in T2DM mice and in IR-challenged hepatocytes in vitro. C9orf72 interacted with and activated cell division cycle 42 (Cdc42) protein in IR-challenged hepatocytes, Which in turn, inhibits lipophagy by promoting neural Wiskott-Aldrich syndrome protein (N-WASP) expression and activation. C9orf72 inhibited lipophagy by activating the Cdc42/N-WASP axis to facilitate hepatic IR; therefore, the knockdown of C9orf72 may be potentially therapeutic for the treatment of IR.

Limitations of Phage Therapy and Corresponding Optimization Strategies: A Review.

Bacterial infectious diseases cause serious harm to human health. At present, antibiotics are the main drugs used in the treatment of bacterial infectious diseases, but the abuse of antibiotics has led to the rapid increase in drug-resistant bacteria and to the inability to effectively control infections. Bacteriophages are a kind of virus that infects bacteria and archaea, adopting bacteria as their hosts. The use of bacteriophages as antimicrobial agents in the treatment of bacterial diseases is an alternative to antibiotics. At present, phage therapy (PT) has been used in various fields and has provided a new technology for addressing diseases caused by bacterial infections in humans, animals, and plants. PT uses bacteriophages to infect pathogenic bacteria so to stop bacterial infections and treat and prevent related diseases. However, PT has several limitations, due to a narrow host range, the lysogenic phenomenon, the lack of relevant policies, and the lack of pharmacokinetic data. The development of reasonable strategies to overcome these limitations is essential for the further development of this technology. This review article described the current applications and limitations of PT and summarizes the existing solutions for these limitations. This information will be useful for clinicians, people working in agriculture and industry, and basic researchers.

Effectiveness of a Guided Internet- and Mobile-Based Intervention for Patients with Chronic Back Pain and Depression (WARD-BP): A Multicenter, Pragmatic Randomized Controlled Trial.

INTRODUCTION: There is neither strong evidence on effective treatments for patients with chronic back pain (CBP) and depressive disorder nor sufficiently available mental health care offers. OBJECTIVE: The aim is to assess the effectiveness of internet- and mobile-based interventions (IMI) as a scalable approach for treating depression in a routine care setting. METHODS: This is an observer-masked, multicenter, pragmatic randomized controlled trial with a randomization ratio of 1:1.Patients with CBP and diagnosed depressive disorder (mild to moderate severity) were recruited from 82 orthopedic rehabilitation clinics across Germany. The intervention group (IG) received a guided depression IMI tailored to CBP next to treatment-as-usual (TAU; including medication), while the control group (CG) received TAU. The primary outcome was observer-masked clinician-rated Hamilton depression severity (9-week follow-up). The secondary outcomes were: further depression outcomes, pain-related outcomes, health-related quality of life, and work capacity. Biostatistician blinded analyses using regression models were conducted by intention-to-treat and per protocol analysis. RESULTS: Between October 2015 and July 2017, we randomly assigned 210 participants (IG, n = 105; CG, n = 105), mostly with only a mild pain intensity but substantial pain disability. No statistically significant difference in depression severity between IG and CG was observed at the 9-week follow-up (ß = -0.19, 95% CI -0.43 to 0.05). Explorative secondary depression (4/9) and pain-related (4/6) outcomes were in part significant (p < 0.05). Health-related quality of life was significantly higher in the IG. No differences were found in work capacity. CONCLUSION: The results indicate that an IMI for patients with CBP and depression in a routine care setting has limited impact on depression. Benefits in pain and health-related outcomes suggest that an IMI might still be a useful measure to improve routine care.

Further development in the assessment of psychological flexibility: validation of the German committed action questionnaire.

BACKGROUND: Psychological flexibility is considered a fundamental aspect of health. It includes six interrelated facets: 1) cognitive defusion, 2) acceptance, 3) contact with the present moment, 4) self-as-context, 5) values, and 6) committed action. To gain further insight into psychological flexibility and its effects on health, reliable and valid instruments to assess all facets are needed. Committed action is one facet that is understudied. A long and short version of a validated measure (CAQ and CAQ-8) have been developed in English. Currently, there are no German versions of the CAQ. Aim of this study is to validate German-language versions of these in a chronic pain population. METHODS: The CAQ instructions and items were translated and evaluated in a chronic pain population (N = 181). Confirmatory factor analysis and Mokken scale analysis were conducted to evaluate the German questionnaires. Correlations with health outcomes, including quality of life (SF-12), physical and emotional functioning (MPI, BPI, PHQ-9, GAD-7), pain intensity, and with other facets of psychological flexibility (CPAQ, FAH-II) were investigated for convergent validity purposes. Scale reliability was assessed by the alpha, MS, lambda-2, LCRC, and omega coefficient. RESULTS: A bifactor model consisting of one general factor and two methodological factors emerged from the analysis. Criteria for reliability and validity were met. Medium to strong correlations to health outcomes and other facets of psychological flexibility were found. Results were similar to the original English version. CONCLUSIONS: The present study presents a valid and reliable instrument to investigate committed action in German populations. Future studies could expand the present findings by evaluating the German CAQ versions in non-pain populations. The role of committed action and the wider psychological flexibility model in pain and other conditions deserves further investigation.

Promoting physical activity through a psychological group intervention in cardiac rehabilitation: a randomized controlled trial.

We examined the long-term effectiveness of a group-based psychological intervention ("MoVo-LISA") to promote physical activity in patients with coronary heart disease. In this randomized controlled trial, N = 202 inactive patients with coronary heart disease were assigned to the control group (n = 102; treatment as usual) or the intervention group (n = 100; treatment as usual plus MoVo-LISA). Physical activity was assessed at baseline, 6 weeks (post-treatment), 6 months, and 12 months after discharge. ANCOVA for repeated measures revealed a significant interaction effect [p < .001; η p 2 = .214] indicating a large effect [d = 1.03] of the intervention on behavior change post-treatment. At 12-month follow-up, the level of physical activity in the intervention group was still 94 min per week higher than in the control group (p < .001; d = 0.57). Results of this RCT indicate that the MoVo-LISA intervention substantially improves the level of physical activity among initially inactive patients with coronary heart disease up to 1 year after the intervention.

Correction: A Web-Based Acceptance-Facilitating Intervention for Identifying Patients' Acceptance, Uptake, and Adherence of Internet- and Mobile-Based Pain Interventions: Randomized Controlled Trial.

[This corrects the article DOI: 10.2196/jmir.9925.].

A Web-Based Acceptance-Facilitating Intervention for Identifying Patients' Acceptance, Uptake, and Adherence of Internet- and Mobile-Based Pain Interventions: Randomized Controlled Trial.

BACKGROUND: Internet- and mobile-based interventions are effective for the treatment of chronic pain. However, little is known about patients' willingness to engage with these types of interventions and how the uptake of such interventions can be improved. OBJECTIVE: The aim of this study was to identify people's acceptance, uptake, and adherence (primary outcomes) with regard to an internet- and mobile-based intervention for chronic pain and the influence of an information video as an acceptance-facilitating intervention (AFI). METHODS: In this randomized controlled trial with a parallel design, we invited 489 individuals with chronic pain to participate in a Web-based survey assessing the acceptance of internet- and mobile-based interventions with the offer to receive an unguided internet- and mobile-based intervention for chronic pain after completion. Two versions of the Web-based survey (with and without AFI) were randomly sent to two groups: one with AFI (n=245) and one without AFI (n=244). Participants who completed the Web-based survey with or without AFI entered the intervention group or the control group, respectively. In the survey, the individuals' acceptance of pain interventions, measured with a 4-item scale (sum score ranging from 4 to 20), predictors of acceptance, sociodemographic and pain-related variables, and physical and emotional functioning were assessed. Uptake rates (log in to the intervention) and adherence (number of completed modules) to the intervention was assessed 4 months after intervention access. To examine which factors influence acceptance, uptake rate, and adherence in the internet- and mobile-based interventions, we conducted additional exploratory subgroup analyses. RESULTS: In total, 57 (intervention group) and 58 (control group) participants in each group completed the survey and were included in the analyses. The groups did not differ with regard to acceptance, uptake rate, or adherence (P=.64, P=.56, P=.75, respectively). Most participants reported moderate (68/115, 59.1%) to high (36/115, 31.3%) acceptance, with 9.6% (11/115) showing low acceptance (intervention group: mean 13.91, SD 3.47; control group: mean 13.61, SD 3.50). Further, 67% (38/57, intervention group) and 62% (36/58, control group) had logged into the intervention. In both groups, an average of 1.04 (SD 1.51) and 1.14 (SD 1.90) modules were completed, respectively. CONCLUSIONS: The informational video was not effective with regard to acceptance, uptake rate, or adherence. Despite the high acceptance, the uptake rate was only moderate and adherence was remarkably low. This study shows that acceptance can be much higher in a sample participating in an internet- and mobile-based intervention efficacy trial than in the target population in routine health care settings. Thus, future research should focus not only on acceptance and uptake facilitating interventions but also on ways to influence adherence. Further research should be conducted within routine health care settings with more representative samples of the target population. TRIAL REGISTRATION: German Clinical Trial Registration DRKS00006183; http://www.drks.de/drks_web/navigate.do ?navigationId=trial.HTML&TRIAL_ID=DRKS00006183 (Archived by WebCite at http://www.webcitation.org/70ebHDhne).

[Internet- and mobile-based approaches : Psycho-social diagnostics and treatment in medical rehabilitation]. / Internet- und mobilebasierte Ansätze : Psychosoziale Diagnostik und Behandlung in der medizinischen Rehabilitation.

Technology-based approaches for psychosocial diagnostics and interventions provide an attractive opportunity to optimize medical rehabilitation. Based on an Internet- and mobile-based assessment of existing functional health impairments, appropriate planning, implementation of corresponding courses of action as well as outcome assessment can take place. This can be implemented in the form of Internet- and mobile-based interventions (IMI).The present article provides an overview of the basic knowledge of IMI and their evidence base both in general and in particular for their use in medical rehabilitation. Important aspects of internet and mobile-based psycho-social diagnostics are discussed subsequently. Finally, an outlook for the use of Internet- and mobile-based diagnostics and interventions in medical rehabilitation is given.

Cognitive-behavioral statuses in depression and internet gaming disorder of adolescents: A transdiagnostic approach.

To investigate the comorbidity of adolescent depression and Internet gaming disorder (IGD) and their shared and unique cognitive-behavioral factors (i.e., self-esteem, dysfunctional attitudes, hopelessness, and coping), a large-scale school-based survey was conducted among 3147 Chinese secondary school students in Hong Kong. Probable depression and IGD were screened using the Centre for Epidemiological Studies-Depression Scale and DSM-5 IGD checklist, respectively. Multinomial logistic regression was performed to identify the associations between different condition statuses and cognitive-behavioral factors. Four groups were identified, including comorbidity group (having probable depression and IGD), IGD group (having probable IGD alone), depression group (probable depression alone), and healthy group (neither condition). Comorbidity group showed the worst cognitive-behavioral statuses, followed by depression group and then IGD group. Compared with healthy group, those with lower self-esteem and higher hopelessness and dysfunctional attitudes were more likely to be classified into depression group and comorbidity group, while maladaptive coping was positively associated with all three disorder groups. The results suggest that depression and IGD may share common cognitive-behavioral mechanisms (e.g., maladaptive coping) but also own their uniqueness regarding specific factors (e.g., hopelessness and self-esteem). A transdiagnostic intervention approach targeting the common factors may effectively address the comorbidity.

Deep learning in the radiologic diagnosis of osteoporosis: a literature review.

OBJECTIVE: Osteoporosis is a systemic bone disease characterized by low bone mass, damaged bone microstructure, increased bone fragility, and susceptibility to fractures. With the rapid development of artificial intelligence, a series of studies have reported deep learning applications in the screening and diagnosis of osteoporosis. The aim of this review was to summary the application of deep learning methods in the radiologic diagnosis of osteoporosis. METHODS: We conducted a two-step literature search using the PubMed and Web of Science databases. In this review, we focused on routine radiologic methods, such as X-ray, computed tomography, and magnetic resonance imaging, used to opportunistically screen for osteoporosis. RESULTS: A total of 40 studies were included in this review. These studies were divided into three categories: osteoporosis screening (n = 20), bone mineral density prediction (n = 13), and osteoporotic fracture risk prediction and detection (n = 7). CONCLUSIONS: Deep learning has demonstrated a remarkable capacity for osteoporosis screening. However, clinical commercialization of a diagnostic model for osteoporosis remains a challenge.

Computer-aided diagnostic system with automated deep learning method based on the AutoGluon framework improved the diagnostic accuracy of early esophageal cancer.

Background: There have been studies on the application of computer-aided diagnosis (CAD) in the endoscopic diagnosis of early esophageal cancer (EEC), but there is still a significant gap from clinical application. We developed an endoscopic CAD system for EEC based on the AutoGluon framework, aiming to explore the feasibility of automatic deep learning (DL) in clinical application. Methods: The endoscopic pictures of normal esophagus, esophagitis, and EEC were collected from The First Affiliated Hospital of Soochow University (September 2015 to December 2021) and the Norwegian HyperKvasir database. All images of non-cancerous esophageal lesions and EEC in this study were pathologically examined. There were three tasks: task A was normal vs. lesion classification under non-magnifying endoscopy (n=932 vs. 1,092); task B was non-cancer lesion vs. EEC classification under non-magnifying endoscopy (n=594 vs. 429); and task C was non-cancer lesion vs. EEC classification under magnifying endoscopy (n=505 vs. 824). In all classification tasks, we took 100 pictures as the verification set, and the rest comprised as the training set. The CAD system was established based on the AutoGluon framework. Diagnostic performance of the model was compared with that of endoscopists grouped according to years of experience (senior >15 years; junior <5 years). Model evaluation indicators included accuracy, recall rate, precision, F1 value, interpretation time, and the area under the receiver operating characteristic (ROC) curve (AUC). Results: In tasks A and B, the accuracies of medium-performance CAD and high-performance CAD were lower than those of junior doctors and senior doctors. In task C, the medium-performance and high-performance CAD accuracies were close to those of junior doctors and senior doctors. The high-performance CAD model outperformed the junior doctors in both task A (0.850 vs. 0.830) and task C (0.840 vs. 0.830) in sensitivity comparison, but there was still a large gap between high-performance CAD models and doctors in sensitivity comparison. In task A, with the aid of CAD pre-interpretation, the accuracy of junior and senior physicians were significantly improved (from 0.880 to 0.915 and from 0.920 to 0.945, respectively); the time spent on film reading was significantly shortened (junior: from 11.3 to 8.7 s; senior: from 6.7 to 5.5 s). In task C, with the aid of CAD pre-interpretation, the accuracy of junior and senior physicians were significantly improved (from 0.850 to 0.865 and from 0.915 to 0.935, respectively); the reading time was significantly shortened (junior: from 9.5 to 7.7 s; senior: from 5.6 to 3.0 s). Conclusions: The CAD system based on the AutoGluon framework can assist doctors to improve the diagnostic accuracy and reading time of EEC under endoscopy. This study reveals that automatic DL methods are promising in clinical application.

Evaluation of super resolution technology for digestive endoscopic images.

Object: This study aims to evaluate the value of super resolution (SR) technology in augmenting the quality of digestive endoscopic images. Methods: In the retrospective study, we employed two advanced SR models, i.e., SwimIR and ESRGAN. Two discrete datasets were utilized, with training conducted using the dataset of the First Affiliated Hospital of Soochow University (12,212 high-resolution images) and evaluation conducted using the HyperKvasir dataset (2,566 low-resolution images). Furthermore, an assessment of the impact of enhanced low-resolution images was conducted using a 5-point Likert scale from the perspectives of endoscopists. Finally, two endoscopic image classification tasks were employed to evaluate the effect of SR technology on computer vision (CV). Results: SwinIR demonstrated superior performance, which achieved a PSNR of 32.60, an SSIM of 0.90, and a VIF of 0.47 in test set. 90 % of endoscopists supported that SR preprocessing moderately ameliorated the readability of endoscopic images. For CV, enhanced images bolstered the performance of convolutional neural networks, whether in the classification task of Barrett's esophagus (improved F1-score: 0.04) or Mayo Endoscopy Score (improved F1-score: 0.04). Conclusions: SR technology demonstrates the capacity to produce high-resolution endoscopic images. The approach enhanced clinical readability and CV models' performance of low-resolution endoscopic images.